Special Publication No.

4
of the Society for Geology
Applied to Mineral Deposits
Geology and
Metallogeny of
Copper Deposits
Proceedings of the Copper Symposium
27th International Geological Congress
Moscow, 1984

Edited by
G.H. Friedrich A.D. Genkin A.J. Naldrett
J. D. Ridge R. H. Sillitoe F. M. Vokes

With 258 Figures

Springer-Verlag
Berlin Heidelberg New York
London Paris Tokyo
Professor Professor
GUNTHER H. FRIEDRICH JOHN D. RIDGE
Institut fUr Mineralogie 1402 NW 18th Street
und Lagerstiittenlehre Gainesville, FL 32605, USA
RWTHAachen
WiillnerstraBe 2
5100 Aachen, FRG Dr. RICHARD H. SILLITOE
8 West Hill Park
Professor Highgate Village
ALEXANDR D. GENKIN London N6 6ND
CCCPIIGEM England
Staromonetny, 35
Moscow, 109017, USSR
Professor
Professor FRANK M. VOKES
ANTHONY J. NALDRETT Geologisk Institutt
Department of Geology Norges Tekniske H0gskole
University of Toronto Universitetet i Trondheim
Toronto, Ontario 7034 Trondheim
Canada M5S lAI Norway

ISBN-13: 978-3-642-70904-3 e-ISBN-13: 978-3-642-70902-9

DOl: 10.1 007/978-3-642-70902-9

Library of Congress Cataloging in Publication Data. Geology and metallogeny of

copper deposits. (Special publication no. 4 of the Society for Geology Applied to
Mineral Deposits) Proceedings of a symposium on copper metallogeny held at the
27th International Geological Congress, Moscow, USSR. I. Copper Ores-Congresses.
I. Friedrich, Giinther, 1929-. II. International Geological Congress (27th: Moscow,
USSR) III. Series: Special publication ... of the Society for Geology Applied to
Mineral Deposits; no. 4. QE390.2.C6G46 1986 553.4'3 86-6745
This work is subject to copyright. All rights are reserved, whether the whole or part of
the material is concerned, specifically those of translation, reprinting, re-use of illustra-
tions, broadcasting, reproduction by photocopying machine or similar means, and stor-
age in data banks. Under § 54 of the German Copyright Law where copies are made for
other than private use a fee is payable to 'Verwertungsgesellschaft Wort', Munich.

® Springer-Verlag Berlin Heidelberg 1986

Softcover reprint of the hardcover 1st edition 1986

The use of registered names, trademarks etc. in this publication does not imply, even in
the absence of a specific statement, that such names are exempt from the relevant pro-
tective laws and regulations and therefore free for general use.

Typesetting: K + V Fotosatz GmbH, Beerfelden.

213213130-543210
Preface

Copper belongs to those metals whose concentrations in nature

arise from a broad diversity of endogeneous and exogeneous pro-
cesses, which applies to essentially all genetic classes of ore
deposits.
This is the first proceedings volume on copper metallogeny to
cover the worldwide distribution of the four main groups of cop-
per deposits, including in Part I: copper-nickel deposits with
cobalt and platinum group elements; Part II: copper-molybde-
num-gold deposits with silver, zinc, and lead; Part III/IV: copper-
zinc-lead deposits (with silver etc.).
On the occasion of the 27th International Geological Congress
in Moscow, USSR, a symposium on copper metallogeny was held,
dealing with metallogenesis and mineral deposits. The symposium
was organized and sponsored by three international societies en-
gaged in the field of ore deposits: The Society of Economic Geol-
ogy (SEG), the Society of Geology Applied to Mineral Deposits
(SGA) and the International Association on the Genesis of Ore
Deposits (IAGOD). Invited papers were presented in four ses-
sions: (1) Copper deposits in mafic and ultramafic complexes, (2)
Porphyry copper deposits, (3) Copper deposits of volcanic-hydro-
thermal association, and (4) Sediment-hosted copper deposits.
The sessions were chaired by A. D. Genkin, A. J. Naldrett, J. D.
Ridge and G. I. Gorbunov; V.1. Sotnikov, A. Soregaroli, R. H.
Sillitoe and V. A. Evstrakhin; F. M. Vokes, A.1. Krivtsov, M.
Solomon and N. I. Eremin; G. H. Friedrich, Yu. V. Bogdanov,
A.C. Brown and F.P. Krendelev.
We wish to express our thanks to the publisher for the work in-
volved in maintaining the known Springer quality of composing
and printing. Beside the authors' work, the expenditure of effort
and time by numerous referees should be mentioned, who re-
viewed the manuscripts in detail, especially those from non-En-
glish speaking countries. They also proposed improvements to the
English translations of foreign manuscripts.

The Editors
Contents

Introduction ....................................... 1

Part I Deposits in Mafic and Ultramafic Complexes

The Physical and Petrologic Setting and Textural and

Compositional Characteristics of Sulfides from the
South Kawishiwi Intrusion, Duluth Complex,
Minnesota, USA
M. FOOSE and P. WEIBLEN(With 5 Figures) .......... 8

Application of Stable Isotopic Studies to Problems of

Magmatic Sulfide Ore Genesis with Special Reference
to the Duluth Complex, Minnesota
E.M. RIPLEY (With 10 Figures) ..................... 25

Controls on the Formation of Komatiite-Associated

Nickel-Copper Sulfide Deposits
C. M. LESHER and D. I. GROVES (With 3 Figures) 43

Depositional Environments of Volcanic Peridotite-

Associated Nickel Sulphide Deposits with Special
Reference to the Kambalda Dome
J. J. GRESHAM (With 8 Figures) .. . . . . . . . . . . . . . . . . . . . 63

Geochemistry of the Sudbury Igneous Complex: A Model

for the Complex and Its Ores
A. J. NALDRETT (With 14 Figures) .................. 91

Sulfide Petrology and Genesis of Copper-Nickel Ore

Deposits
V. V. DISTLER, A.D. GENKIN, and O.A. DYUZHIKOV
(With 5 Figures) .................................. 111

Types and Distinctive Features of Ore-Bearing Formations

of Copper-Nickel Deposits
M.N. GODLEVSKYandA.P. LIKHACHEV ............ 124
VIII Contents

On the Role of Metamorphism in the Formation of

Nickel-Copper Sulfide Deposits in the Kola Peninsula
Yu.N. YAKOVLEV and A.K. YAKOVLEVA
(With 4 Figures) .................................. 135

The Gabbro-Wehrlite Association in the Eastern Part of

the Baltic Shield
E. HANSKI(With 12 Figures) ....................... 151

Peridotitic Komatiites and the Origin of Ores (South-

eastern Part of the Baltic Shield)
V.1. KOCHNEV-PERVUKHOV, E. S. ZASKIND, and
V. V. PROSKURYAKOV(With 3 Figures) .............. 171

Geologic Setting of Selected Chromium and Nickel

Deposits of China
P.-F.FAN(With7Figures) ......................... 179

Part II Porphyry Deposits

Recent Advances in Porphyry Base Metal Deposit

Research
J.M. GUILBERT (With 4 Figures) .................... 196

The Geochemical Behaviour of Copper and Molybdenum

in Ore-Forming Processes
A. V. KUDRIN, L.N. VARYASH, Yu.N. PASHKOV, and
V.1. REKHARSKY (With 3 Figures) . . . . . . . . . . . . . . . . . .. 209

Batholith-Volcano Coupling in the Metallogeny of

Porphyry Copper Deposits
P.E. DAMON (With 7 Figures) ...................... 216

Space-Time Distribution, Crustal Setting and Cu/Mo

Ratios of Central Andean Porphyry Copper Deposits:
Metallogenic Implications
R.H. SILLITOE(With 6 Figures) ..................... 235

Metallogenic Zoning of Volcano-Plutonic Belts and

Porphyry-Copper Mineralization
V. A. E VSTRAKHIN, A.1. KRIVTSOV, and
I.F. MIGACHYOV(With 7 Figures) .................. 251

Geological and Structural Conditions of Localization of

the High-Grade Ores of Porphyry Copper Deposits
I.M. GOLOVANOV, E.1. NIKOLAYEVA, and
M. A. KAZIKHIN (With 4 Figures) ................... 261
Contents IX

Elements Determining in the Geological-Structural Model

of the Erdenetuin-Obo Copper-Molybdenum Ore Field,
Mongolia
M. ZHAMSRAN, V.1. SOTNIKOV, A.P. BERZINA,
D. GARAMZHAV, and Yu. A. SARYANOV
(With 4 Figures) .................................. 271

Genetic Aspects of the Recsk Mineralized Complex,

Hungary
Cs. BAKSA(With 11 Figures) ....................... 280

Ore-Magmatic Systems of Copper-Molybdenum Deposits

V.1. SOTNIKovandA.P. BERZINA(With3 Figures) .... 291

Part III Deposits of Volcanic-Hydrothermal Association

Base Metal Deposits in the Iberian Pyrite Belt

G.K. STRAUSS and K. G. GRAY (With 12 Figures) ... . .. 304

Copper-Pyrite and Pyrite Base Metal Deposits of the

Caucasian Region
N.K. KURBANOV(With 5 Figures) . . .... . . ... ........ 325

The Hidden Mineralogical and Geochemical Zoning and

the Ore-Forming Conditions of Copper and Copper-
Lead-Zinc Massive Sulfide Deposits
N.1. EREMIN(With 2 Figures) ...................... 338

Zoning of Massive Sulphide Deposits and their Origin

N.S. SKRIPCHENKO, V.1. SHCHEGLOV, G. V. RYABOV,
and V. L. ANDREYEV (With 5 Figures) ............... 350

Physico-Chemical Conditions of Base Metal Sulphide Ore

Formation
A.G. TVALCHRELIDZE(With 6 Figures) .............. 358

On the Genesis of Barite Associated with Volcanogenic

Massive Sulfides, Fukazawa Mine, Hokuroku District,
Japan
St.1. KALOGEROPOULOS and S. D. SCOTT
(With 11 Figures) ................................. 370

Part IV Sediment-Hosted Deposits

Diagenetic Features at White Pine (Michigan), Redstone

(N. W. Territories, Canada) and Kamoto (Zaire).
Sequence of Mineralization in Sediment-Hosted
Copper Deposits (Part 1)
A. C. BROWN and F. M. CHARTRAND (With 1 Figure) 390
x Contents

Diagenetic Sulphide Mineralization Within the Stratiform

Copper-Cobalt Deposit of West Kambove (Shaba-
Zaire). Sequence of Mineralization in Sediment-Hosted
Copper Deposits (Part 2)
J. CAILTEUX(With 7 Figures) ...................... 398

Geochemical Aspects of Stratiform and Red-Bed Copper

Deposits in the Catskill Formation (Pennsylvania,
USA) and Redstone Area (Canada). Sequence of
Mineralization in Sediment-Hosted Copper Deposits
(Part 3)
A. W. ROSE, A. T. SMITH, R. L. LUSTWERK,
H. OHMOTO, and L.D. Hoy (With 7 Figures) ......... 412

Stratabound Copper Deposits in East South-Central

Alaska: Their Characteristics and Origin
M. K. SOOD, R. J. WAGNER, and H. D. MARKAZI
(With 8 Figures) .................................. 422

Major Element Geochemistry of the Host Rocks in Some

Sediment-Hosted Copper Deposits
B. MOINE, L. GUILLOUX, and D. AUDEOUD
(With 7 Figures) .................................. 443
Zechstein Copper-Bearing Shales in Poland. Lagoonal
Environments and the Sapropel Model of Genesis
C. HARANCZYK (With 11 Figures) ................... 461
Formation Conditions of Copper Sandstone and Copper-
Shale Deposits
A.M. LUR'YE(With 7 Figures) ...................... 477
Major Types of Copper-Bearing Zones in the Soviet
Union
Yu. v. BOGDANOV(With 3 Figures) . . .... ... . .. ... . .. 492
Red-Colored Terrigenous Sediments - Specific Copper-
Forming Systems
B. B. SUSURA, V. O. GLYBOVSKY, and A. V. KISLITSIN
(With 3 Figures) .................................. 504
Cupriferous Sandstones and Shales of the Siberian
Platform
F.P. KRENDELEV, L.F. NARKELYUN, A.I. TRUBACHEV,
V. S. SALIKHOV, P. N. VOLODIN, V. V. KUNITSIN,
V. S. CHECHETKIN, and N. N. BAKUN(With 4 Figures) 513
Genetic Types of Copper Mineralization in the Igarka
Area, West of the Siberian Platform
I. F. GABLINA (With 7 Figures) ..................... 524
Contents XI

The Diverse Styles of Sediment-Hosted Copper Deposits

in Australia
LB. LAMBERT, J. KNUTSON, T.H. DONNELLY, and
H. ETMINAN(With 12 Figures) ..................... 540

Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide

Deposits - A Quantitative Kinetic Approach
E. MERINO, C. MOORE, P. ORTOLEVA, and
E. M. RIPLEY (With 4 Figures) .. . . . . . . . . . . . . . . . . . . . . 559

Results of Recent Exploration for Copper-Silver Deposits

in the Kupferschiefer of West Germany
F.-P. SCHMIDT, C. SCHUMACHER, V. SPIETH, and
G. H. FRIEDRICH (With 10 Figures) .. . . . .. . .. . . . . . . . . 572

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 583

List of Contributors

You will find the addresses at the beginning of the respective contribution

Andreyev, V. L. 350 Kislitsin, A. V. 504

Audeoud, D. 443 Knutson, J. 540
Baksa, Cs. 280 Kochnev-Pervukhov, V.1. 171
Bakun, N.N. 513 Krendelev, F. P. 513
Berzina, A. P. 271, 291 Krivtsov, A.I. 251
Bogdanov, Yu. V. 492 Kudrin, A. V. 209
Brown, A. C. 390 Kunitsin, V. V. 513
Cailteux, J. 398 Kurbanov, N.K. 325
Chartrand, F. M. 390 Lambert, I. B. 540
Chechetkin, V. S. 513 Lesher, C.M. 43
Damon, P.E. 216 Likhachev, A. P. 124
Distler, V.V. 111 Lur'ye, A. M. 477
Donnelly, T. H. 540 Lustwerk, R. L. 412
Dyuzhikov,O.A. 111 Markazi, H. D. 422
Eremin, N.1. 338 Merino, E. 559
Etminan, H. 540 Migachyov, I. F. 251
Evstrakhin, V. A. 251 Moine, B. 443
Fan, P.-F. 179 Moore, C. 559
Foose, M. 8 Naldrett, A. J. 91
Friedrich, G.H. 572 Narkelyun, L. F. 513
Gablina, I. F. 524 Nikolayeva, E. I. 261
Garamzhav, D. 271 Ohmoto, H. 412
Genkin, A.D. 111 Ortoleva, P. 559
Glybovsky, V. O. 504 Pashkov, Yu. N. 209
Godlevsky, M.N. 124 Proskuryakov, V. V. 171
Golovanov,I.M. 261 Rekharsky, V.I. 209
Gray, K.G. 304 Ripley, E. M. 25, 559
Gresham, J. J. 63 Rose, A. W. 412
Groves, D.1. 43 Ryabov, G. V. 350
Guilbert, J.M. 196 Salikhov, V. S. 513
Guilloux, L. 443 Saryanov, Yu.A. 271
Hanski, E. 151 Schmidt, F.-P. 572
Haranczyk, C. 461 Schumacher, C. 572
Hoy, L. D. 412 Scott, S.D. 370
Kalogeropoulos, St.1. 370 Shcheglov, V.1. 350
Kazikhin, M. A. 261 Sillitoe, R. H. 235
XIV List of Contributors

Skripchenko, N. S. 350 Varyash, L. N. 209

Smith, A. T. 412 Volodin, P.N. 513
Sood, M. K. 422 Wagner, R.J. 422
Sotnikov, V. I. 271, 291 Weiblen, P. 8
Spieth, V. 572 Yakovlev, Yu.N. 135
Strauss, G. K. 304 Yakovleva, A. K. 135
Susura, B. B. 504 Zaskind, E. S. 171
Trubachev, A. I. 513 Zhamsran, M. 271
Tvalchrelidze, A. G. 358
Introduction

The symposium on copper deposits and the publication of its Proceedings are a
good example of international scientific cooperation. The extensive exchange of
opinions between specialists in different types of copper deposits has provided
important new data on their geological setting and condition of formation.
A remarkable number of authors writing about Cu - Ni deposits in Part I of
this volume reach the conclusion that these are the consequence of intet;'action of
mafic or ultramafic magma and near-surface country rocks. Foose and Weiblen
draw attention to the contrast between the organized nature of the unmineralized
rocks forming the main part of the Kawishiwi intrusion of the Duluth Complex,
which have crystallized as a consequence of repetitive influxes of new magma,
and the heterogeneous nature of the sulfide-bearing rocks, which show indica-
tions of having reacted with volatiles to varying degrees in different places. The
volatiles are believed to have been derived from country rocks, and a model is
presented of rocks intruding a dynamic tectonic environment during rifting.
Ripley's S, 0, Hand C isotopic data also indicate that devolatilization of country
rocks has accompanied the introduction of sulfur to mafic intrusions at Duluth.
At the Dunkha Road deposit, this has occurred in situ, but at Babbitt, devolatil-
ization and partial melting of country is believed to have occurred in the feeder to
the present deposit. Lesher and Groves emphasize the association of mineralized
komatiites with sulfide-bearing sediments amongst the country rocks. They point
to the ability of komatiites to erode thermally sediments and basalts on which
they extrude, and propose that erosion and assimilation of sulfide-bearing
sediments makes a major contribution to the sulfides of the ore zones, while
erosion of basalt produces the structures in which the ores reside. Gresham takes
a different view, stressing the contrast between ore-bearing and ore-free zones of
komatiites, and proposing that the ore-bearing units are related to rift-produced
irregularities in the pre-komatiite volcanic topography. Reporting a study of the
Sudbury structure and its rocks, Naldrett demonstrates that the Sudbury magmas
have assimilated in excess of 500/0 of country rocks. He proposes that this
massive contamination has been responsible for "salting out" sulfides from
mafic magma, thus accounting for the high incidence of deposits at Sudbury.
In contrast to these views emphasizing a near-surface origin for immiscible
sulfides, both Distler, Genkin, and Dyuzhikov as well as Godlevsky and Lik-
hachev emphasize the separation of liquid sulfides from associated mafic magma
at depth. All of these authors propose that sulfide-rich magma was introduced
separately to the main mass of the associated intrusion. Distler et al. explain
2 Introduction

zoning at Talnakh partly as the consequence of immiscibility between two sulfide

liquids. Godlevsky and Likhachev believe that the shape of a mafic intrusion,
chonolithic or lacolithic, depends on the relative density of the magma with
respect to its country rocks, which in turn is a function of how much sulfide the
magma contains.
Yakovlevand Yakovleva emphasize the magmatic origins of deposits of the
Kola peninsula, pointing out that metamorphism has changed them, but has not
generated any deposits on its own. Hanski draws parallels between the gabbro-
wehrlite rocks hosting the Proterozoic Pechanga deposits of Kola and those of
the Archean Kuhmo belt in Finland, but points out that the latter have distinctive
chemical compositions.
Kochnev-Pervukhov, Zaskind and Proskuryakov compare komatiites of the
SE Baltic shield with those elsewhere and relate the presence of mineralization to
the composition of the lavas and to their tectonic setting. Fan describes for the
first time in western literature chromite and nickel deposits of the People's
Republic of China.
The section on porphyry copper deposits, Part II, commences with Guilbert's
overview of recent progress in the understanding of porphyry base-metal deposits
in the western Americas.
Emphasis is placed on the plate-tectonic settings of porphyry deposits, and on
the histories, compositions, sources and flow patterns, and physicochemistry of
the mineralizing fluids. A Soviet view of selected physicochemical parameters in-
volved in porphyry copper formation is then presented in two contributions by
Kudrin, Varyash, Pashkov, and Rekharsky, and by Sotnikov and Berzina. Using
northern Mexico as an example, Damon presents a hypothesis to explain ob-
served spatial and temporal relationships between batholiths, stocks, and por-
phyry copper deposits. Sillitoe explores the metallogenic implications of the
space-time distribution, crustal setting and Cu/Mo ratios of porphyry copper
deposits in the central Andes. In contrast, Evstrakhin, Krivtsov, and Migachyov
provide a lithotectonic rather than a plate-tectonic analysis of selected porphyry
copper provinces in the USSR. Golovanov, Nikolayeva, and Kazikhin appraise
the geologic controls of high-grade core zones in porphyry copper deposits, using
the Almalyk deposits in the USSR as an example. The porphyry copper section is
concluded with descriptions of the regional and local geology of the Erdenetuin-
Obo deposit in Mongolia, and Zhamsran, Sotnikov, Berzina, Garamzhav,
Saryanov, and the Recsk deposit in Hungary by Baksa.
Part III was planned to comprise a series of papers dealing with sulfide ore
deposits hosted in volcanic and volcano sedimentary complexes, characterized
metallurgically, chiefly by copper, with varying proportions of zinc and lead,
with or without precious metal contents. Mineralogically, the deposits are dom-
inated by the iron sulfides, pyrite and/or pyrrhotite.
These ores, as envisaged by the conveners of the Session, are mainly of the
stratabound volcanic-exhalative type and comprise both concordant, often strati-
form, often massive, components and cross-cutting, disperse to semi-massive,
components representing feeder zones to the massive ores.
By mutual agreement, magmatic ores characterized by Ni - (Cu) occurring in
mafic and ultramafic volcanic sequences (Kambalda type) were not included in
the papers chosen for this Session.
Introduction 3

The final selection of papers published here is, through various causes, much
less comprehensive than those originally contemplated, and does not represent all
the papers delivered during Session III at the 27th IGC in Moscow on August
1984. The coverage of volcanite-hosted stratabound sulfide ores of
Cu - (Zn - Pb) as defined above is admittedly very patchy, both as regards geo-
graphical distribution and subject matter.
However, the papers now presented cover a fair range of aspects of the ores
defined above. There are descriptions and discussions of two important
European stratabound sulfide ore districts, as well as a group of papers which
discuss central genetical aspects of the ores, touching on such matters as
physicochemical conditions of formation and the causes of chemical and
mineralogical zoning both on district- and individual ore deposit scales.
Of the ore district descriptions, the chapter by Strauss and Gray deals with
one of the classical copper ore fields of western Europe, where exploitation
began in pre-Roman times, viz., the Iberian Pyrite Belt.
The chapter is in the form of a general review of the geology of the pyrite belt
and of the main ore deposits, including a description of the various types of
sulfide ore encountered. The ores are considered to be closely linked with a
single, submarine, acid-alkaline volcanic cycle within a volcanic-sedimentary
complex of Upper Visean to Lower Westphalian age.
The chapter by Kurbanov analyses the metallogeny of two pairs of ore-bear-
ing zones in the Caucasian segment of the Mediterranean fold belt, each
apparently characterized by sulfide ores of different paleotectonic settings and
related to different types of volcanism. The relation of the pyritic deposits to por-
phyry and vein-type copper deposits can be shown.
The chemical and mineralogical zoning which is exhibited by base-metal-
bearing stratabound sulfide ores in many parts of the world is dealt with in other
chapters by Russian authors. Eremin discusses what he refers to as the "hidden
mineralogical and geochemical zoning", which he recognizes in ore deposits from
USSR and Japanese ore districts. Two types of zoning are distingushed;
repeated, exhibited by hydrothermal replacement ores, and nonrepeated, ex-
hibited by ores of volcanic-sedimentary origin. The latter type of zoning is at-
tributed to variations in temperature and sulfur activity during ore deposi-
tion.
Skripchenko, Shcheglov, Ryabov, and Andreyev also discuss lateral and
vertical zoning shown by individual ores hosted in volcanic and sedimentary
sequences. Two models to explain this zoning are presented, relating respectively
to volcanite-hosted and to sediment-volcanite-hosted ores.
The origin of base-metal zonation of sulfide ores deposited on the sea floor is
also touched on by Tvalchrelidze in a consideration of the physicochemical con-
ditions of their formation. This author assigns an important role here to the
reaction between pore waters and sulfide rich sediments during diagenesis.
Kalogeropoulos and Scott in contrast to the preceeding chapters, deal with a
specific restricted zone of one of the ores of the famous Hokuroku ore district in
the Eocene Green Tuff Region of Japan. These authors interpret the textural and
chemical evidence they find in the barite zone at the Fukazawa mine as indicating
the mixing of a circulating barium-rich hydrothermal solution with cold sea
4 Introduction

water within an unconsolidated pyrite tuff. Of especial interest in respect to the

genesis of these ores is their evidence, reinforcing earlier observations, of con-
tinuing hydrothermal activity after the "post-ore" hanging wall rocks had been
deposited.
Part IV contains 14 chapters from 42 authors and co-authors, dealing with
two main subjects. First, mineralogical, petrographical and chemical studies
(Brown and Chartrand, Cailteux, Rose et al., Sood et al., Moine et al.,
Haranczyk and Lur'ye); second, a detailed description and synthesis of copper
deposits including genetic interpretations (Bogdanov, Susura et al., Krendelev et
al., Gablina and Lambert et al.). In addition, results of thermodynamical inves-
tigations on sulfide zoning by Merino et al. and results of recent exploration of
Kupferschiefer-type deposits are presented by Schmidt et al.
Brown and Chartrand interpret the sequence of mineralization in the
sediment-hosted copper deposits at White Pine (Michigan), Redstone (North-
West Territories), and Kamoto (Zaire). They suppose an influx of dissolved
copper into initially pyritic basal units of grey-bed host rocks during early
diagenesis. The plausible sources of copper include the red beds themselves, or
pene-exhalative solutions which "exhale" into the red beds.
Cailteux points out the stratiform copper-cobalt mineralization in the Shaban
Mines Group with its micro- and macrocycles. This mineralization results from
diagenetic processes involving a mixing of stagnant interstitial water with brines
which were enriched in metals and migrated through still porous sediments. He
additionally gives reaction equations which characterize secondary reactions
taking place during chalcocitization and carrolitization.
Rose, Smith, Lustwerk, Ohmoto, and Hoy point out geochemical aspects of
stratiform redbed copper deposits in the Catskill formation, Pennsylvania, and
Redstone Area, North-West Territories. They suppose that the copper minerals
were emplaced after earliest diagenesis, but before induration and formation of
quartz overgrowth. The copper has been redistributed, depleted from nearly all
red (oxidized) sediments. After mobilization of copper from the red beds it has
been transported by chloride-bearing pore fluids of marine or evaporative origin.
This flow was driven by compaction, sea-level changes, and gravity flow down
the alluvial plain.
Sood, Wagner, and Markazi describe the characteristics and origin of Kenne-
cott-type copper deposits in the McCarthy Quadrangle, Alaska. They suppose a
liberating of copper from the metal oxides and the mafic minerals by hydration-
dehydration reactions associated with metamorphism. The circulation and
transport of metals took place in meteoric water. The activity in such systems has
been driven by the heat of Tertiary intrusives.
Moine, Guilloux, and Audeoud investigated the major elements of the host
rocks in some Kupferschiefer-type deposits at the Zambian Copperbelt. They
presume that the high contents of Mg and K-feldspar cannot be contained by
sedimentary processes only and that a large part of these is probably of dia-
genetic/hydrothermal origin.
Haranczyk shows the complexity of processes active in forming stratiform
copper deposits. He gives new data and new ideas, e.g., maps of redox state and
paleogeography.
Introduction 5

Lur'ye presents a chapter summarizing new data of formation conditions of

copper sandstone and shale-type deposits. He points out the relation of ore thick-
ness-grade to underlying lithology and the importance of hydrodynamics, com-
bined with some data of flow rate and rock porosity. Additionally, he believes
that there are no copper occurrences where flows of high copper concentration
killed sulfate-reducing bacteria.
Bogdanov described major types of copper-bearing zones in the Soviet
Union, including orogenic copper-bearing zones, copper-bearing areas in
platform regions, and copper-bearing zones of geosynclinal areas. Additionally,
he gives information on major epochs of copper accumulation and a prognostic
evaluation of copper-bearing areas.
The chapter of Susura, Glybovsky, and Kislitsin shows interesting ideas on the
processes forming the Dzhezkazgan ores, and a summary of sedimentary tectonic
environments for numerous sedimentary copper deposits in the western USSR.
Krendelev, Narke/yun, Trubachev, Salikhov, Volodin, Kunitsin, Chechetkin,
and Bakun show the variety of mineral deposits in the cupriferous sandstones
and shales of the Siberian Platform and distinguish six major copper provinces.
Within these at least 15 major zones of deposits and occurrences are recognized,
occurring in Proterozoic and Paleozoic detrital sediments. The sources of copper
are considered to be in the uplifted blocks that supplied the sediments and in
volcanic effusives.
Gablina gives an overview of different genetic types of copper mineralization
in the Igarka area/Soviet Union. She describes very interesting geological, sedi-
mentological and mineralogical relations of significant copper ores and low
grade mineralization.
Lambert, Knutson, Donnelly, and Etminan point out a synthesis of features
and genesis of various types of sediment-hosted copper Deposits in Australia.
Besides hydrothermal fluids associated with igneous activity and regional meta-
morphism (e.g., Mt. Isa) they describe the widespread, bedded-disseminated
mineralization, formed at relatively low temperatures by reactions of cupriferous
fluids with biogenic HzS and early diagenetic iron sulfides.
Merino, Moore, Ortoleva, and Ripley give a quantitative kinetic approach,
based on systems of thermodynamical and differential equations and equilibrium
activity diagrams. The zoning copper/chalcocite/bornite/chalcopyrite/pyrite is
portrayed by a log!<h versus pH diagram at 100°C, 0.1 m Cl- and 10 -4 m total
sulfur. It is shown that in the chalcocite field most of the sulfur must be sulfate,
and the maximum total of copper concentration is only 0.1 mg/l in the chalco-
pyrite field.
The chapter of Schmidt, Schumacher, Spieth, and Friedrich combines scien-
tific investigations (geochemical, mineralogical, petrographical, geological) with
results of actual exploration for copper-silver deposits in the Kupferschiefer of
West Germany. The relationship between paleoenvironment, the metal distri-
bution, and the diagenetic process of the "Rote Faule" is shown.
The results of the chapters included in Part IV, indicate a new way to consider
the problems of sediment-hosted copper deposits. All authors assume underlying
red-bed formation as a plausible source of copper. The controlling factors of
mobilizing and transporting copper may be compaction, sea-level changes,
6 Introduction

gravity flow, hydrodynamic activities, the influence of heat of intrusives and

metamorphism. The infiltration and precipitation of the copper-bearing flows
(subsurface waters) took place during diagnesis and not during sedimentation.
Part I
Deposits in Mafic and Ultramafic Complexes
The Physical and Petrologic Setting and Textural and
Compositional Characteristics of Sulfides from the South
Kawishiwi Intrusion, Duluth Complex, Minnesota, USA
M. FOOSE 1 and P. WEIBLEN 2

Abstract

The South Kawishiwi intrusion in the northern part of the Duluth Complex,
Minn. is divisable into a sulfide-free upper zone and a sulfide-bearing lower
zone. The sulfide-free zone is mostly troctolite containing laterally persistent
anorthosite layers. Sequences of troctolite and anorthosite form cyclic units in
which the crystallization of plagioclase was followed by precipitation of plagio-
clase plus olivine. Chemical variations associated with cycles show trends either
similar to or reversed from those expected under normal conditions of fractional
crystallization. These trends, and the absence of any systematic chemical
evolution, indicate that crystallization of the sulfide-free rocks occurred in a
chamber that was continuously replenished by compositionally similar liquids. In
contrast, the sulfide-bearing zone consists of a heterogeneous mixture of
troctolite, picrite, dunite, anorthosite, oxide cumulates, and hornfels in which
sulfides are ubiquitously disseminated. The sulfide-bearing zone has no dis-
cernible internal stratigraphy. The contact with the sulfide-free zone is sharp.
The sulfides are predominantly pyrrhotite, chalcopyrite with cubanite, and
pentlandite occurring mostly as interstitial and included grains. The close asso-
ciation of sulfides with hydrous phases or with grains that show reverse composi-
tional zoning indicates that volatiles derived from adjacent metasedimentary in-
clusions were important in sulfide formation. The wide dispersion of sulfides,
heterogeneous mix of rock types, and absence of traceable units indicates that the
rocks accumulated in an extremely dynamic environment that is thought to have
been associated with rifting in this area. Faults presumed to be rift-related were
active throughout the emplacement of this intrusion and, in part, control the dis-
tribution of the sulfide-bearing zone and the overlying rocks.

Introduction

The 1.1 Ga Duluth Complex in the north-central United States (Fig. 1) is a large
composite mafic intrusion that contains significant amounts of copper-nickel

I US Geological Survey, Reston, VA 22092, USA

2 Minnesota Geological Survey, Minneapolis, MN 55455, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et aL)
© Springer-Verlag Berlin Heidelberg 1986
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 9

92°

® Ely 0 IOKM

CJ Undivided rocks. mostly troctolite

N

0 Bold Eogle intrusion

~ South Kowishiw i intrusion

D.. Partridge River troctolite

'.
GJ Ra ilroad troctolite

a Powerline gabbro

[] Anorthositic series 47"45'

Hornfels

D Ma jor Cu-Ni

47"

oI
50KM
,

Fig. lA, B. Index (A) and generalized geologic map (B) of the Duluth Complex near Ely, Minnesota,
USA. Also shown are major copper-nickel sulfide occurrences (shaded), the outline of the area
shown in Fig. 2, and location of drill holes shown in Figs. 3 and 4. Location of samples from which
sulfur isotopic compositions of + 7.5 and + 1.7 per mil were collected are shown by solid square and
solid triangle, respectively. (Modified from Morey and Cooper 1977, and Watowich et al. 1981)

sulfides. The full extent of the resource is unknown, but along a 50-km distance
at the base of the complex southeast of Ely, Minn., over 4 billion metric tons of
mineralized rock containing 0.66% copper and 0.2010 nickel has been identified
(Listerud and Meineke 1977). In addition to their economic potential, these
sulfides are of interest because they are one of the few large concentrations of
magmatic sulfides in non-Archean rocks (Naldrett 1981). This contribution
describes the physical and petrologic setting of sulfide-bearing rocks and the
textures, types, and compositions of sulfide minerals that occur within the South
Kawishiwi intrusion which is one part of this composite complex.
10 M.FooseandP. Weiblen

Geologic Setting

The early general studies of Grout (1918a, b) and later more specific work of
Taylor (1964), Green et al. (1966), Bonnichsen (1970, 1972a), Phinney (1969,
1972), and Weiblen and Morey (1975, 1980) show the Duluth Complex to be a
composite intrusion composed predominantly of anorthositic, trocolitic, and
gabbroic rocks. These intrusions are divisable into two series. Anorthositic rocks
of the older anorthositic series are enclosed in and cut by younger rocks that are
predominantly troctolites which contain minor interlayered anorthosite. These
younger rocks are known as the troctolitic series (Bonnichsen 1972a). These two
series intrude diverse Archean and Proterozoic volcanic, metasedimentary, and
igneous rocks that define the northwest margin of the complex. In the northern
part, the footwall rocks are either 2.7 Ga granites of the Giants Range Batholith,
Early Proterozoic Biwabik Iron Formation, or Early Proterozoic Virginia
Formation. To the southeast, the intrusion is overlain by the North Shore
Volcanic Group.
Emplacement of the Duluth Complex occurred about 1.1 Ga ago (Goldich et
al. 1961, Faure et al. 1969) during an episode of regional crustal rifting which
produced what is now referred to as the Midcontinent rift (Wold and Hinze
1982). This rift, as expressed by the Midcontinent gravity high, extends over
1,000 km from Lake Superior to northeast Kansas (King and Zietz 1971, Chase
and Gilmer 1973). The Duluth Complex is located on the northwest flank of this
gravity anomaly. Magma was injected into fault-bounded voids formed during
rifting; crystallization within these separated magma chambers resulted in the
multiple intrusions that make up this composite complex (Weiblen and Morey
1975, 1980).
In the part of the Duluth Complex southeast of Ely, Minn. (Fig. 1) rocks of
both the older anorthositic series and the younger troctolitic series are exposed.
The latter form at least three intrusive bodies which have been given the informal
names of the South Kawishiwi intrusion (SKI), the Partridge River troctolite
(PRT), and the Bald Eagle intrusion (Green et al. 1966, Bonnichsen and Tyson
1975, Weiblen 1965). The Bald Eagle intrusion is a funnel-shaped body (Weiblen
1965) that clearly cross-cuts anorthositic series rocks. The SKI and the PRT, in
contrast, are adjacent and compositionally similar bodies that lack a clearly rec-
ognizable mutual contact and which are identified as separate intrusions prin-
cipally because of slightly different textures.
In this region, significant amounts of copper-nickel sulfide have been located
only along the base of the SKI and the PRT. Both general descriptions of these
sulfides (Bonnichsen 1972b, 1974, Listerud and Meineke 1977, Watowich et al.
1981, Iwasaki et al. 1983) and specific studies (Boucher 1975, Fukui 1976,
Weiblen and Morey 1976, Bonnichsen et al. 1980, Tyson and Chang 1984, and
Pasteris 1984) suggest that sulfides in both intrusions have similar mineralogy
(dominantly pyrrhotite with less abundant chalcopyrite and cubanite, and minor
pentlandite), textures (dominantly interstitial or included), and mode of
occurrence (disseminations associated with the footwall or inclusion-rich zones).
The composition of the sulfides do, however, change along strike as the copper-
nickel ratios increase toward the southwest (Fig. 1) from the Spruce Road deposit
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 11

(2.7), to the Maturi deposit (3.2), to the Minnamax deposit (4.1), to the Dunka
Road deposit (5.1) (Tyson and Chang 1984).

Petrologic and Structural Setting of Sulfides

The combination of large amounts of surface and subsurface information from

the SKI allows the development of an integrated petrologic and structural
picture. These relationships are summarized below.

Surface Relations

In outcrop, the SKI is composed mostly of plagioclase-olivine cumulates contain-

ing minor (less than 100/0) interstitial augite, oxides, or biotite (Bonnichsen and
Tyson 1975, Foose and Cooper 1978, 1981). Madallayering is not common and
generally is not traceable for more than 100 m. Even less common are phase
layers made by plagioclase cumulates. However, many of these phase layers are

Troctolites of the
/ Faults
Bald Eagle Intrusion

D Undivided anorthosites

~ Troctoite with minor

Intercumulus pyroxene
and oxide

lZ:J Traceable anonhOSlle layer

~ Troctoile containing
pyroxene oikocrysts

[Z] Troctoite wrth 3 · tO

percent IIltercumulus
pyroxene and oxides

~ Anorlhosllic series

o I KlOMrnR
~

Fig. 2. Generalized geologic map of area outlined in Fig. 1

12 M. Foose and P. Weiblen

laterally persistent and can be followed. In areas of good exposure, mapping

these layers enables the geologic relationships to be established (Fig. 2).
Most important, the offsetting of these layers shows this intrusion to be
highly faulted (Foose and Cooper 1981). Faults strike in three principal direc-
tions (N 35°E, N 05°W, and N 40 0 W), and generally have small offsets. The
general lack of offset of one fault set by others indicates that most displacements
involved nearly vertical movement of blocks. The faulting parallels regional
lineament and joint patterns that have been related to east-west tension (Cooper
1978). The overall displacement, in which movement of blocks is down to the
southeast is consistent with the expected motion on the northwest flank of a rift.

Subsurface Relations

Drill holes near the surface exposures allow projection of these relationships into
the subsurface. Three major lithologic zones are recognized (Fig. 3). From
bottom to top: (a) older, granitic footwall rocks, (b) basal zone sulfide-bearing
rocks, and (c) generally sulfide-free (i.e., found in trace amounts only) troctolite
with interlayered anorthosite (Bonnichsen et al. 1980, 1984). Contacts between
these zones are the only structural and stratigraphic features which have been
traced with certainty.

Sulfide-Free Zone. As on the surface, the upper sulfide-free zone is composed

predominantly of plagioclase-olivine cumulates with thin plagioclase cumulate
interlayers. In addition, some pegmatoids are present. Most pegmatoids are thin
( < 10 cm) and consists of 50 to 90070 plagioclase crystals that are 1 to 4 cm in
length with 10 to 40% coarse-grained clinopyroxene, and minor olivine.
Typically these pegmatoids form bottoms of 2 to 40 m thick laterally persistent
successions which grade upward into medium-grained plagioclase cumulates and
then into plagioclase-olivine cumulates. These successions define packages of
rocks which can be correlated among drill holes; correlation of individual succes-
sions within these packages is less certain (Fig. 3). The most clearly traceable
layer is a 30 to 100 m thick plagioclase-rich and commonly pegmatoidal section at
the base of the sulfide-free zone which is in sharp contact with the underlying
sulfide-bearing zone.
The pegmatoids that form the bottoms of these successions are not always
present and, in fact, decrease in abundance to the northeast (Fig. 3). Where
pegmatoids are absent, the bottoms are made by medium-grained plagioclase
cumulates. The sharp lower contact between the base of these successions and
underlying plagioclase-olivine cumulates" indicates a break at which the magma
changed from crystallizing plagioclase and olivine to crystallizing only plagio-
clase. The gradation of plagioclase cumulates into overlying plagioclase-olivine
cumulates show that olivine eventually rejoined plagioclase on the liquidus.

Sulfide-Bearing Zone. In contrast to overlying rocks, the sulfide-bearing basal

zone is a heterogeneous, 70- to 350-m thick sequence of rocks. The most abun-
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 13

17

METERS
8
o 13

16

12

1:-:':':':':'1 Olivine-rich TrocToliTe

c::::::J TrOCTol iTe
amnm Anorthosite
c==J PegmoToid
(;sss;J Sulfide-beoring zone
~ FooTwoll

2 KILOMETERS
L-_....1..._---',

Fig. 3. Generalized subsurface stratigraphy of the South Kawishiwi intrusion. Drill hole locations are
shown here in inset and in Fig. 1. Major sequences of rocks are correlated by solid lines and are:
I sulfide-bearing zone; II anorthositic and plagioclase-rich pegmatoidal layer; III a laterally discon-
tinuous olivine-rich troctolite; IV troctolite with few anorthositic layers; V troctolite with many
anorthositic and pegmatoidal layers; VI upper troctolitic sequence having rare pegmatoidal zones.
The base of the sulfide-free zone defines the surface to which all holes are referenced. (Modified from
Foose 1984)
14 M. Foose and P. Weiblen

dant rock type is medium- to fine-grained troctolite; less common are picrites,
norites, anorthosites, and thin layers of oxide cumulates. Hornfels inclusions are
abundant, and most of the unit contains disseminated sulfides. The sulfide zone
lacks a coherent internal stratigraphy and even distinctive lithologic breaks
cannot be traced between closely spaced drill holes. The sulfide-bearing zone is
distinguished from overlying rocks by its internal heterogeneity, generally greater
olivine content, large number of hornfels inclusions, presence of oxide
cumulates, and ubiquitous sulfides.

Geochemical Trends

The pegmatoid-anorthosite-troctolite successions in the sulfide-free zone

provides important information about the emplacement and evolution of this
intrusion. Mineral and whole rock analyses show these sequences to have
compositional trends that are either similar to, or reversed from those expected
under normal conditions of fractional crystallization (Fig. 4).
Plots of whole-rock nickel show widely variable and relatively high values in
the basal zone, reflecting the sporadic distribution of sulfides. In the overlying
sulfide-free zone, nickel is partitioned mostly into olivine, and the whole-rock
nickel content in these rocks is controlled by the abundance of olivine and by the
amount of nickel in olivine. Because nickel content in olivine is, in turn, con-
trolled by the nickel content of the magma from which it crystallized, variations
in whole-rock nickel content among rocks having similar amounts of olivine
reflect changes in magma composition. Nickel contents from compositionally
similar rocks that the have between 70/0 and 15% olivine are shown separated
(dots) from more plagioclase - or olivine-rich rocks (slashes) and define the
trends shown by dashed lines (Fig. 4).
Nickel values define two distinct patterns within rocks above the sulfide-
bearing zone. Depletion patterns that are expected in normal fractional crystal-
lization in which nickel content decreases upward are more common. Segments
in which nickel content increases upward are somewhat less common. These
nickel variations parallel trends in mineral compositions as shown by detailed
study through a number of these cycles (Foose 1982).
Similar normal and reversed compositional trends have been noted both in
the SKI and in the adjacent PRT (Fukui 1976, Bonnichsen et al. 1980, Grant and
MoIling 1981, Tyson and Chang 1984). Trace element studies in the PRT by
Grant and MoIling (1981) indicate that some of the reversed trends result from
variations in the proportion of intercumulus liquids. Grant and MoIling suggest
that larger amounts of interstitialliquid$ are trapped near bottoms of cycles and
that subsolidus equilibration with these interstitial liquids causes shifts to com-
positions that are more sodic and iron-rich than in overlying rocks where less
intercumulus liquid was present.
However, subsolidus equilibration with interstitial liquids does not appear to
have caused the reversed patterns seen in Fig. 4. Strongly zoned minerals and
large amounts of various intercumulus phases are present in the pegmatoids and
 MElERS
o
o 300 600 900 1200 1500 1800 2400 2700
\ , o-j
1\1 2100 :r
+ 01 • r- '"1:1"
o \ :r
'<
o
1 ~
\ .~ •I 1
\ r • I ['"
~ I»
\ '. I I.
I -.-- I 8-
.\ V····.. ··I "1:1
t \ ~
I Oliv ine·rich ...
I
J
I . I
, \ 'I
J. troctolife
o
0"
100 I I ra.
I c:J o
~ I I I. Trocto lite w ith
./ I I 7· 15 percent
., J\l I I· ol ivine (JQ
~
'J~;. L· I r· I»
]r: I hr••1 op..
I I Olivine - poor
~ Iroctolite
l' .1 ~
IUlIIIID S-
. -1- \ Anorthosite a
r -riC'""': I
I CJ ~
p..
-- . L = I. Plagioclase-rich ("')
200 f pegmatoid o
I 8
\ 01 ~ '0
• o
., I Sulfide·bearing ~.
I zone o·o
-c.:' J: t' I - _ I. E'Z.ZZa e?.
("')
. I . Footwall :r
) I»

\ I" ... ~
..' (I)

r I ::J.
u..........u... ~ u..........u... L.u........L... r......1...L... e.o
o 500 o 500 o 500 o 500 o SOC '"
o
-.
JOO NICKEL CONTENT (PPM) [I)
c:
Fig. 4. Whole-rock nickel content and detailed stratigraphy of drill hole 12. Nickel content of rocks having between 70/0 and 15% olivine (dots) are 5
shown separately from more plagioclase- or olivine-rich rocks (slashes) and locally define trends shown by thin dashed lines. Dashed horizontal lines ~
mark distinct breaks in crystallization. (Scale in meters; tic marks and numbers in feet. Analyses on splits from drill core made by atomic absorption
.-
methods in USGS laboratory; W.D. Angelo analyst) v.
16 M. Foose and P. Weiblen

clearly show that large amounts of intercumulus liquid were locally present at the
bottoms of some cycles. However, the overlying anorthosite and troctolite are
virtually unzoned adcumulates, indicating relatively small and uniform amounts
of trapped interstitial liquids. Compositional changes within them, therefore,
cannot be easily attributable to varying amounts of sub solidus equilibration.
Further, in troctolites that have between 7 to 15D7o olivine, there is good correla-
tion (coefficient = 0.782) between whole-rock nickel content and olivine MgO
content, whereas there is no correlation (coefficient = - 0.104) between whole-
rock nickel and phosphorous contents. Because phosphorous is an incompatible
element that is concentrated in the intercumulus phase, a close correlation of Ni
and P would be expected if varying amounts of intercumulus liquids had con-
trolled olivine compositions and thereby caused the trends shown in Fig. 4.
It is more likely that formation of the diverse cyclic patterns resulted from the
injection and mixing of liquids of slightly different compositions. Episodic injec-
tions of liquids more primitive than those within a magma chamber could
produce repeated patterns of decreasing nickel content upward. In contrast, in-
creases upward in nickel, Ca/Na, and Mg/Fe could indicate crystallization from
injections of more fractionated liquid that were mixed with the larger and com-
positionally more primitive magma reservoir.
Difficulties in generating, storing, and then emplacing more fractionated
liquids into a magma chamber that also would be recharged with batches of more
primitive liquids would seem to make injections of batches of more fractionated
liquids unlikely. However, because the separate intrusions that constitute the
Duluth Complex were emplaced and crystallized in separate fault-bounded
chambers created during rifting (Weiblen and Morey 1975, 1980), it is possible
that compositionally similar liquids in adjacent magma chambers fractionated to
differing degrees. Disruption of these isolated magma chambers by continued
rift-related faulting could allow mixing of more evolved and more primitive
liquid batches between adjacent chambers.
Support for injection of liquids from adjacent magma chambers is given by
the distribution of cyclic sequences (Fig. 3). Cycles are more abundant in the
southwestern part of the SKI and become less common to the northeast. Accom-
panying this lateral change is a decrease in abundance of pegmatoids at the
bottoms of cycles. The compositionally similar PRT is present along the south-
west boundary of the SKI and may have been a source for injections of diverse
batches of liquids. As observed, abundant thin, but not laterally extensive cycles
would be expected near this source area, while thicker and more laterally con-
tinuous cycles would extend to the northeast. Furthermore, the rapid rates of
crystal accumulation expected near areas of magma injection would produce
conditions ideal for trapping intercumulus liquids and the formation of the
coarse-grained bottoms of cycles. Less rapid accumulation of crystals further
from the source of magma injection would produce the more texturally uniform
adcumulates that occur to the northeast.
Finally, one should note that mineral compositional changes within in-
dividual cycles often exceed the overall average compositional change within the
entire exposed portion of SKI. This lack of systematic chemical differentiation
provides further evidence that this portion of the intrusion crystallized in an open
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 17

magmatic system in which there was nearly continuous replenishment of

compositionally similar liquids.
Mapping on the surface shows that significant faulting occurred after crystal
consolidation. Additionally, some fault-related fractures are filled with pyrox-
ene, indicating that movements occurred prior to complete crystal consolidation.
However, the subsurface relationships reveal the importance of syn-intrusion
faulting in controlling the emplacement of these rocks. For example, between
drill holes 12 and 17, the orientation of the basal contact of the intrusion changes
abruptly (Figs. 1 and 3), and the basal zone thickens to over 300 m. Overlying the
thicker part of the basal zone is a laterally discontinuous olivine-rich troctolite
layer and a thicker sequence of overlying troctolite. These changes indicate that
faults were active at the start of the emplacement of the intrusion and formed a
fault-bounded trough that accumulated a thicker section of the sulfide-bearing
basal zone. The discontinuous lens of olivine-rich troctolite, and thicker
sequences of overlying rocks show that faulting continued during the emplace-
ment of the overlying sulfide-free zone.

Characteristics of Sulfides

Three major concentrations of sulfides have been recognized in the troctolitic

and anorthositic successions that form the lower parts of the SKI and the ad-
jacent PRT. These are the Spruce Road deposit, the Maturi deposit, and the Min-
namax deposit (Fig. 1). All have broadly similar mineralogy, textures, and style
of mineralization. Most of the specific features discussed here are in the Spruce
Road deposit (Weiblen and Morey 1976), but reference is also made to features
that have been observed elsewhere. The Spruce Road deposit has proven re-
sources of 248 million metric tons grading 0.40/0 eu and 0.17% Ni (Watowich et
al. 1981). The sulfides are mostly in a medium-grained plagioclase-olivine cu-
mulate that contains variable amounts of interstitial augite, orthopyroxene, iron-
titanium oxides, and biotite. Particularly important is the relatively high biotite
content in the sulfide-bearing rocks as compared with that of the overlying,
unmineralized rocks (Weiblen and Morey 1980). Additional rock types are cross-
cutting augite troctolite pegmatoids, mafic troctolite, picrite, and abundant
hornfels. Because of the great numbers of inclusions, the Spruce Road area may
best be described as an igneous breccia.
The sulfides may be grouped into four textural types: (1) interstitial sulfides;
(2) included sulfides; (3) sulfide-silicate intergrowths; and (4) fine-grained sulfide
veinlets. Representative photomicrographs of these sulfides are shown in Fig. 5.

Interstitial Sulfides. The great majority of sulfide grains occupy roughly

triangular shaped primsatic voids that are bounded by a plagioclase framework
and less commonly by anhedral olivine. The void fillings may be entirely sulfide
or a mixture of silicates, mostly clinopyroxene plus ilmenite, biotite, potassium
feldspar, and apatite. Particularly common are void fillings formed by frame-
works of plagioclase. Where plagioclase forms the boundary of the voids, it com-
18 M. FooseandP. Weiblen

Fig. SA - D. Sulfide textures. A Interstitial sulfides between plagioclase (bar is 1 mm). B Included
sulfides in plagioclase. Sulfides are mostly chalcopyrite and are concentrated along 001 (bar is
100 11m). C Myrmekite-like intergrowths of sulfide with orthopyroxene (bar is 200 11m). D Fine-
grained veinlets of chalcopyrite cutting interstitial clinopyroxene (bar is 100 11m). Cb, cubanite; Cp,
chalcopyrite; Po, pyrrhotite; Pn, pentlandite; Mk, Mackinawite; PI, plagioclase; Cpx, clinopyrox-
ene; Opx, orthopyroxene; 01, olivine; Sp, serpentine. (After Weiblen and Morey 1976)
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 19

Table 1. Size distribution of interstitial and included sulfides a

Phase Length (mm) Width (mm)

Min Max Av Min Max Av

Chalcopyrite
Interstitial 0.00100 1.29000 0.11979 ± 0.20878 0.00100 0.48000 0.04783 ± 0.07542
Included 0.00067 2.64000 0.06635 ±0.17783 0.00060 0.83850 0.03034 ± 0.06459
Pyrrhotite
Interstitial 0.00120 0.72000 0.10151 ± 0.16146 0.00120 0.28800 0.03719±0.05440
Included 0.00100 0.72000 0.01919 ± 0.05358 0.00100 0.09100 0.00825 ± 0.01867
Pentlandite
Interstitial 0.00174 0.9051 0.01472 ± 0.01457 0.00174 0.04177 0.00811 ± 0.00668
Included 0.02400 0.05280 0.04080 ± 0.01060 0.00480 0.00720 0.00660±0.00104

a Measurements of apparent maximum length and width made in thin section on 2,419 sulfide grains.
(Modified from Weiblen and Morey 1976).

monly contains included sulfide. Olivine grains are mantled with myrmekite-like
overgrowths of orthopyroxene and sulfide. The sizes of sulfides in voids varies
widely as a function of the size of the plagioclase framework (Table 1). Sulfide
modes vary greatly but averaged about 55070 pyrrhotite, 37% chalcopyrite-
cubanite, and 8% pentlandite. This average mode is much less copper-rich than
the other sulfide types and must indicate a unique sulfide composition. Poly-
phase assemblages and single phase assemblages of chalcopyrite or pyrrhotite are
COmmon; monomineralic occurrences of pentlandite are rare (Pasteris 1984).

Included Sulfides. Included sulfides are the second most important type and
occur mostly in plagioclase and clinopyroxene. Some grains also occur in
ilmenite, but no inclusions were found in olivine. Inclusions in clinopyroxene are
irregular in shape and are scattered throughout the grain, whereas those in
plagioclase have rectangular forms and are concentrated near grain margins. The
included sulfides average 0.061 mm in length and 0.028 mm in width, which is
smaller than the average size of the interstitial sulfides. A striking feature of the
included sulfides is that more than 99% of the grains are chalcopyrite. The chal-
copyrite is associated with bornite, pyrrhotite, or minor pentlandite.

Sulfide-Silicate Intergrowths. Both orthopyroxene and amphiboles are inter-

grown with sulfides. Sulfide-orthopyroxene intergrowths typically mantle olivine
and, in some cases, are separated from it by serpentine. Interstitial biotite and/or
ilmentite may also be present. The sulfides, primarily chalcopyrite, typically have
a myrmekitic texture.
The sulfide-amphibole intergrowths occur in interstitial voids and are charac-
terized by an interdigitated texture. The amphiboles are hornblende or actinolite;
the sulfides include pyrrhotite, chalcopyrite, cubanite, and pentlandite. Some
intergrowths occur either as an independent interstitial phase, or are closely asso-
ciated with interstitial clinopyroxene and sericititized plagioclase. In this latter
20 M. Foose and P. Weiblen

occurrence the textural evidence suggests the reaction clinopyroxene + plagio-

clase + H 20 -+ hornblende.

Fine-Grained Veinlets. Thin sulfide veins (less than 1 to 50 j..lm in width) cut all
the host silicate phases. Although relatively common, veins constitute only a
small fraction of the total sulfide population. The sulfides in the veins are pre-
dominantly chalcopyrite and cubanite.
Average sulfide compositions are given in Table 2. Pyrrhotite consists of both
monoclinic and hexagonal forms, contains an average of 0.17 wtOJo nickel and
less than 0.05 wtOJo cobalt. Pentlandite contains between 33 and 37 (average
34.67) wt% nickel and an average of 1.84 wt% cobalt. Pasteris (1984) examined
other pentlandites from this deposit and noted a somewhat larger range in nickel
contents (29.9 to 42.7 wt%) with the highest values occurring in the most copper-
rich assemblages.
Plagioclase and olivine are the two cumulus phases present in these rocks.
Olivine associated with sulfide ranges in compositions from F0 45 to F0 55 and
contains between 0.06 to 0.08 wt% nickel. Plagioclase has compositions near
An55' Plagioclase is typically overgrown by rims that contain sulfide inclusions.
The calcium-sodium ratio of plagioclase overgrowths that contain inclusions is
consistently much higher than cores of grains (for example, An55 vs An7o).
Similarly, the magnesium-iron ratios of sulfide-bearing pyroxene overgrowths
around olivine are larger than those of framework pyroxene grains that do not
include sulfides. The trends are reversed from those expected in a normal
differentiation sequence, but are consistent with the depression ofihe liquidus by
introduction of water.

Table 2. Average composition of sulfide phases from the Spruce Road deposita

Pyrrhotite Pentlandite Mackinawite Chalcopyrite Cubanite

(hexagonal) (monoclinic)
wt 070

S 36.67 39.41 33.37 36.06 32.90 36.94

Fe 62.65 60.02 30.55 54.95 31.29 39.46
Co 0.28 0.03 1.84 0.96 0.08
Ni 0.01 0.24 34.67 7.68 0.06 0.06
Cu 0.01 35.76 21.97
Total 99.61 99.71 100.43 99.65 100.09 98.43

Normalized atomic percentages

S 50.372 53.241 47.102 49.857 47.490 52.238

Fe 49.411 46.553 24.758 43.621 26.041 32.038
Co 0.209 0.022 1.413 0.722 0.063
Ni 0.007 0.177 26.727 5.800 0.047 0.046
Cu 0.007 26.157 15.678
MIS 0.985 0.878 1.123 1.005 1.097 0.914

a Wavelength dispersive electron microprobe analyses made at 20 kV, 0.03 IJ.A, 3 - 5 ~m beam diam-
eter, and 20 s counting interval.
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 21

The textural relationships described above are inconsistent with a simple

magmatic accumulation of sulfides. Such a process would require significant
amounts of gravity accumulation in order to produce the observed ore zone.
However, the disseminated, sporadically distributed, and interstitial nature of
this mineralization indicates that this accumulation has not happened. Instead,
the setting of most sulfides in intercumulus spaces or as inclusions in adcumulus
rims shows that sulfide separation occurred well after the accumulation of olivine
and plagioclase. Further, the association of sulfides with reversely zoned silicates
and hydrous phases (particularly biotite and amphiboles that texturally appear to
have formed by hydration reactions) indicates that sulfides formed when
volatiles were introduced into the crystallizing magma and that the volatiles were
sulfur-bearing. The textural relationships suggest that the volatiles were
introduced when the rocks were approximately 750/0 crystallized.
The relationships are consistent with isotopic studies in the adjacent PRT
(Ripley 1981, Rao and Ripley 1983). Sulfides in the PRT show sulfur isotopic
compositions between +0.2 to + 15.3 per mil, with averages near + 7.5 per mil.
These values differ significantly from the approximately 0.0 per mil value
expected from magmatic sulfides, but are similar to values found in the underly-
ing Virginia Formation. Additionally, 01SO values in troctolite near footwall
rocks have values of up to 9.6 per mil that are similar to 0 180 values of the
Viriginia Formation and are much greater than the 5.8 to 7.0 per mil in uncon-
taminated troctolite. The isotopic similarity between sulfides and country rock
clearly shows that assimilation of country rock introduced volatiles into the crys-
tallizing troctolite and that sulfur derived from the breakdown of sedimentary
pyrite was added to the magma and was then reprecipitated as copper-nickel sul-
fides.
Detailed isotopic studies have not been done within the SKI. However, posi-
tive values of + 7.5 and + 1.7 per mil have been determined from sulfides at two
different locations (Fig. 1; Ault and Kulp 1959) and are consistent with values
observed in the PRT. The footwall of the SKI is mostly granite, however, the
abundant metasedimentary inclusions within the sulfide-bearing basal zone in-
dicate that either metasediments capable of contributing sulfur were once present
or that metasediments have been transported from another area. Weiblen and
Morey (1980) note a consistent, but sharp decrease in biotite content away from
the footwall of the SKI, a change attributed to upward diffusion of water into a
crystallizing magma. The oxygen isotopic data of Churchill (1978) also provide
evidence for introduction of water as they indicate a correlation in anomalous
0 180 values in the troctolites with increased sulfide abundance. These observa-
tions support the textural data by indicating that introduction of a water-sulfur
rich vapor phase from metasedimentary rocks was an essential part of the genesis
of sulfides within the South Kawishiwi intrusion.
The sharp contrast between the relatively pyrrhotite-rich interstitial sulfides
and the much more copper-rich included and vein sulfides indicates two episodes
of mineralization. The polyphase mineralogy of the interstitial sulfides is con-
sistent with sulfide exsolution from typical magmatic Fe - Cu - Ni sulfide liq-
uids. The chalcopyrite dominated mineralogy of the included and vein sulfides
crosscut and replace these polyphase sulfides and are, thus, clearly later. The in-
22 M. Foose and P. Weiblen

troduction of the copper-rich solutions that deposited these later sulfides has
been attributed to a variety of processes including fractional crystallization of
sulfide liquids (Bonnichsen et al. 1980), fractional crystallization of the inter-
stitial silicate liquid (Pasteris 1984), and copper introduction from the adjacent
footwall rocks (Tyson and Chang 1984).

Conclusions

The disseminated and sporadically distributed sulfides in the basal zone of the
South Kawishiwi intrusion separated from a troctolitic liquid after cumulus
olivine and plagioclase had established a crystal framework. Accumulation of
sulfide was accompanied by the introduction of water and sulfur-rich volatiles
derived from metasedimentary xenoliths and footwall rocks. Reversely-zoned
rims on cumulus grains indicate that these volatiles were introduced when the
magma was approximately 750/0 crystallized.
The volatiles could have been introduced into the basal zone by a combina-
tion of three mechanisms. These are (1) direct vapor diffusion across the footwall
contact, (2) direct vapor diffusion from included metasedimentary rocks, or (3)
indirect transfer along faults and fractures. The decrease in biotite away from the
footwall contact is consistent with the first process, but the general upward
decrease in sulfide abundance which would also be expected has not been docu-
mented. Sulfides occur sporadically, but ubiquitously up to the sharp contact
with the sulfide-free zone. On the other hand, the apparent lack of concentration
of metasedimentary rocks along the footwall of this area and the evidence for
dynamic magmatic processes offered by the absence of any traceable silicate
layering in the sulfide-bearing zone suggests that conditions were appropriate for
the widespread mixing of sulfide-bearing inclusions in the basal zone. Faulting
that in part controlled the distribution and thickness of the basal zone also may
have provided channels along which volatiles could have concentrated.
The abrupt contact between the sulfide-bearing zone and the overlying
sulfide-free zone indicates a radical change in magmatic conditions. It is sug-
gested that the sulfide-bearing zone reflects dynamic magma mixing and assimi-
lation processes in a restricted magma chamber or a restricted part of a larger
chamber that was formed during rifting. Surges of magma into the chamber
entrained, assimilated, and dispersed pieces of the Virginia Formation and the
Biwabik Iron Formation and caused a complete disruption of any internal layer-
ing within the chamber. Crystallization of the heterogeneous mix was accom-
panied by devolatilization of the enclosed metasedimentary xenoliths; vapors
derived from the metasediments were important in forming the disseminated and
sporadically distributed sulfides. With further rifting, the restricted magmatic
conditions yielded to a more quiescent, open magmatic system in which the over-
lying sulfide-free troctolitic rocks accumulated. The homogeneous sequence of
troctolitic rocks that contain thin, "but laterally persistent anorthosite layers can
be attributed to nearly continuous replenishment of compositionally similar
liquids.
The Physical and Petrologic Setting and Textural and Compositional Characteristics of Sulfides 23

References

Ault WU, Kulp JL (1959) Isotopic geochemistry of sulphur. Geochim Cosmochim Acta 16:201- 235
Bonnichsen B (1970) Geologic map of the Allen, Babbitt, Babbitt NE., Babbitt S.E., and Babbitt
S.W.74 minute quadranges. Minn Geol Surv Open File Maps
Bonnichsen B (1972a) Southern part of Duluth Complex. In: Sims PK, Morey GB (eds) Geology of
Minnesota: A centennial volume. Minnesota Geol Surv, St. Paul, Minn, pp 361 - 387 (see Phinney
WC,1972)
Bonnichsen B (1972b) Sulfide minerals in the Duluth Complex. In: Sims PK, Morey GB (eds)
Geology of Minnesota: A centennial volume. Minnesota Geol Surv, St. Paul, Minn, pp 388 - 393
Bonnichsen B (1974) Copper and nickel resources in the Duluth Complex, northeastern Minnesota.
Minn Geol Surv Inf Circ 10:24
Bonnichsen B, Tyson RM (1975) Geology of the Ely-Hoyt Lakes region of the Duluth Complex, Min-
nesota. Geol Soc Am Field Trip Guide, Penrose Conference on Mafic Plutons and Magmatic
Sulfides. Ely, Minn, September 14-19, p 20
Bonnichsen B, Fukui L, Chang L (1980) Geologic setting, mineralogy, and geochemistry of magmatic
sulfides, South Kawishiwi intrusion, Duluth Complex, Minnesota. Proc 5th Q IAGOD Symp,
Stuttgart, Germany, pp 545 - 565
Boucher ML (1975) Copper-nickel mineralization in a drill core from the Duluth Complex of
northern Minnesota. US Bur Mines Rep Invest 8084:55
Chase CG, Gilmer TH (1973) Precambrian plate tectonics: The midcontinent gravity high. Earth
Planet Sci Lett 21 :70 -78
Churchill RK (1978) A geochemical and petrological investigation of the Cu - Ni sulfide genesis in the
Duluth Complex, Minnesota. MS Thesis, Univ Minn, Minneapolis, 101 pp (unpublished)
Cooper RW (1978) Lineament and structure analysis of the Duluth Complex, Hoyt Lake-Kawishiwi
area, northeastern Minnesota. Ph D Diss, Univ Minn, Minneapolis, 220 pp (unpublished)
Faure G, Chaudhuri S, Fenton MD (1969) Age of the Duluth Complex and the Endion Sill, Duluth,
Minnesota. J Geophys Res 74:720-725
Foose MP (1982) Structural, stratigraphic, and geochemical features of the South Kawishiwi
. Intrusion, Duluth Complex. Geol Soc Am (Abstract with Programs) 14:490
Foose MP (1984) Drill logs and correlations within part of the South Kawishiwi Intrusion, Duluth
Complex, Minnesota. US Geol Surv Open-File Rep 84-14:230
Foose MP, Cooper RW (1978) Geology of the Harris Lake area, northeastern Minnesota. US Geol
Surv Open-File Rep 78-385:34
Foose MP, Cooper RW (1981) Faulting and fracturing in part of the Duluth Complex, northeastern
Minnesota. Can J Earth Sci 18:810 - 814
Fukui LM (1976) The mineralogy and petrology of the South Kawishiwi intrusion, Duluth Complex,
Minnesota. MS Thesis, Univ Ill, Chicago Circle, 110 pp (unpublished)
Goldich SS, Nier AO, Baadsgaard H, Hoffman JH, Krueger HW (1961) The Precambrian geology
and geochronology of Minnesota. Minn Geol Surv Bull 41:193
Grant NK, MoIling PA (1981) A strontium isotope and trace element profile through the Partriolge
River troctolite, Duluth Complex, Minnesota. Contrib Mineral Petrol 77:296 - 305
Green JC, Phinney WC, Weiblen PW (1966) Gabbro Lake quadrangle, Lake County, Minnesota.
Minn Geol Surv Misc Map M-2
Grout FF (1918a) Internal structures of igneous rocks; their significance and origin; with special
reference to the Duluth Gabbro. J Geol 26:439 - 458
Grout FF (1918b) The lopolith; an igneous form exemplified by the Duluth Gabbro. Am J Sci 4th Ser
46:516-522
Iwasaki I, Malicsi AS, Lipp RJ (1983) By-product recovery from copper-nickel bearing Duluth
gabbro flotation tailings. Soc Min Eng AIME, paper presented at SME-AIME meeting. Atlanta
Georia, March 6-10, preprint No 83-20, p 11
King ER, Zietz I (1971) Aeromagnetic study of the Midcontinent gravity high of central United
States. Geol Soc Am Bull 82:2187 - 2208
Listerud WH, Meineke DG (1977) Mineral resources of a portion of the Duluth Complex and
adjacent rocks in St. Louis and Lake Counties, northeastern Minnesota. Minn Depart Nat Res Div
Miner Rep 93:74
Morey GB, Cooper RW (1977) Bedrock geology of the Hoyt Lakes-Kawishiwi area, St. Louis and
Lake Counties, northeastern Minnesota. Minn Geol Surv Open-File Map 1:48,000
24 M. Foose and P. Weiblen: The Physical and Petrologic Setting

Naldrett AJ (1981) Nickel sulfide deposits: classification, composition, and genesis. Econ Geol 75th
Anniversary Vol, pp 628-685
Pasteris JD (1984) Further interpretation of the Cu - Fe - Ni sulfide mineralization in the Duluth
Complex, northeastern Minnesota. Can Miner 22:39 - 53
Phinney WC (1969) The Duluth Complex in the Gabbro Lake quadrangle, Minnesota. Minn Geol
Surv Rep Invest 9:20
Phinney WC (1972) Northwestern part of Duluth Complex. In: Sims PK, Morey GB (eds) Geology of
Minnesota: A centennial volume. Minnesota Geol Surv, St. Paul,· Minn, pp 335 - 345 (see
reference by Bonnichsen, for example)
Rao BV, Ripley EM (1983) Petrochemical studies of the Dunka Road Cu - Ni deposit, Duluth Com-
plex, Minnesota. Econ Geol 78:1222 -1238
Ripley EM (1981) Sulfur isotopic studies of the Dunka Road Cu - Ni deposit, Duluth Complex, Min-
nesota. Econ Geol 76:610-620
Stopler E (1980) A phase diagram for mid-ocean ridge basalts: preliminary results and implications
for petrogenesis. Contrib Mineral Petrol 74:13 - 27
Taylor RB (1964) Geology of the Duluth Gabbro Complex near Duluth, Minnesota. Minn Geol Surv
Bul144:63
Tyson RM, Chang LL Y (1984) The petrology and sulfide mineralization of the Partridge River
troctolite, Duluth Complex, Minnesota. Can Miner 22:23 - 38
Watowich SN, Malcolm JB, Parker PO (1981) A review of the Duluth Gabbro Complex of
Minnesota as a domestic source of critical and strategic metals. Soc Min Eng AIME, paper
presented at the SME-AIME fall meeting. Denver, Col, Nov. 18 - 20, preprint 81-351, p 9
Weiblen PW (1965) A funnel-shaped, gabbro-troctolite intrusion in the Duluth Complex, Lake
County, Minnesota. Ph 0 Diss, Univ Minn, 161 pp (unpublished)
Weiblen PW, Morey GB (1975) The Duluth Complex - a petrologic and tectonic summary. Proc 48th
Annu Meet, Minn Sect Am Inst Min Eng 36th Annu Mining Symp, Univ Minn, Minneapolis,
pp72-95
Weiblen PW, Morey GB (1976) Textural and compositional characteristics of sulfide areas from the
basal contact zone of the, South Kawishiwi intrusion, Duluth Complex, northeastern Minnesota.
Proc 49th Annu Meet, Minn Sect Am Inst Min Eng 37th Annu Mining Symp, Univ Minn,
Minneapolis, p 24
Weiblen PW, Morey GB (1980) A summary of the stratigraphy, petrology, and structure of the
Duluth Complex. Am J Sci 280-A:88 -133
Wold RJ, Hinze WJ (eds) (1982) Geology and tectonics of the Lake Superior Basin. Geol Soc Am
Mem 156:280
Application of Stable Isotopic Studies to Problems of
Magmatic Sulfide Ore Genesis With Special Reference
to the Duluth Complex, Minnesota
E.M. RIPLEy1

Abstract

Analyses of the isotopic composition of S, C, 0, and H in mafic igneous rocks

may provide valuable information regarding the extent of contamination by con-
tinental crust material. Partial melting and devolatilization of country rocks are
important processes in the generation of many Cu - Ni deposits associated with
mafic igneous rocks. Sulfur isotopes provide a means of evaluating the pos-
sibility of extraneous sulfur addition to a melt. Variability and distribution of
o34g values may also provide data relating to the timing and mechanisms of
sulfur production and incorporation into the melt. Oxygen, hydrogen, and
carbon isotopes also can be sensitive indicators of isotopic contamination,
whether by partial melting, devolatilization, or solid state exchange. Zones of
contamination within an igneous sequence are likely to be areas where factors
that control sulfide solubility (T, fS2, /°2 , melt composition) have been per-
turbed, and are thus key sites for ore generation.
Stable isotopic studies of two deposits within the Duluth Complex have high-
lighted the importance of country rock contamination in the generation of
Cu - Ni ores. For the Dunka-Road deposit devolatilization of country rocks has
resulted in essentially in situ contamination, and the formation of ore that is
variable in both its spatial distribution and o34g values. Partial melting and
major element contamination are restricted to areas near the margins of
xenoliths. At the Babbitt deposit, contamination via both partial melting and de-
volatilization has been significant. Sulfur isotopic distribution between igneous
and metasedimentary rocks suggests that sulfur must have been derived prior to
or during magma ascent. Partial melting of country rocks and oxygen isotopic
exchange may have occurred either before or after magma emplacement.
Chemical diffusion, as well as fluid dynamic properties of mixing are thought to
control the isotopic inhomogeneity that characterizes sulfide ore zones in the
Duluth Complex.

1 Geology Department, Indiana University, Bloomington, IN 47405, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et aL)
© Springer-Verlag Berlin Heidelberg 1986
26 E.M. Ripley

Introduction

The chemical interaction between mafic magma and crustal country rocks has
received considerable attention in recent years. In particular, isotopic and trace
element data has been utilized in evaluating the importance of wall rock assimila-
tion and magma mixing in the generation of compositionally varied igneous
suites (Carter et al. 1978, Taylor et al. 1979, Gray et al. 1981, Thompson et al.
1982, Jacquemin et al. 1982). Modelling of trace element and isotopic systematics
produced by fractional crystallization and assimilation has been outlined by
Taylor (1980) and De Paolo (1981). Most of these studies have focused on large-
scale interaction between crust and primitive mafic magma, and its importance in
magma petrogenesis.

EXPLANATION
[IJ Cretaceous IUP£d I Upper Precambrian
Duluth Complex
IT] Paleozoic I MP£ I Middle Precambrian

[]ill Upper Precambrian [}ill Lower Precambrian

Fig. 1. Generalized bed-rock geology of Minnesota (After Cooper et al. 1981) showing location of the
Duluth Complex. The outlined area (Fig. 2) contains most of the Cu - Ni reserves of the Duluth
Complex
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 27

Contamination of mafic magma by sedimentary country rocks has also been

shown to be a potentially important process in the generation of magmatic
Cu - Ni and Pt-group element deposits. Mainwaring and Naldrett (1977) have
suggested that assimilation of sulfidic country rocks was instrumental in
producing ores associated with the Water Hen intrusion of the southern Duluth
Complex. Sulfur isotopic studies of sulfide occurrences in the Bushveld Complex
(Liebenberg 1970), Muskox intrusion (Sasaki 1969), Norils'k (Vinogradov and
Grinenko 1964), and Patridge River troctolite of the Duluth Complex (Ripley
1981) all suggest that most of the sulfur found in the intrusives is of sedimentary
derivation. Magma contamination by CO 2 - H 20 fluids has been suggested by de
Waal (1977) to explain sulfide mineralization in the base of the layered sequence
in the Bushveld Complex. Page (1979), basing his conclusions principally on
petrochemical and textural studies, suggests that partial melting of country rocks
has occurred in the vicinity of sulfide mineralization in the Basal Zone of the
Stillwater Complex. Bonnichsen (1972), Weiblen and Morey (1976), and Rao and
Ripley (1983) use petrographic and major element data to illustrate the associa-
tion between ore zones and areas of contamination in the Duluth Complex.

\ +WATER
; HEN
Fig. 2. Location of the Dunka Road and Babbitt
orth Shore deposits within the Patridge River troctolite of the
\blconic Group
Duluth Complex, Minnesota. (Modified from
Cooper el al. 1981)
28 E.M.Ripley

Groves et al. (1979) outline several models for the generation of Australian
Archaean Ni-sulfide deposits, and include sulfur derivation from metasedimen-
tary host rocks for many of the deposits. Addition of sedimentary sulfur and
carbon along contact zones has also resulted in the localization of sulfide
minerals in the Rana intrusion of Norway (Boyd and Mathiesen 1979). Recent
isotopic and chemical studies of the Sudbury ores (Rao et al. 1984) also highlight
the role of country rock contamination in the genesis of these important deposits.
The primary purpose of this paper is to review the principles involved in the
application of stable isotopic studies to problems of Cu - Ni ore genesis in mafic
igneous rocks. Particular emphasis is placed on the origin of sulfide occurrences
at two localities in the Duluth Complex (Figs. 1 and 2), where our research group
has been involved in stable isotopic and petrochemical studies over the past 6
years.

General Considerations

Evaluation of magma contamination by country rocks is normally based on the

premise that one of several possible mixing processes has occurred, and resulted
in the production of a melt with an isotopic, trace element, or major element
composition ranging between values found in the primary melt and the original
contaminant. The process may be a relatively simple one, with mixing of end-
members with constant compositions, or extremely complex, characterized by
mixing of end-members with variable compositions. A wide spectrum of inter-
mediate results is possible, depending on the degree of partial melting of the
country rock and the extent of fractional crystallization of the melt. Devolatiliza-
tion of country rocks and incorporation of volatiles in the melt may lead to
major and trace element, and isotopic systematics that are quite distinct from
those that result from a partial melting mechanism.
Haughton et al. (1974), Shima and Naldrett (1975), Buchanan and Nolan
(1979), and Danckwerth et al. (1979) have determined that sulfide solubility is
dependent on variables such as temperature, sulfur and oxygen fugacities, and
melt composition. Clearly all of these variables may be affected by magma
mixing or addition of volatiles to a melt. A decrease in magma temperature due
to heat transfer associated with contamination may lead to sulfide precipitation
or separation of an immiscible sulfide melt. In a dry system sulfur apparently dis-
solves in a mafic magma by displacing oxygen and bonding to metallic ions, prin-
cipally Fe2+ (Cu 2+, Ni 2+, etc. may also bond with sulfur, but their abundance is
far less than Fe 2+) (Fincham and Richardson 1954, Shima and Naldrett 1975).
Reactions such as
FeO + S2- = FeS + 0 2-
or
FeO + tS2 = FeS + t02
describe the process. Related to 10 2 conditions are iron redox state and overall
melt composition. Haughton et al. (1974) suggest that the Fe2+ content of mafic
magma is proportional to sulfide solubility. A decrease in Fe 2+ caused either by
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 29

oxidation (Fe 2 + --+ Fe 3 +) or magma mixing may lead to a decrease in the sulfide
capacity of the melt. Irvine (1977) has also suggested that an increase in Si0 2
content of a melt, accompanied by magma mixing, may lead to sulfide satura-
tion. Such a process is particularly important with respect to partial melting, as
Si0 2 constitutes a low melting component in the partial fusion of crustal
material. In addition, Si0 2 may be added to a melt in response to volatile produc-
tion in the country rocks (Walther and Orville 1982). In systems where water or
hydrogen is added to a mafic melt, the dissolution mechanism of sulfur is not
clear, and may involve reactions such as:
H 2S(v) + Ol,;) + Na(~) = HS(~) + ONa(~) + H(~)
H 2S(v) + FeO(m) = FeS(m) + H 20(m)
where (v) and (m) refer to vapor and melt, respectively. In such instances, sulfide
saturation is reached due to continued incorporation of volatiles from the
country rock.

Stable Isotopes as Indicators of Magma Contamination

Stable isotopic studies constitute one type of tool that can, and ideally should, be
applied to problems of melt contamination. However, for a comprehensive
evaluation such studies should also be combined with detailed petrographic,
trace element, bulk chemical, mineral chemical, and radiogenic isotope studies.
As an aid in evaluating the types of data that may be obtained from stable iso-
topic investigations, a brief review of the systematics of S, C, 0, and H isotopes
in magmatic systems is included below.

Sulfur. For sulfide ores perhaps the key element, and the one that is most com-
monly isotopically analyzed in evaluating the role of contamination, is sulfur.
Primary, or mantle-derived sulfur is normally considered to be characterized by
J34g values near the meteoritic value of ::::0%0. Slight fluctuations are possible
due to fractionation occurring during magmatic crystallization. Although sulfur
isotopic fractionation between crystals, silicate liquid, and sulfide liquid are
expected to be small at temperatures in excess of ::::900°C, particularly where
conditions are reducing enough that oxidized sulfur species are minimal,
evidence for high temperature fractionation in mafic systems does exist. For
example, trends in J34g values of sulfides found within layered intrusions have
been documented (Shima et al. 1963, Ripley 1983), and may be explained by
minor fractionation between Ni-, Cu-, and Fe-sulfide, accompanied by a
Rayleigh-type process of 34g enrichment during crystallization. Variations caused
by such mechanisms rarely exceed 4 to 5%0. In practice, J34g values from sulfides
in mafic rocks that fall outside a range of 0 ± 3%0 have generally been attributed
to contamination by sulfur derived from sedimentary country rocks. By simply
comparing observed J34g values to those that could likely be produced by pristine
magmatic processes, external addition of sedimentary sulfur (or other sulfur
component characterized by a J34g value far removed from 0%0) is relatively easy
30 E.M. Ripley

to document. Evaluation of kinetic isotope effects associated with sulfur libera-

tion in the wall rocks, mechanisms of sulfur transport, and degree of homo-
geneity within the melt, are impeded if 034s values and distribution within the
probable contaminant are not well known. Confirmation of external sulfur addi-
tion is significantly more complex if 034s values of sulfur in the contaminant are
not markedly different from 0 ± 3%0. Such may be the case for early Archaean
deposits, where sulfur in surrounding sedimentary rocks is commonly near 0%0.
Variation in the isotopic composition of sedimentary sulfur is linked to the devel-
opment of sulfate-reducing bacteria, whose origin is not easily documented prior
to about 2.6-2.7 b.y. (Ripley and Nicol 1981, Thode and Goodwin 1983).

Oxygen. The successful use of oxygen isotopes in evaluating magma contamina-

tion is dependent on the extent of contamination, the difference in 0 180 between
primary melt and contaminant, and the ratio of melt to contaminant, at any
stage of the process. In a simple two end-member mixing scheme, mass balance
expresssions of the form:
o180mix = o180parentXparent+ 0180contarn.Xcontam.
describe the resulting 0 180 value of the hybrid magma. In order to properly
evaluate the expression the atomic fraction of oxygen (X) contributed from each
end-member must be computed. Considering that 0 180 values of most mafic
igneous rocks range from 5.5 to 7.0%0 the expression illustrates that a small
degree of contamination may not be recognized unless the contaminant 0 180 is
far different than that of the melt. For instance, a 200/0 contribution of oxygen
from a country rock characterized by a 0 180 value of 10%0 to a mafic melt with a
0 180 of 6%0, produces a 0 180 mix of only 6.8%0. Careful and detailed isotopic
analyses are required to detect such variations.
Oxygen isotopic distribution within the contaminant must be known to
evaluate the extent of partial melting or devolatilization, and possible isotopic
exchange mechanisms. Analysis of individual minerals are also useful in inves-
tigating the timing of contaminant introduction, particularly in cases of small
scale or in situ assimilation, or when intrusion of a crystal mush is suspected.
Such a simple end-member mixing process is probably quite rare in geological
environments. A more realistic model may be the situation where one or both
end-members changes in composition due to oxygen addition or depletion, and
to isotopic fractionation. Taylor (1980) and De Paolo (1981) have described the
combined effects of fractional crystallization and assimilation on oxygen isotopic
systematics. Evaluation assumes a fixed 0 180 of the contaminant, changing pro-
portions of crystallized igneous rock to assimilated contaminant, and mayor
may not include fractionation between crystals and melt. In a simple situation
where one mineral is crystallizing from a melt and crystal-melt isotopic fractiona-
tion occurs, oxygen isotopic distributions may be computed by first determining
the isotopic effect of removing a specified amount of mineral on the remaining
liquid, followed by calculation of the resultant 0 180 value caused by assimilation
of a defined amount of country rock. A simple iterative routine allows calcula-
tions of this type to be done on most programmable calculators. Necessary ex-
pressions are similar to the following:
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 31

J melt = J(o) - Li(AIL)

J(o) = Initial isotopic composition of the primary melt (this value is
replaced by J me1t (see below) for successive iterations
Li = Fractionation factor between crystal and melt
A = Grams of mineral crystallized
L = Grams of remaining liquid
L(i) = L - 4 (in the case of 5 g of mineral crystallizing to 1 g of country rock
dissolving
X ml = (L(i)-1)IL(i); mass fraction of melt
Xcr = 1- X ml ; mass fraction of contaminant added to the melt
Jcr = J I80 of contaminant
J me1t = (Jme1t • X m1 ) + (Jcr ' X cr ) •
Figure 3 illustrates the specific case of assimilation of cpuntry rock with a J I80
value of 14%0 by a magma with an initial J I80 of 5.7%0, at a ratio of 5 g olivine
crystallized to 1 g country rock assimilated. Computation is only slightly more
complex for the fractionation of additional minerals. A family of such curves
may be generated [see Taylor (1980) and De Paolo (1981)], all with results quite
different from the simple mixing model.
The degree of homogeneity in a melt may also be investigated by detailed
oxygen isotopic studies. Diffusion plays an important role in the interaction
between mafic magma and silicic partial melts (Watson 1982), and may be the
limiting process in the case of in situ assimilation. Attainment of isotopic homo-
geneity is aided by turbulent mixing of liquids, but zones of isotopically normal
melt may persist within an otherwise anomalous melt depending on density,
velocity, viscosity, etc. which control the fluid dynamics of the system. Oxygen
isotopic homogeneity is discussed further below with reference to the Duluth
Complex.

100

80

..
c
.!2
0
N Fig. 3. Curve 1 shows the 0180melt path for a
60 specific case of assimilation, fractional crystal-
2
<II
lization of olivine, and fractionation between
>.
melt and crystals. Conditions are as follows:
..
'-
U
40 0180initial melt = 5.7%0
c
Q)
0
0180country rock = 14.0%0
'-
Q) 5 g olivine crystallized per 1 g country rock
a.. assimilated
20
Llmelt.olivine = 1.2%0
T= 1100°C
Curve 2 is a similar situation for a melt of
0
5 0 180(1) = 6.4%0, but without fractionation of a
specific mineral and no mineral-melt frac-
tionation
32 E.M. Ripley

Isotopic variation may also result if the J 180 value of the contaminant, as
well as that of the parent melt, progressively changes. For stable isotopes the
magnitude of such an effect may be small, because as temperature increases the
fractionation between species decreases. Therefore when partial melting of the
contaminant begins, the oxygen isotopic difference between produced melt and
residual crystals is expected to be small. Because of possible mineral-fluid frac-
tionations considerable variations in the isotopic composition of the contaminant
may result if devolatilization is the primary process responsible for mass transfer.
Kinetic effects could in large part control isotopic distributions in such cases.

Hydrogen. Hydrogen isotopes are potentially sensitive indicators of magma con-

tamination because of the small amount of hydrogen present in most basaltic to
ultramafic rock types. For isotopic investigation a primary requisite is a signif-
icant difference between the JD of the contaminant rocks or released water and
that of hydrogen in the primary melt. Often the mere presence of anomalous
amounts of hydrous phases in mafic host rocks indicates the introduction of
extraneous water. However, JD values of minerals, such as amphibole or biotite
(particularly if associated with sulfide ore minerals), may provide direct evidence
for melt contamination. If sufficient hydrogen is present in inclusions formed
within early crystallizing silicates, such as olivine, pyroxene, or plagioclase,
hydrogen isotopic composition of these phases relative to late crystallizing
hydrous phases, such as biotite or amphibole, may aid in reconstructing the
timing of introduction of country rock hydrogen.

Carbon. Analysis of J 13C in mafic rocks provides valuable information as to the

relative importance of igneous versus sedimentary-derived carbon in the ore-
forming process. Many deposits are thought to contain primary graphite asso-
ciated with sulfides, whereas in deposits, such as those of the Duluth Complex,
much of the graphite was derived from country rocks (Hollister 1980, Mainwar-
ing and Douthitt 1981). Primary igneous carbon is normally characterized by
J 13C values near - 5%0. Sedimentary carbon may be much lighter or heavier, de-
pending on its genesis as biogenic or carbonate carbon of marine or nonmarine
origin. Carbon may be incorporated in mafic magmas by partial melting, solid
ingestion, or devolatilization of graphite-bearing country rocks. As with the
other light stable isotopes, devolatilization provides an opportunity for isotopic
variation due to kinetic or equilibrium distributions among produced volatile
species (e.g., CH 4 , CO 2 ) and residual carbon. Because of its possible effect on
magma oxidation state, and hence on sulfide solubility, knowledge of the mecha-
nism and timing of carbon introduction is desirable in evaluating ore genesis.

Stable Isotopic Studies Applied to Cu - Ni Deposits of the Duluth Complex

Our research group has been involved in detailed mineralogic, chemical, and
isotopic investigations of two deposits located in the 1.1 b.y. - old Duluth
Complex of Minnesota. Both the Dunka Road and Babbitt deposits occur in the
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 33

southern portion of the Duluth Complex, within an intrusive sequence known as

the Patridge River troctolite (Fig. 2). Rock types present include troctolite, peri-
dotite, norite, olivine gabbro, and gabbro. The sequence constitutes part of what
is known as the Troctolitic series (Bonnichsen 1972) and is overlain by a thick
succession of predominantly anorthositic rock types (Anorthosite Series). Field
evidence suggests that the Troctolitic Series is intrusive into the Anorthositic
Series. Mineralogy of the Partridge River troctolite is varied (Bonnichsen 1972,
Grant and MoIling 1981, Rao and Ripley 1983, Tyson and Chang 1984), with
most rock types composed of euhedral to subhedral olivine, plagioclase, or
orthopyroxene, and interstitial phases which may include the same minerals,
along with clinopyroxene, ilmenite, biotite, and major sulfides. Country rocks in
the area include graphitic, and locally sulfidic pelitic metasediments of the
Virginia Formation that are underlain by the Biwabik Iron Formation and
granitic rocks of the Giant's Range Batholith.
Sulfides are generally localized near the basal contact with the Virginia For-
mation, but may occur several hundred meters above the lowermost contact. In
virtually all cases, however, xenoliths of country rock are found in the general
proximity of sulfide accumulations. It is possible that upper ore zones represent
previous basal contacts between metasedimentary and igneous rocks. Principal
ore minerals are pyrrhotite, chalcopyrite, cubanite, and minor pentlandite.
Copper: nickel ratios in this area of the Complex average ::::: 4 - 5 : 1.
Mineralization is dominantly disseminated in form, with sulfides associated with
interstitial silicates and oxides, primarily pyroxene, biotite, and ilmenite.
Detailed textural descriptions of sulfide occurrences in the Partridge River
troctolite are given by Rao and Ripley (1983), Pasteris (1984), and Tyson and
Chang (1984). Within most ore zones sulfide volume percentage averages :::::2- 3,
although massive occurrences of pyrrhotite or chalcopyrite-cubanite occur
locally. Massive ore is more abundant at the Babbitt deposit that at Dunka Road.

25

o Ore zone sulfide

20 ~ Non-ore sulfide

[!] Hornfelsed Virginia Formotion

>- 15
G Unmetamorphosed Virginia Fm.
(.)
c
<I>
::J
0-
<I>
It 10

Fig. 4. Histogram of 0 34S values

• for sulfides in troctolitic rocks of
the Dunka Road deposit, Virginia
20 25 Formation hornfels, and unmeta-
morphosed Virginia Formation
34 E.M. Ripley

Sulfur isotopic composition of minerals in the Dunka Road deposit are in-
structive with regard to sulfur origin and ore formation. The J 3"s values of all
sulfide minerals range from 0.2 to 15.3%0, with both mean and mode values near
7.5%0 (Ripley 1981; also see Fig. 4). No trends in sulfur isotopic composition
within troctolites are apparent, and values are in general very erratic. Pyrrhotite
in the underlying Virginia Formation and pyrite from unmetamorphosed
Virginia Formation are characterized by a smilar range in J 3"s values. The high
positive J 3"s values and similarity to values found in the underlying Virginia
Formation strongly suggest that most of the sulfur in the Dunka Road deposit
was derived from the breakdown of pyrite in the Virginia Formation. Sulfur
isotopic homogeneity was certainly not attained, and as such is consistent with a
relatively late introduction of external sulfur. Nickel contents of olivine (Rao and
Ripley 1983) do not indicate that olivine equilibrated with a coexisting sulfide-
rich melt. The disseminated form of mineralization, sulfur isotopic variability,
and nickel contents of olivine all suggest that sulfide saturation was achieved late
in the crystallization sequence of the Dunka intrusive. Such late introduction of
sulfur also would have inhibited the development of massive sulfide zones.
Oxygen isotopic results from Dunka (Fig. 5) shed light on the mechanisms
responsible for magma contamination. The J I80 values of troctolites and norites
range from 5.8 to 9.6%0. Anomalous JIO values (>8%0) are found only near the
margins of country rock xenoliths. The J I80 values of Virginia Formation horn-
fels in the Dunka Road area range from 8.8 to 11.1 %0. High J I80 values corre-
spond to the presence of orthopyroxene rather than olivine in the igneous rocks.
Bulk chemical studies verify the introduction of Si0 2 in these localities, and

Whole Rocks Plagioclase

Unit

40a

] .:::::~;.~

<:;
0 300
C
0
u
E
~ 200
'"
u
c:
0
Ui
0
100

VF
I
0
VF
5 7 9 11 5 7 9

Fig. 5. Representative whole rock and plagioclase 15 180 values for troctolites and Virginia Formation
inclusions, Dunka Road deposit
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 35

together with isotopic data are suggestive of partial melting along the rims of
xenoliths. However, bulk chemical and isotopic anomalies are restricted to
within = 3 m of inclusion contacts. The 0 180 values of mineral separates confirm
the whole rock systematics, with olivine, pyroxene, plagioclase, ilmenite, and
biotite values paralleling those of whole rocks.
Sulfur and oxygen isotopic data, together with mineralogical and chemical
analyses, strongly suggest that assimilation of nonvolatile elements via partial
melting was not a strong control on sulfide genesis at Dunka Road. Isotopic
systematics are more consistent with major contamination caused by devolat-
ilization of country rocks. Reactions such as:
FeS2+3H20+5I2C = FeS+3/2C0 2 +CH 4 +H 2S
are thought to characterize volatile production in the contact rocks and
xenoliths. Experiments done in our lab (Ripley and Snyder, in preparation)
indicate that under hydrous conditions, little sulfur isotopic fractionation accom-
panies the production of H 2S from pyrite. Therefore an in situ model of volatile
transfer and melt contamination may best explain the disseminated nature of
sulfide mineralization and variable 03~ values. Sulfur is thought to be derived
from local accumulations of pyrite in the Virginia Formation, which are them-
selves characterized by variable 03~ values. Because of the late introduction of
H 2S into the crystal mush, sulfur mobility and isotopic homogenization were
impeded.
Several other mechanisms that may have contributed to sulfur isotopic vari-
ability at Dunka Road are discussed by Ripley (1981). Included are mixing of
sedimentary and igneous sulfur, isotopic fractionation caused by fluctuations in
oxidation states during sulfur volatilization and assimilation, and periodic intro-
duction of isotopically distinct magma batches. Although none of these processes
can be entirely eliminated, isotopic, chemical, and mineralogical data are more
consistent with the devolatilization model described above.
The Babbitt deposit, located to the northeast of the Dunka Road area (Fig. 2)
shares many common characteristics, although isotopic systematics suggest a
somewhat different history of ore genesis. Ore at Babbitt is generally dis-
seminated in form, as at Dunka, but massive, high-grade areas occur in greater
abundance, and render the deposit economically more attractive. Concentrations
of platinum group elements at Babbitt far exceed those at Dunka. Areas of mas-
sive to semi-massive (> 100/0 sulfides) ore are locally found in both norites and
cordierite-orthopyroxene hornfels. Contacts between igneous and metasedimen-
tary rocks in such areas are difficult to discern, and norite is far more abundant
than troctolite, suggesting appreciable introduction of silica. Based on these
criteria alone, the degree of probable contamination at Babbitt would appear to
be greater than at Dunka.
Oxygen isotopic values of whole rocks and mineral separates from Babbitt
confirm mineralogical and chemical evidence for contamination. Igneous rock
0 180 values range from 5.4 to 12.1%0, with a mode at 7 -7.5%0 (Ripley and AI-
Jassar 1983, and in preparation). Metamorphosed Virginia Formation 0 180
values range from 8.0 to 14.5%0, as do representative samples from mildly meta-
morphosed Virginia Formation graywacke, slate, and argillite. The 0 180 values
36 E.M. Ripley

COPPER DISTRIBUTION - W.R. 6180

NW 189 156 SE

PWftMI +.60% Cu
8.1
7.8 ~> 1.00%Cu
mrrrm
6.0
Virginia Fm.
Ft M
300LO
6.4 ~Biwabik Iron Fm.

90M
o
o 300Ft
8.1
5.5

Fig. 6. Representative whole rock "ISO values for igneous rocks and Virginia Formation hornfels
along a drill section, Babbitt deposit

Table 1. Oxygen isotope values through two igneous rock-metasedimentary

xenolith profiles, Babbitt deposit

Sample depth (m) "ISO (%oSMOW) Rock type

Locality 1
279.8 7.5 Troctolite
280.0 9.1 Troctolite
281.3 11.6 Hornfels
282.8 14.1 Hornfels
284.3 8.8 Hornfels
284.7 8.4 Hornfels
285 .9 5.8 Troctolite
Locality 2
516.0 6.7 Troctolite
517.5 6.5 Troctolite
519.7 11.7 Hornfels
520.9 10.4 Hornfels
523.6 11.1 Hornfels
525.1 12.5 Hornfels
526.7 to.7 Hornfels
528.2 11.0 Hornfels
529.7 10.9 Hornfels
531.2 12.1 Hornfels
532.9 10.9 Hornfels
534.3 8.3 Troctolite
535.8 8.8 Troctolite
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 37

show a sporadic distribution within ore zones (Fig. 6). Trends are difficult to
distinguish, in part due to the complex mixture of rock types present and in part
due to sampling dictated by drill core location. Detailed oxygen isotopic studies
around Virginia Formation inclusions show J 180 variations in both the metasedi-
ments and surrounding igneous rocks. Table 1 illustrates the results of two
profiles through xenoliths and surrounding igneous rock. Although inclusion
hornfels are mineralogically very similar, their J 180 values may vary by as much
as 5%0 over a 2-m interval. Some of this isotopic variability may be attributed to
differences in the sedimentary prototypes (graywacke, shale, calcareous units),
and some must be due to interaction with the mafic magma. The J 180 values of
igneous rocks in contact with the inclusions decrease away from the contact. This
trend is similar to that observed at Dunka Road, but the width of isotopically
anomalous zones is generally greater. Within sulfide ore zones over 80070 of the
igneous rocks analyzed are considered isotopically anomalous. Areas of iso-
topically normal (5 .5 - 7.0%0) igneous rocks occur sporadically within areas
having anomalous values.
Locations of isotopically normal troctolites and norites within extensive
anomalous zones are difficult to evaluate. Two possibilities seem plausible. One
is that assimilation occurred essentially in place, with mass transfer limited by
diffusion within the melt. In this manner isotopic "exchange" zones may be pro-
duced around xenoliths, some in overlapping fashion. Figure 7 illustrates a case
where areas of igneous rock characterized by anomalous J 180 values surround
isotopically normal troctolite.
A second alternative is that contamination occurred during and after intru-
sion, and zones of turbulent mixing were limited by the presence of xenoliths.
The fluid dynamics of partial melting of country rock around an intrusive have
been modelled by Kovach and Marsh (1981). Their results suggest that extensive
partial melting may occur only if a critical velocity is maintained by the magma.
A mafic magma intruding a metasedimentary sequence may remove blocks of
wall rock during ascent, but thorough mixing of the partial and primary melts is
dependent upon flow patterns as well as diffusion. Local instabilities may be
generated along the interface between the two melts (e.g., Huppert and Sparks
1980, Irvine 1980), with the entrainment of contaminants restricted to well-

r;:;t II II
~ Normal magma Formation

Fig. 7. Oxygen isotopic "exchange" zones developed around the margins of xenoliths, with inter-
vening areas of isotopically " normal" igneous rock
38 E.M. Ripley

~ Virg inia Fm.

~ inclusions

c=J Maf ic magma

!:.{:/':I ~~et~p~~ 1~~~~eadnge

... and mixing

Fig. 8. Hypothetical flow and mixing patterns around xenoliths in a magma of fixed velocity. Asmall
area of essentially unaffected melt occurs between the xenoliths

defined localities. Figure 8 illustrates a situation where flow and mixing patterns
around metasedimentary xenoliths produce a small intervening area of melt only
moderately affected by the partial melting process. The oxygen isotopic composi-
tion obtained in the zo~e of mixing is dependent on the 0 180 values of xenoliths,
diffusion coefficients, fluid dynamic properties of the melts, and time of the
mixing process. Variations in any of these parameters could lead to the inhomo-
geneity observed in isotopically "anomalous" zones.
Sulfur isotopic studies at Babbitt also reveal a pattern of sulfur evolution that
differs from that at Dunka. The range and mean of sulfur isotopic values is
approximately the same as at Dunka (Fig. 9), and leave little doubt that most of
the sulfur in' the deposit is characterized by a strong sedimentary component.
However, two distinctions are noteworthy. First is the difference between 0 3"5
values of sulfides in the igneous rocks versus those of pyrrhotite in the Virginia
Formation hornfels. Unlike Dunka, where isotopic values were similar to those
in troctolites, hornfels from the Babbitt area have 0 3"5 values that range from
1.5 to 6.2%0, and are lower than over 95070 of the values for sulfides in ore zones.
A simple explanation of in situ devolatilization or partial melting does not ade-
quately explain the 0 3"5 distribution. Although the 0 3"5 values of pyrrhotite
from the hornfels are low, they do not fall outside the range for pyrite in weakly
metamorphosed Virginia Formation. It is apparent that sulfur has been lost in
the conversion to pyrrhotite, but evidence is lacking for the introduction of suf-
ficient sulfur from this source to cause a decrease in ore zone 0 3"5 values.
Two possible mechanisms are considered for the derivation of sulfur at
Babbitt. First is the rather unlikely possibility that partial melting of Virginia
Formation has removed all traces of sulfides having 0 3"5 values in excess of
:::::6%0. The fact that 0 3"5 values of Virginia Formation pyrrhotite from several
localities in the Babbitt area show a similar range of values argues against such a
hypothesis. A second mechanism involves the derivation of sulfur prior to or
during magma ascent. Such a process is similar to that proposed for magma con-
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 39

28 fig. 9. Histogram of tS 34S values for

• >1.00 "1. Cu sulfides in igneous rocks and Virginia
24 D 0.20 - 1.00 "1. Cu Formation hornfels from the Babbitt
[!] < 0.20 "1. Cu deposit
(v. F. hornfels)
III 20
G)
a.
E

-
~ 16
o
Iii
.0
12
E
:::l
Z 8

O~,r~~~~~~~~~~~-r~-+
o 2 4 12 14

tamination based on ~180 studies. The ~3~ values at Babbitt are consistent with
a model of major sulfur derivation from rocks not now represented in the area of
the Babbitt deposit. Assimilation of crustal sulfur during ascent, with limited
sulfur communication in the melt, could produce the observed ~3~ systematics.
In addition, a model of sulfur derivation prior to or during magma ascent
provides a means for immiscible sulfide liquids to be brought into contact with a
greater amount of silicate liquid. Campbell et al. (1983) have presented a model
for the generation of Pt-rich sulfide deposits that emphasizes the importance of
large silicate/sulfide melt ratios. Analyses of Pt-group elements in the Babbitt
and Dunka Road deposits indicate a greater than four times enrichment in the
Babbitt deposit versus Dunka Road. Extraction of metal from a larger mass of
magma is one mechanism which may account for the compositional differences.
A second unique feature of the ~34s values found at Babbitt is the distribu-
tion of isotopes between low grade (Cu-poor) and high grade (> 1% Cu) ore
(Figs. 9 and 10). Although ~3~ values show an overall variability, there is a
distinct isotopic fractionation consistent with ore grade. Copper-rich zones are
characterized by ~3~ values in excess of 9%0, whereas lower grade sulfide occur-
rences show ~3~ values between 6 and 10%0. This isotopic zonation strongly
suggests that significant sulfur isotopic fractionation occurred between a
separated Cu-enriched sulfide liquid and a Cu-poor residual melt (including any
sulfur still dissolved in the silicate melt). A similar 3~ enrichment in copper-rich
sulfides from the Levack West deposit of the Sudbury areas has been described
by Naldrett (1981).
Both sulfur and oxygen isotopic studies suggest that significant contamina-
tion via both partial melting and devolatilization, has occurred at Babbitt. Sulfur
isotopic values are difficult to explain in terms of in situ sulfur derivation. How-
ever, both sulfur and oxygen isotopic compositions are consistent with a model
of major contamination prior to or during emplacement, with isotopic homoge-
40 E.M. Ripley

/ .... , ....
D 34
+0.60% Cu, 6 S = 6.5-10.0%. /'~---
I .,~--
"',

D- >1.00% Cu, 6 34 S >10.0%. ('.

\
':)
l
Grade contact ----"}
/
I
~ Troctolite/hornfels /
,....... -_-- -
/
f
Somple sile, 634 S
.... ....
,
\
\
1
o
,,600 FT I
I
I
oI " 150 M
I
I
/
/
I

Fig, 10. Copper grade and 0 34S distribution, 1700-ft level, Babbitt deposit

neity prohibited by fluid dynamic properties of the melts. Critical parameters are
the size and distribution of crystals and xenoliths, diffusion rates, temperature,
density and viscosity of the primary melt and contaminant partial melt, and
magma velocity.

Acknowledgments. Appreciation is expressed to the geological staffs of United States Steel Corpora-
tion (Dunka Road deposit) and AMAX and Kennecott Company (Babbitt deposit) for cooperation
during all phases of our studies in the Duluth Complex. Many of the ideas presented in this paper
have benefited by discussions with A. J. Naldrett, B. V. Rao, R. Bonnichsen, P. Weiblen, M. Foose,
M. Tyson, N. Grant, T. Al-Jassar, J.D. Pasteris, and D. Cole. Research on the genesis of Cu- Ni
ores in the Duluth Complex has been supported through National Science Foundation grants EAR
78-01699 and EAR 81-08536.

References

Bonnichsen BG (1972) Southern part of the Duluth Complex. In: Sims PK, Morey GB (eds) Geology
of Minnesota: A centennial volume. Minn Geol Sur, St. Paul, pp 361- 388
Boyd R, Mathiesen CO (1979) The nickel mineralization of the Rana mafic intrusion, Nordland,
Norway. Can Miner 17, 2:287-299
Buchanan DL, Nolan J (1979) Solubility of sulfur and sulfide immiscibility in synthetic tholeiitic
melts and their relevance to Bushveld-complex rocks. Can Miner 17, 2:483 - 495
Campbell IH, Naldrett AJ, Barnes SJ (1983) A model for the origin of the platinum-rich sulfide
horizons in the Bushveld and Stillwater Complexes. J Petrol 24: 133 - 165
Carter SR, Evensen NM, Hamilton PT, O'Nions RK (1978) Nd- and Sr-isotopic evidence for crustal
contamination of continental volcanics. Science 202:743 -747
Cooper RW, Morey GB, Weiblen PW (1981) Topographic and aeromagnetic lineaments and their
relationship to bedrock geology in glaciated Precambrian terrane, northeastern Minnesota. In:
O'Leary 0 (ed) Proc 3rd Int Conf Basement Tectonics. Basement Tecton Commun Inc 3:137 -148
Danckwerth PA, Hess PC, Rutherford MJ (1979) The solubility of sulfur in high-Ti02 mare basalts.
Proc Lunar Planet Sci Conf 10:517 - 530
Application of Stable Isotopic Studies to Problems of Magmatic Sulfide Ore Genesis 41

DePaolo DJ (1981) Trace element and isotopic effects of combined wallrock assimilation and frac-
tional crystallization. Earth Planet Sci Lett 53:189-202
DeWaal SA (1977) Carbon dioxide and water from metamorphic reactions as agents for sulfide and
spinel precipitation in mafic magmas. Geol Soc S Afr Trans 80:193 -197
Fincham CJB, Richardson FD (1954) The behavior of sulfur in silicate and aluminate melts. Proc R
Soc London Ser A 223:40-62
Grant NK, Molling PA (1981) A strontium isotope and trace element profile through the Partridge
River troctolite, Duluth Complex, Minnesota. Contrib Mineral Petrol 77:296 - 305
Gray CM, Cliff RA, Goode ADT (1981) Neodymium-strontium isotope evidence for extreme con-
tamination in a layered basic intrusion. Earth Planet Sci Lett 56:189-198
Groves 01, Barrett RM, McQueen KG (1979) The relative roles of magmatic segregation, volcanic
exhalation, and regional metamorphism in the generation of volcanic-associated nickel ores of
Western Australia. Can Miner 17, 2:319-337
Haughton D, Roder PL, Skinner BJ (1974) Solubility of sulfur in mafic magmas. Econ Geol
69:451-567
Hollister VF (1980) Origin of graphite in the Duluth Complex. Econ Geol 75:765 -766
Huppert HE, Sparks RSJ (1980) The fluid dynamics of a basic magma chamber replenished by hot,
dense ultrabasic magma. Contrib Mineral Petrol 75:279 - 289
Irvine TN (1977) Origin of chromite layer in the Muskox Intrusions and other stratiform intrusions: a
new interpretation. Geology 5:273 - 277
Irvine TN (1980) Magmatic infiltration metasomatism, double diffusive fractional crystallization,
and adcumulus growth in the Muskox intrusion, and other layered intrusions. In: Hargraves RB
(ed) Physics of magmatic processes. Princeton Univ Press, Princeton, pp 325 - 385
Jacquemin H, Sheppard SMF, Vidal P (1982) Isotopic geochemistry (0, Sr, Pb) of the Golda Zuelva
and Mboutou anorogenic complexes, North Caneroun: mantle origin with evidence for crustal
contamination. Earth Planet Sci Lett 61 :97 -111
Kovach LA, Marsh BD (1981) Magma flow rate and partial fusion of wall rock; Hungtington Lake,
CA. Geol Soc Am (Abstracts with Program) 13:490
Liebenberg L (1970) The sulphides in the layered sequence of the Bushveld Igneous Complex. Geol
Soc S Afr, Spec Publl:l08-207
Mainwaring PR, Douthitt CB (1981) Carbon isotope study of graphitic sulfides in the Duluth
Complex, Minnesota, USA (Abstracts). Geol Assoc Can 6:A-37
Mainwaring PR, Naldrett AJ (1977) Country rock assimilation and the genesis of Cu - Ni sulfides in
the Water Hen intrusion, Duluth Complex, Minnesota. Econ Geol 72:1269-1284
Myers JD, Sinha AK, Marsh BD (1983) Assimilation of crustal material by basaltic magma:
Strontium isotopic and trace element data from the Edgecumbe Volcanic Field, S.E. Alaska. J
Petrol 25, 1:1- 26
Naldrett AJ (1981) Nickel sulfide deposits: classification, composition, and genesis. Econ Geol 75th
Anniversary Vol 2:628-685
Page NJ (1979) Stillwater Complex, Montana - structure, mineralogy, and petrology of the Basal
zone with emphasis on the occurrence of sulfides. US Geol Surv Prof Pap 1038:1- 69
Pasteris JD (1984) Further interpretation of the Cu-Fe-Ni sulfide mineralization in the Duluth
Complex, northeastern Minnesota. Can Miner 22:39 - 53
Rao BV, Ripley EM (1983) Petrochemical studies of the Dunka Road Cu-Ni deposit. Econ Geol
78:1222 -1238
Rao BV, Naldrett AJ, Evensen NM (1984) Crustal contamination in the sublayer, Sudbury igneous
complex: A combined trace element and strontium isotope study. Geol Assoc Can (Abstrat with
Program) 9:98
Ripley EM (1981) Sulfur isotopic studies of the Dunka Road Cu-Ni Deposit, Duluth Complex,
Minnesota. Econ Geol 76:610-620
Ripley EM (1983) Sulfide mineralogy and sulfur isotope geochemistry of layered sills in the Deer Lake
Complex, Minnesota. Mineral Depos 18:3 -15
Ripley EM, AI-Jassar T (1983) Oxygen isotopic studies of the Babbitt Cu - Ni prospect, Duluth
Complex, Minnesota (Abstracts). Geol Soc Am (Abstract with Program) 15:671
Ripley EM, Nicol DL (1981) Sulfur isotopic studies of Archean slate and graywacke from Northern
Minnesota: Evidence for the existence of sulfate-reducing bacteria. Geochim Cosmochim Acta
45:839-847
42 E. M. Ripley: Application of Stable Isotopic Studies

Sasaki A (1969) Sulfur isotope study of the Muskox intrusion, District of MacKenzie. Geol Surv Can
Pap 68-46:68 pp
Shima H, Naldrett AJ (1975) Solubility of sulfur in an ultramafic melt and the relevance of the system
Fe-S-O. Econ Geol60:960-967
Shima M, Gross WH, Thode HG (1963) Sulfur isotope abundances in basic sills, differentiated
granites, and meteorites. J Geophys Res 68:2835 - 2847
Taylor HP Jr (1980) The effects of assimilation of country rocks by magmas on 180/ 160 and
87Sr/86Sr systematics in igneous rocks. Earth Planet Sci Lett 47:243 - 254
Taylor HP Jr, Giannetti B, Turi B (1979) Oxygen isotope geochemistry of the potassic igneous rocks
from the Roccamonfina volcano, Roman comagmatic region, Italy. Earth Planet Sci Lett
46:81-106
Thode HG, Goodwin AM (1983) Further sulfur and carbon isotope studies of late Archaean Iron-
Formations of the Canadian Shield and the rise of sulfate-reducing bacteria. Precambrian Res
20:337-356
Thompson RN, Dickin AP, Gibson IL, Morrison MA (1982) Elemental fingerprints of isotopic con-
tamination of Hebrideau Paleocene mantle-derived magmas by Archaean sial. Contrib Mineral
Petrol 79: 159 - 168
Tyson RM, Chang LL Y (1984) The petrology and sulfide mineralization of the Patridge River
troctolite, Duluth Complex, Minnesota. Can Miner 22:23 - 38
Vinogradov AP, Grinenko LN (1964) On the influence of including rocks on the isotopic composition
of ore sulfides. Geokhimiya 9:491 - 499
Walther JV, Orville PM (1982) Volatile production and transport in regional metamorphism. Contrib
Mineral Petrol 79:252 - 257
Watson EB (1982) Basalt contamination by continental crust: some experiments and models. Contrib
Mineral Petrol 80:73 - 87
Weiblen PW, Morey GB (1976) Textural and compositional characteristics of sulfide ore from the
basal contact zone of the South Kawishiwi intrusion, Duluth Complex, northeastern Minnesota.
Proc 49th Annu Minn Sect AIME, 24 pp
Controls on the Formation of Komatiite-Associated
Nickel-Copper Sulfide Deposits
c. M. LESHER 1 and D. I. GROVES 2

Abstract

Nickel-copper sulfide deposits associated with rocks of komatiitic affinity may

be divided into two associations on the basis of host rock composition and ore
distribution: (1) stratiform, massive-matrix-disseminated sulfides hosted by
cumulate komatiite lava flows (lava conduits), or strata-bound, coarse dis-
seminated sulfides hosted by cumulate komatiite bodies near volcanic vents, and
(2) strata-bound, finely disseminated sulfides and more restricted massive sul-
fides hosted by subvolcanic komatiitic dunites (high-level magma chambers).
These deposits represent a continuum in environments of emplacement from
distal volcanic through proximal and central volcanic to subvolcanic.
Field relationships and geochemical data suggest that the sulfides in these de-
posits are magmatic in origin. Previous models for their formation involve
derivation of sulfides directly from the mantle or from low-level magma cham-
bers, but there are several field, physical, and thermodynamic constraints which
suggest that this may not be feasible: (1) the spatial concentration of mineraliza-
tion within certain parts of otherwise virtually barren host units suggests that the
latter were not uniformly saturated in sulfide at the time of emplacement; (2)
there are physical problems in transporting dense sulfides in low viscosity
komatiite magmas, although dispersed sulfide droplets may be supported during
rapid ascent; and (3) high temperature komatiitic melts generated by high-degree
partial melting, or sequential melting, of mantle source rocks are unlikely to be
sulfide-saturated in the source region or to remain sulfide-saturated during
adiabatic ascent. These deposits are, however, commonly associated with specific
country rocks, particularly sulfidic sediments, and the distribution of the ores
appears to be related to the distribution of these rocks. It is most likely that the
sulfide ores exsolved from the komatiitic magmas in situ, in response to assimila-
tion of the country rocks during emplacement. The distribution and texture of
the mineralization was probably influenced by (1) the mode of emplacement
(volcanic vs subvolcanic) and amount of assimilation; (2) the relative volume of
magma (lava conduit vs magma chamber) and its capacity to dissolve sulfur; and
(3) the timing of sulfide separation and opportunity for segregation.

1 Department of Geology, University of Alabama, University, AL 35486, USA

2 Department of Geology, University of Western Australia, Nedlands 6009, Western Australia

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
44 C. M. Lesher and 0.1. Groves

A fundamental aspect of this model is the coincidence of cumulate komatiitic

host units (representing dynamic lava conduits and magma chambers) and
siliceous, sulfidic, or ferric iron oxide-rich sediments, the former representing a
source of heat for assimilation and of ore metals, and the latter a source of sulfur
or of components to induce sulfide separation. Mineralized greenstones appear
to have formed in deeper water under conditions of greater extension than un-
mineralized greenstones; rifting environments represent the most favorable
tectonic environment for voluminous komatiite eruption/irruption and for ac-
cumulation of sulfidic sediments. Variations in the degree of extension along or
within the rift may influence the mode of komatiite emplacement and therefore
the type of deposit.

Introduction

Nickel-copper sulfide deposits associated with komatiitic rocks in Archean green-

stone belts currently account for over 25070 of the world's identified resources of
sulfide Ni with greater than 0.8% Ni (Ross and Travis 1981). Field relationships
and geochemical data suggest that the sulfides in these deposits are magmatic in
origin (Groves et al. 1979, Naldrett 1979). However, several fundamental aspects
of ore genesis remain poorly constrained and many details of ore localization are
still debated (cf. Lesher et al. 1984, Gresham, this volume).
The most important issue is the timing of separation of the sulfides in these
deposits. Most models for the formation of the Ni - Cu ores have stressed their
restrictive association with highly magnesian host rocks and have inferred that
sulfides were an intrinsic component of the precursor magma (e.g., Nesbitt 1971,
Naldrett 1973, Arndt 1976a). For example, Naldrett (1973) suggested that mantle
source regions were enriched in sulfides through downward percolation in areas
away from major convective overturn. Nesbitt (1971) and Naldrett and Cabri
(1976) noted that some deposits contain concentrations of sulfides that are too
great to have been dissolved in the overlying host units, and proposed that the
sulfides were carried in suspension directly from the mantle. Under the assump-
tion that the olivine in the host units is intratelluric, Ross and Hopkins (1975) and
Marston et al. (1981) suggested that immiscible sulfides and olivine crystals
accumulated (in a subsurface magma chamber) prior to eruption, and that
sulfides were erupted as a segregated phase.
There are several problems with such models. Firstly, the tectonic setting and
petrogenesis of komatiites in the Archean have not been established with any
degree of certainty. If komatiites in greenstone belts were formed in settings
similar to marginal basins (probably the best modern analog: Weaver and Tarney
1979), were generated by fusion of subducted komatiitic ocean crust (e.g.,
Allegre 1982, Arndt 1983), or were derived from a laterally extensive mantle
melt-layer (Nisbet and Walker 1982), then sulfides are unlikely to have become
concentrated in the manner suggested by Naldrett (1973). Secondly, iron-nickel
sulfide liquids are very dense (4.2 - 4.3 g cm -3: Dobrovinski et al. 1969) and
should settle to the base of a slowly ascending diapir and be lost in precisely the
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 45

same manner that they were originally proposed to have concentrated in the
source region. Thirdly, the present distribution of sulfides in the ore zones is not
necessarily representative of their abundance in the original magma, because the
host units are laterally more extensive than the ores (Lesher et al. 1984). Mass
balance calculations, therefore, provide only minimum values of magma: sulfide
ratios. Finally, dense sulfide liquids that have segregated at the base of a magma
chamber (or diapir) are unlikely to be erupted through the less dense
lithosphere/ crust.
Most models for the formation of these deposits are based on specific
features of a few individual deposits and have not considered the more funda-
mental features of the deposits as a group. There are also several important field,
physical, and thermodynamic constraints on the generation and transport of
sulfides in komatiitic magmas that require evaluation before models involving
eruption of mantle-derived sulfides can be accepted. The aims of this paper are
(1) to summarize the fundamental characteristics of komatiite-associated nickel-
copper deposits as a group; (2) to discuss some of the field, physical, and thermo-
dynamic constraints on the generation and eruption of sulfide-bearing komati-
ites; and (3) to propose an alternative, general model for their formation within
that context.

Fundamental Features

Komatiite-associated nickel-copper sulfide deposits may be divided into two

associations on the basis of host rock composition and ore distribution:
1. Cumulate komatiite-associated deposits:
a) stratiform, massive-matrix-disseminated sulfides hosted by cumulate komati-
ite lava flows (e.g., Kambalda: Gresham and Loftus-Hills 1981; Wannaway:
McQueen 1981b; Windarra: Schmulian 1984; Scotia: Page and Schmulian
1981; Langmuir: Green and Naldrett 1981; Trojan: Chimimba 1984),
b) strata-bound, coarse-disseminated sulfides hosted by cumulate komatiite
bodies near volcanic vents (e.g., Shangani: Viljoen et al. 1976; Damba-
Silwane: Williams 1979), and
2. Komatiitic dunite-associated deposits:
strata-bound, finely disseminated sulfides and more restricted massive sulfide
lenses hosted by subvolcanic komatiitic dunites (e.g., Mt. Keith-Betheno:
Groves and Keays 1979; Six Mile Well: Naldrett and Turner 1977; Honey-
moon Well: Donaldson and Bromley 1982; Agnew: Billington 1984; For-
restania: Porter and McKay 1981; Dumont: Eckstrand 1975; Katiniq: Barnes
et al. 1981).
The first association has been referred to previously as the "volcanic peridotite"
association, the second as the "intrusive dunite" association (Marston et al.
1981). The host rocks are here recognized as members of the komatiitic suite (see
Arndt and Brooks 1980, Arndt and Nisbet 1982), and are named accordingly.
46 C. M. Lesher and D.l. Groves

Table 1. Comparison of the cumulate komatiite and komatiitic dunite hosts to nickel sulfide mineral-
ization in the Norseman-Wiluna Belt, Western Australia

Cumulate komatiite Komatiitic dunite

Structure:
Thickness Thin (30-100 m) Thick (300-1000 m)
Layering Both: fine-scale cryptic-rhythmic layering
Textures Orthocumulate to Mesocumulate to
mesocumulate, crescumulate adcumulate
Grain size Fine-medium Coarse
Composition:
Whole rock 40-45010 MgO 45 - 50% MgO
Modal olivine 50-75% 75 -95070
Olivine Fo 90-95 Fo 90-95
Inferred parent 28-32% MgO 28-32070 MgO
Mineralization: High-grade, Low-grade,
stratiform, strata-bound,
massive-disseminated disseminated,
rarely massive
Probable nature: Lava conduit Subvolcanic sill

In Western Australia enough detailed work has been done that a comparison
(Table 1) can be made between the two types that are found in the Norseman-
Wiluna Belt (see Lesher et al. 1982). The cumulate komatiites and komatiitic
dunites in Western Australia appear to have formed from parental magmas of
similar composition (28 - 32070 MgO: Naldrett and Turner 1977, Donaldson and
Lesher 1982) and are interpreted to result from flow-through fractional crystal-
lization and in situ accumulation of olivine: the former as linear lava conduits
and the latter as subvolcanic magma chambers.
It is important to recognize that these deposits represent a continuum in en-
vironments of emplacement from distal volcanic through proximal and central
volcanic to subvolcanic. For example, the Shangani, Damba-Silwane, and
Langmuir #2 deposits appear to overlie feeders and probably occur in central
volcanic settings. The host komatiite sequence at the Scotia deposit contains
komatiitic intrusives and komatiitic pyroclastics, and it is considered to occur in
a proximal volcanic setting. In contrast, no feeders, komatiitic intrusives, or
pyroclastic breccias have been identified at the Kambalda or Widgiemooltha
deposits and these most likely occur in distal volcanic settings. The komatiites
were probably erupted from discrete volcanic centers within crustal rift zones (see
Groves et al. 1984). The absence of sheeted dikes and the presence of multiple
horizons of interflow sediments indicateS that this was not a midoceanic ridge en-
vironment. The setting of most komatiitic dunite-associated deposits is less
certain. The better preserved and more extensively studied examples (e.g., Six
Mile Well, Honeymoon Well, Forrestania, and Katiniq), exhibit rhythmic or
cryptic layering, are conformable with enclosing volcanics, and! or appear to be
petrogenetically related to overlying komatiitic volcanics. They are interpreted by
most workers as subvolcanic sills that fed the overlying komatiitic volcanics.
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 47

The two major deposit types appear to be concentrated in specific tectonic-

volcanic provinces and are normally spatially separated from each other. For
example, cumulate komatiite-associated deposits are concentrated in the
southern part of the Norseman-Wiluna Belt (Marston et al. 1981, Lesher et al.
1982), in the eastern part of the Abitibi Belt (Naldrett and Gasparini 1971, Coad
1979), and in Zimbabwe (Williams 1979, Clutten et al. 1981, Hammerbeck 1984).
Komatiitic dunite-associated deposits are concentrated in the northern part of the
Norseman-Wiluna Belt (Marston et al. 1981, Lesher et al. 1982), in the southern
part of the Southern Cross Province (Porter and McKay 1981), and in the
Circum-Superior Belt (Kilburn et al. 1969). Mineralized komatiitic dunites may
be overlain by barren komatiite lavas (e.g., Six Mile Well, Honeymoon Well,
Forrestania) and barren komatiitic dunites may be overlain by poorly miner-
alized cumulate komatiites (e.g., Mt. Clifford: Travis 1975), but rarely are both
mineralized.
The most important and distinctive characteristic of these deposits, as a
group, is their association with cumulate rocks of komatiitic affinity. Although
the ultramafic portion of the komatiitic suite comprises a complete continuum
from aphyric pyroxenitic komatiite (16 - 20070 MgO) and ultramafic komatiite
(20 - 32% MgO) through porphyritic komatiite (28 - 36% MgO) to ortho-
cumulate-mesocumulate-crescumulate komatiite (36 - 44% MgO) and meso-
cumulate-adcumulate komatiitic dunite (44 - 50% MgO), the nickel sulfide ores
are associated exclusively with the cumulate rocks. This restriction is observed in
Western Australia, Canada, and Zimbabwe (references above); greenstone belts
that contain komatiites, but only minor cumulate komatiites (e.g., Barberton
belt, South Africa; Murchison, Southern Cross, and East Pilbara provinces,
Western Australia) generally do not contain significant nickel sulfide deposits. It
should be noted that the composition of the liquid from which the cumulates
formed does not appear to be very important. Although most Archean deposits
are found associated with aphyric lavas ranging between 24 and 32% MgO
(Lesher et al. 1981), some Proterozoic deposits are associated with rocks derived
from liquids with as low as 19% MgO (Barnes et al. 1981). The petrogenetic
affinity of the komatiites (see Nesbitt et al. 1979) also does not appear to be
critical (cf. Marston et al. 1981). Although most important deposits are asso-
ciated with aluminum-undepleted komatiites (e.g., Kambalda: Lesher et al.
1981), some are associated with aluminum-depleted komatiites (e.g., For-
restania: Porter and McKay 1981; Ruth Well: Barley and Bickle 1982; E. G.
Nisbet and G. A. Chinner, personal communication, 1984).
As originally noted by Prider (1970) in Western Australia, most of these de-
posits are also intimately associated with certain country rocks, particularly sul-
fidic sediments. Most importantly, the distribution of the ores appears to be
related to the distribution of these rocks. These relationships are most pro-
nounced for stratiform deposits. For example, the ores at several Kambalda
shoots (e.g., parts of Juan, Lunnon, and Jan shoots: Paterson et al. 1984) grade
laterally into sulfidic quartz-albite sediments. The ores at Redstone (Robinson
and Hutchinson 1982) grade laterally into sulfide facies iron-formation. The ores
at Langmuir #2 (Green and Naldrett 1981), Mt. Edwards (INAL Staff 1975),
and Foster (Gresham and Loftus-Hills 1981) lap onto sulfidic sediments. The
48 C. M. Lesher and D. I. Groves

ores at several Kambalda shoots (e.g., parts of Ken, McMahon, and Cruickshank
shoots: Gresham and Loftus-Hills 1981), Scotia (Page and Schmulian 1981), Mt.
Windarra (Schmulian 1984), Trojan (Chimimba 1984), Texmont and Hart (Coad
1979) directly overlie sulfidic sediments that may be locally thinned beneath the
ores. The footwall rocks at the Alexo and Sothman deposits (Coad 1979) are sul-
fidic volcanics. The ores at Wannaway (McQueen 1981b) overlie altered rem-
nants of sulfidic sediments that are better preserved along strike. The ores at
many Kambalda shoots (Gresham and Loftus-Hills 1981), Nepean (Sheppy and
Rowe 1975), Widgie 3 (Groves and Lesher 1982), and Miriam (Gemuts 1975) are
stratigraphically equivalent to sulfidic sediments. The stratiform ores in several
of the Forrestania deposits occur in areas where oxide facies iron-formation is
locally thinned or absent (e.g., New Morning, Digger Rocks, Mt. Hope, and
Liquid Acrobat: Porter and McKay 1981). The ores at Cosmic Boy (Porter and
McKay 1981) apparently occur exclusively on contacts between oxide facies iron-
formation and komatiitic dunite; sulfidic sediments are also present in the foot-
wall rocks at Forrestania.
The stratigraphic relationships between ores and specific country rocks are
necessarily less specific for strata-bound deposits, but most of these deposits also
occur in sequences that contain sulfidic country rocks. For example, the footwall
sequence at Damba-Silwane (Williams 1979) contains pyritic shales in the upper
part. The Shangani deposit (Viljoen et al. 1976) occurs in a volcanic neck that
truncates a thick sulfide facies iron-formation. Mineralization occurs on the
upper and lower margins of the komatiitic dunite at Honeymoon Well
(Donaldson and Bromley 1982) and both the footwall and hanging wall
sequences contain pyritic black shales. The komatiitic dunites at Mt. Keith (Burt
and Sheppy 1975) and Six Mile (Naldrett and Turner 1977) are bounded to the
west by a thick sulfidic chert.
Massive nickel-copper sulfide ores, irrespective of whether associated with
cumulate komatiites (e.g., Kambalda) or with komatiitic dunites (e.g., Agnew,
Kataniq), normally occur in footwall embayments at the base of the host units.
The geometry of the embayments ranges from shallow, open flexures (e.g.,
Scotia, Wannaway, Agnew, Katiniq, Forrestania) to reentrant troughs (e.g.,
most Kambalda shoots). Lesher et al. (1984) argue that the embayments which
localize the ores at Kambalda are probably volcanic topographic features, modi-
fied by thermal erosion and subsequent deformation. On the basis of theoretical
and laboratory experimental studies, Huppert et al. (1984) suggested that the em-
bayments at Kambalda may have been generated entirely by thermal erosion be-
neath linear lava conduits. This is conceivable, but there are several problems
with this interpretation that remain to be resolved (Lesher 1983, Lesher et al.
1984):
1. The original geometry of all embayments has been modified to some degree by
superimposed polyphase deformation;
2. Some embayments are elliptical in outline (e.g., parts of the Ken and Juan
complexes) and could not have formed beneath linear lava conduits;
3. Although there is evidence of thermal erosion on the lateral margins of some
ore horizons, the basal contacts normally appear to be conformable with foot-
wall pillow basalts;
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 49

4. The basal contacts of some embayments correlate with cooling-unit bound-

aries in the adjacent basalt sequence and therefore appear to be strati-
graphically conformable within the footwall sequence;
5. The basal host units which define the embayments at Kambalda are derived
from the most magnesian liquids (28 - 31070 MgO) in the sequence which sug-
gests that thermal erosion and contamination may be less significant than cal-
culated by Huppert et al. (1984);
6. The embayments at the Scotia (Page and Schmulian 1981) and Wannaway
(McQueen 1981b) deposits contain preserved or partially preserved sediments
that predate the emplacement of the host unit.
The embayments that localize komatiitic dunite-associated deposits are most
likely tectonic or thermal-erosional; many (e.g., Agnew) are now so highly
tectonized that their original form and origin are virtually impossible to deter-
mine. Disseminated ores, including both fine intercumulus grains and coarse
blebs, may be marginally or centrally disposed (see Lesher et al. 1982: Table B5).
Although many host units, especially komatiitic dunites, contain ubiquitous
disseminated sulfides, some host units contain negligible mineralization outside
of the ore zones. For example, some host units at Kambalda contain widespread
low-grade disseminated mineralization (e.g., Lunnon shoot: Ross and Hopkins
1975, Keays et al. 1981), whereas others are virtually barren of sulfides outside of
the basal ore zone (e.g., Juan Main shoot: Marston and Kay 1980). At Honey-
moon Well (Donaldson and Bromley 1982) only the margins of the host unit
contain disseminated sulfides. Similarly, some parts of the mineralized dunites at
Forrestania (Porter and McKay 1981) contain only trace amounts of sulfide.
Zones of disseminated mineralization at Six Mile (Naldrett and Turner 1977)
alternate with zones containing negligible sulfides.
The majority of komatiite-associated deposits, including most large ones, are
of late Archean age (2.7 - 2.9 Ga), but they range from early Archean (3.5 Ga;
e.g., Ruth Well: Nisbet and Chinner 1981) to early Proterozoic (1.9 Ga; e.g.,
Katiniq: Barnes et al. 1982).

Field Constraints

The effects of sulfur mobility during metamorphism are difficult to evaluate (see
Groves et al. 1974, Eckstrand 1975, Groves and Keays 1979), but the spatial con-
centration of nickel sulfide mineralization in certain parts of otherwise virtually
barren host units, regardless of the nature of the metamorphism, suggests that
those host units were not uniformly saturated in sulfur at the time of
emplacement. If sulfur-saturated, the entire host unit should contain ubiquitous
disseminated mineralization: (1) sulfides should have been trapped in the chilled
margins of the host units; (2) sulfides should have continued to exsolve through-
out the host units as the lavas/magmas cooled, oxidized, and crystallized after
emplacement; and (3) flow-through fractional crystallization of sulfur-saturated
komatiitic lavas/magmas should have resulted in significant accumulation of
sulfide as well as olivine. The absence of ubiquitous mineralization in many host
50 C. M. Lesher and D.l. Groves

units, therefore, suggests that the sulfides separated locally, subsequent to

emplacement. Note, however, that the presence of widespread disseminated
mineralization does not necessitate early sulfide saturation, as the sulfides could
have separated at a late stage during crystallization.

Geochemical Constraints

Geochemical data have been used to infer that the precursor magmas at some
cumulate komatiite-associated deposits were sulfur-saturated prior to eruption
(Lesher and Groves 1984):
1. Aphyric and spinifex-textured komatiites from throughout the komatiite
sequence at Kambalda, Western Australia are significantly depleted in Ni and
Co relative to unmineralized komatiite sequences and to compositions pre-
dicted by theoretical sulfide-free fractional crystallization models (Lesher et
al. 1981);
2. Chromites in mineralized komatiite sequences in Western Australia are sys-
tematically enriched in Zn compared to those from unmineralized sequences
(Groves et al. 1977, Groves et al. 1981).
Because the geochemical features are present throughout the host komatiite
sequence(s), irrespective of whether the sampled units are mineralized or barren,
Lesher et al. (1981) concluded that they probably reflect equilibration of pre-
cursor komatiite magmas with sulfides prior to eruption. It has been recently re-
cognized, however, that the host komatiite flows are probably laterally very
extensive (see Lesher et al. 1984). As the structure and dimensions of the miner-
alized volcanic piles are not known, it is quite possible that the barren upper parts
of a mineralized sequence in one area of the volcanic belt may be the facies
equivalent of mineralized parts of a sequence in another area of the belt.
Although the geochemical data may indeed reflect equilibration between the host
magmas and sulfides, this does not necessarily restrict that process to the pre-
eruptive stage.

Physical Constraints

Despite the fact that sulfides in these deposits are widely assumed to have been
derived from the mantle or from low level magma chambers, the feasibility of
transporting dense sulfides in low-viscosity komatiite magmas has not been pre-
viously assessed. This problem has normally been dismissed under the assump-
tion that the precursor magmas were enriched in intratelluric olivine which would
significantly increase their effective viscosity (Marston et al. 1981) and, there-
fore, their ability to transport sulfides. Relict olivine is rare at most deposits, but
the systematic compositional and textural variations exhibited by that which has
been studied (Stolz 1981, Lesher 1983, Donaldson 1983) indicates that most is not
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 51

intratelluric. This, plus the absence of significant amounts of olivine phenocrysts

in the chilled margins of most komatiite lava flows, suggests that the precursor
komatiitic magmas were phenocryst-poor. The cumulate nature of these units is
attributed to flow-through fractional crystallization and in situ accumulation of
olivine.
As noted above, dense sulfides that have segregated at the base of a magma
chamber are unlikely to be erupted. It is possible, however, that minor amounts
of dispersed sulfides may be entrained in a komatiitic magma without signif-
icantly increasing its bulk density. Assuming that some concentration of sulfides
has occurred, regardless of the precise model for sulfide emplacement (see review
by Lesher et al. 1984), the maximum amount of sulfides is that presently
preserved in the ore zones - which Lesher et al. (1981) calculate to be of the
order of 20/0. Concentrations of sulfides less than 2% would have negligible
influence on the bulk density (and viscosity) of the komatiite magma.
The feasibility of transporting sulfides in an ascending komatiite magma may
be evaluated in terms of fluid mechanics. Iron-nickel sulfides of the primary
compositions inferred for these deposits will be molten (Kullerud 1963) at all
temperatures above the komatiite solidus (Arndt 1976b). Extensive studies of the
motion of solid particles, fluid droplets, and gas bubbles in an infinite medium
have shown, in general, that the drag coefficient, and consequently the terminal
setting velocity, are a function of the dimensionless Reynolds' number:

Re = (}dU (1)
J.l
where (} is the density of the continuous phase, d the diameter of the dispersed
phase, U the velocity of the dispersed phase, and J.l the viscosity of the continuous
phase. Analytical solutions are only available when inertial effects are ignored,
that is, when Re ~ 1, but empirical solutions and theoretical models are available
at higher Reynolds' numbers.
Hubbert et al. (1984) have noted that low viscosity komatiites should ascend
turbulently. Experimental studies have shown that particles respond to turbulent
fluctuations 3 with a scale larger than the particle diameter. The terminal velocity
of small particles is not affected by free-stream turbulence, but that of large
particles may increase or decrease as a result of changes in the flow regime (Clift
et al. 1978). The overall effect of turbulence appears to be to increase the drag
coefficient and consequently to decrease the settling velocity. As such, the
analytical solutions and empirical correlations discussed below can be considered
to represent maximum settling velocities.
The terminal settling velocity of small sulfide droplets at low Reynolds'
numbers (Re < 2: Stokes flow regime) in a komatiite magma may be estimated
from Stokes' law:

3 Other effects of turbulence that are more difficult to evaluate are (1) increases in the probability of
hydrodynamic interference between droplets, leading to coalescence and (2) increases in shear
stresses at the surface of the droplet, leading to breakup. The opposing influences of these pro-
cesses lead to the development of a steady state droplet size distribution.
52 c. M. Lesher and D. I. Groves

U _ gd2({}p_{})
(2)
TS - 18.u
where U TS is the terminal settling velocity, g the gravitational acceleration, d the
droplet diameter, {}p the density of the sulfide droplet, {} the density of the
komatiite, and.u the viscosity of the komatiite. There are solutions which contain
terms to account for internal circulation in a fluid droplet (i.e., Hadamard-
Rybczynski formula: see Clift et al. 1978, Eq. 3 -15), but Eq. (2) is sufficiently
accurate for our purposes. Note that terminal settling velocity in the Stokes'
regime is directly proportional to the difference in density between the two
phases and inversely proportional to the viscosity of the continuous phase, but
that it is proportional to the surface area of the dispersed phase and, therefore, to
the square of the diameter of the droplet. At higher Reynolds' numbers the in-
fluences of the viscosity of the continuous phase and the dimension of the dis-
persed phase become less important.
The terminal settling velocity of moderate-sized droplets at moderate
Reynolds' numbers (2 < Re < 500: laminar flow regime) may be estimated from
the following empirical correlation of the standard drag curve (McCabe and
Smith 1976: Eq. 7 - 51):
0.153 gO.71 d1.14({}p_ {} )0.71
U TL = -------,.-=----;;-...,..:----
{} 0.29 .u 0.43
(3)

The terminal settling velocity of larger droplets at higher Reynolds' numbers

(500 < Re < 2 X 10 5: turbulent flow regime) may be estimated from the following
empirical correlation (McCabe and Smith 1976: Eq. 7 - 52):

U TN = 1.75 gd({}p_{})1/2 (4)

{}

Using values of 4.2 g cm - 3 for {}p of an iron-nickel sulfide melt (Dobrovinski

et al. 1969), and 2.75 g cm -3 for {} and 1 poise for .u of a 32070 MgO komatiite
(Nisbet 1982, Lesher 1983), terminal settling velocities and Reynolds' numbers
for different-sized sulfide droplets are given in Table 2. For a range of plausible
fissure widths (0.3 -10 m), Huppert et al. (1984) calculated that the ascent
velocity of komatiites should be within the range 190 - 1100 cm s -1. Although
this range may be realistic for well-established conduits, most of the ores in

Table 2. Terminal settling velocities and Reynolds' numbers of sulfide droplets in a komatiite magma

d(cm) Re

0.01 0.0079 0.00022

0.03 0.071 0.0059
0.10 0.79 0.22
0.30 5.0 4.1
1.0 20 54
3.0 69 570
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 53

cumulate komatiite-associated nickel sulfide deposits are associated with the

initial flow of the komatiite sequence. The first magmas probably expended more
energy during intrusion and ascended more slowly than later magmas which
passed through previously established conduits. The magnitude of the difference
in ascent velocity is difficult to evaluate, but it is clear that at least fine sulfide
droplets could be supported and transported in an ascending komatiite magma.

Thermodynamic Constraints

Although it appears physically feasible to transport dispersed sulfide droplets in

an ascending komatiite magma, there are several thermodynamic factors that
militate against the eruption of sulfide-bearing komatiites:
1. Firstly, komatiitic melts are generated by high-temperature, high-degree
partial melting of mantle source rocks. As the dissolution of mantle sulfide
will depend primarily on the amount of melt and the temperature of the melt,
komatiites are the least likely of all melts to remain sulfur-saturated during
formation. For example, the amount of sulfide that can dissolve in a 50"10
komatiitic melt with a sulfur content of the order of 3000 ppm at sulfide
saturation (Shima and Naldrett 1975) is 15 times that of a 10"10 basaltic melt
with a sulfur content of the order of 1000 ppm at sulfide saturation
(Haughton et al. 1974). Generation of a 50% partial melt containing
3000 ppm S would require a' source with a minimum sulfur content of
1500ppm. The sulfur content of the mantle is not precisely known, but most
estimates are very much lower, of the order of 350 ppm (Ringwood and
Kesson 1976, Sun 1981).
2. Secondly, independent experimental studies by Helz (1977), Huang and
Williams (1980), and Wentlandt (1982) have shown (a) that there is a strong
negative pressure dependence on sulfur solubility in basaltic melts and (b) that
sulfur saturation isopleths are parallel to anhydrous silicate liquidi (see
Fig. 1). If sulfur-saturated at depth, magmas which ascend along liquidus
paths should be sulfur-saturated during ascent and may therefore exsolve an
immiscible sulfide liquid during ascent and after eruption. However, as noted
above, there is no evidence of significant intratelluric phenocryst contents in
komatiites, and magmas of such low viscosity should ascend rapidly and lose
little heat. Komatiites should ascend along steep P - T trajectories, nearer to
the adiabatic gradient, and should arrive at the surface superheated and
sulfur-undersaturated. This is consistent with the field observations.
3. Finally, if komatiites are formed by multistage melting or mixing processes
(e.g., Arndt 1977, Bickle et al. 1977), they are even less likely to remain sulfur-
saturated. Sulfur may partition into early melts and/or be physically lost as
sulfide during melt segregation, and residual melts will be driven off the sulfur
saturation surface by continued partial melting/mixing.
54 C. M. Lesher and 0.1. Groves

0 0
POSSIBLE P- T
10
5 TRAJECTORIES
FOR KOMA TIITES 20

10 30

40
15
kb km 50

20 60

70
25
80

30L-~1~2~00~~1~3~0~0~1~4~0~0--1~5~0~0~~16~OVO'-~~-'~--~~--~ 90
1900
°c
Fig. 1. Sulfide saturation model showing limits on possible pressure-temperature trajectories for
ascending komatiite magmas. The dashed curves are liquidus ascent paths (Arndt 1976b, Bickle et at.
1977. Bickle 1978). The solid curves are adiabatic ascent paths (dTIdP = 3 °C/kb). Sulfide saturation
isopleths are assumed to be parallel to the silicate Iiquidii (see references in text). $ = sulfide-saturated
komatiites, 0 = sulfide-undersaturated komatiites

Discussion

Together, the above field, physical, and thermodynamic constraints suggest that
most komatiites should not have been sulfide-saturated on emplacement. It may
be physically feasible to transport dispersed sulfide droplets in a rapidly ascend-
ing komatiite magma, but the field evidence and thermodynamic constraints sug-
gest that this probably did not occur. In the case of finely-disseminated mineral-
ization in komatiitic dunites, sulfide saturation must have occurred late in the
crystallization history of the host units in order to produce fine-grained, inter-
stitial textures. The sulfur in these deposits may have been mantle-derived,
through dissolution during partial melting and exsolution during crystallization.
In contrast, massive sulfides in cumulate komatiites must have formed early in
the crystallization history of the host units in order to provide the opportunity
for sulfide separation and segregation. If the komatiitic hosts for these deposits
were not sulfide-saturated on eruption, as inferred above, then some process
must have induced sulfide separation at a later stage.
Sulfide separation occurs in a silicate melt when it becomes saturated in S2- .
Sulfur solubility in silicate melts have been shown to decrease with: (1) decreasing
temperature; (2) decreasing aFeO or increasing aSi0 2; and/or (3) decreasingjS2
or increasing 10 2 (MacLean 1969, Haughton et al. 1974, Shima and Naldrett
1975, Buchanan and Nolan 1979). As the cumulate komatiite host units are inter-
preted as dynamic lava conduits and the komatiitic dunite host units as dynamic
magma chambers, temperatures were probably maintained at relatively high
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 55

levels during emplacement. Similarly, variations in 102 or fS2 are unlikely to

have been very significant, although sulfur may have been volatilized during
eruption (Lesher et al. 1981). Direct contamination of the lava/magma by silica,
ferric iron, or sulfide is the most feasible mechanism for inducing rapid, signif-
icant sulfide exsolution. The cherty, sulfidic sediments at Kambalda, Windarra,
Widgiemooltha, and Nepean, for example, are rich in silica and sulfide (Groves
et al. 1976, Bavinton 1981). The oxide facies iron-formations at Forrestania are
rich in silica and ferric iron oxide.
On the basis of stratigraphic relationships and physical volcanology, Lesher
et al. (1984) proposed a volcanic assimilation model for Kambalda. In light of the
intimate spatial and stratigraphic relationships between cumulate komatiite-asso-
ciated nickel-copper sulfide deposits and certain types of country rocks, assimila-
tion of country rocks is considered to be the most probable mechanism for induc-
ing localized sulfide saturation and sulfide separation in these deposits. The
necessity of an external sulfur source or sulfide precipitating mechanism is less
pronounced in the case of the disseminated mineralization in komatiitic dunites,
but the distribution of sulfides in some of those deposits also appears to be
intimately related to the distribution of sulfidic country rocks . Contamination of
mafic-ultramafic magmas by either sulfide or siliceous country rocks has been
suggested as a mechanism for inducing sulfide saturation for most other mag-
matic sulfide deposits (see summary by Naldrett 1981), and it is likely that
komatiite-associated nickel-copper sulfide deposits are no exception.
In such a model (Fig. 2), the texture and distribution of mineralization will be
influenced by (1) the mode of emplacement (volcanic vs subvolcanic) and degree
of assimilation; (2) the relative volume of magma (lava flow vs magma chamber)
and its capacity to dissolve sulfur; and (3) the timing of sulfide saturation and
opportunity for sulfide segregation. For example, local assimilation of siliceous,
sulfidic, or ferric iron oxide-rich footwall rocks by thermal erosion beneath

CENTRAL PROXIMAL DISTAL

DISSEMINA TED
: . ~LEBBY SULFIDES
.... ! .
cumu la te komatiite
MASSIVE SULPHID
SULFUR ASSIMILATION
komatiitic dunite
SUBVOLCANIC

sulfidic sediment

' SULFUR ASSIMILATION

NOt TO SCALE

Fig. 2. Generalized model for the formation of komatiite-associated nickel sulfide deposits by assimi-
lation of crustal sulfur. Sulfides are shown in black: solid for sulfide ores (dotted = disseminated;
spotted = blebby; filled = massive), striped for sulfidic sediments. Note that sulfide ores may occur in
a range of settings from distal volcanic through proximal and central volcanic to subvolcanic
56 c. M. Lesher and 0.1. Groves
narrow lava conduits, would result in sulfide separation prior to crystallization
of the host unit - producing basal stratiform massive ores with overlying dis-
seminated mineralization. Variations in the mode of emplacement of the host
unit would influence the degree of assimilation and may account for variations in
the intensity of mineralization and for differences in the stratigraphic relation-
ships with the sulfidic sediments that are observed in different cumulate komati-
ite-associated deposits. In contrast, assimilation of siliceous, sulfidic, or ferric
iron oxide-rich country rocks during intrusion of a more voluminous komatiitic
dunite body may not induce immediate sulfide saturation and sulfide separation
could be delayed until late in the crystallization history - producing dis-
seminated ores. Blebby sulfides are too coarse (av 1 em diam) to remain sus-
pended in the magma and too abundant to have exsolved in situ (Lesher 1983);
they must have formed during flow-through crystallization of the host unit. In all
cases, it is the unusual thermal and physical characteristics of komatiitic magmas
that influenced their mode of emplacement (Lesher et al. 1984) and promoted
thermal erosion and assimilation (Huppert et al. 1984).
At present there are few data available to properly test this model, but the fol-
lowing may be considered to be strong corroborative evidence:
1. Sulfur Isotopes. Sulfur isotopic data for ores and possible sulfide sources are
available for only a few of the better characterized komatiite-associated
deposits. These data (Fig. 3) indicate that sulfides in ores and country rocks at
individual deposits generally exhibit indistinguishable sulfur isotopic ratios
that are systematically different from those of other deposits of this type.
Sulfur/selenium ratios exhibit similar trends. Although the sulfur in the ores
and country rocks could have formed independently from the same sulfur
source, the sulfur isotopic variations between deposits are not consistent with
derivation from a homogeneous mantle source. Fractionation of sulfur iso-
topes in the hydrothermal-exhalative-sedimentary environment with sub-
sequent incorporation in and exsolution from komatiitic magmas is the most
likely explanation for the variation between deposits.

10
Alexo, CIJ--ORE Fig. 3. Sulfur isotopic data for
Ontario --+-
2
COUNTRY ROCK several cumulate komatiite-asso-
ciated nickel sulfide deposits.
62 Data from Naldrett (1966). Sec-
Kambalda, ~ORE
combe et al. (1978, 1981), and
Western Australia ~SEDIMENT
21 Green and Naldrett (1981). Data
18
given as mean (solid bar), stan-
ORE ----c::::::I::: Langmuir, dard deviation (closed box), and
--IC==:J'==::Jf.- COUNTRY ROCK Ontario total range (horizontal line). Note
5
that the ores and sulfidic country
31 rocks at most individual deposits
ORE ------[IJ-- Mt Windarra,
SEDIMENT -----l , I--- Western Australia exhibit similar and commonly in-
28
distinguishable sulfur isotopic
ratios that are systematically
-8 -6 -4 -2 0 2 4 6 8
different from those of the other
8
34 8%0 deposits
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 57

2. Ore Tenor Variations. The composition of the sulfide fraction of different ore
shoots in Western Australia varies between different deposits and between dif-
ferent shoots within a single deposit (Ross and Keays 1979, Marston and Kay
1980, Woolrich et al. 1981). The limited range of parental liquid compositions
for the host units (generally 28 - 32070 MgO) indicates that this cannot be at-
tributed to variations in the compositions of the precursor magmas. Further-
more, positive correlations of Ni/Cu, Ni/Co, and Pd/Pt with increasing Ni
indicates that this cannot be attributed to variations inj02 andfS2 among dif-
ferent magma batches, as major variations in these parameters are unlikely
and cannot be expected to change the relative partitioning of the chalcophile
elements. The thicker ore shoots in Western Australia are generally of lower
grade than the thinner ones (Marston et al. 1981), which indicates a relation-
ship between ore tenor and the amount of sulfide that may be attributed to
variations in the degree of sulfide assimilation. Campbell and Naldrett (1979)
have shown that the compositions of sulfide melts are markedly influenced by
variations in the magma:sulfide ratio.

Lesher et al. (1984) present stratigraphic, volcanologic, and theoretical argu-

ments that specifically support the possibility of sulfidic sediment assimilation at
Kambalda. Careful stratigraphic and volcanologic studies at other deposits, in
conjunction with systematic sulfur and lead isotopic studies will be required to
fully evaluate this as a generalized model.
The metallogenesis of komatiite-associated nickel-copper sulfide deposits is
probably tectonically controlled. Groves and Batt (1984) define two general types
of Archean greenstone belts that differ in terms of timing and degree of crustal
extension. Platform-phase greenstones are characterized by: equispaced grani-
toid batholiths with intervening stellate greenstone belts, a coherent volcanic
stratigraphy, abundant komatiitic basalts, and shallow water volcaniclastics or
oxide facies iron-formation. They are interpreted to have formed in relatively
shallow water under conditions of low crustal extension. Rift-phase greenstones
are characterized by: elongate granitoid domes, linear tectonic patterns, complex
volcanic stratigraphy, abundant komatiites (especially cumulate komatiites), and
sulfidic shales and cherts. They are interpreted to have formed in relatively deep
water under conditions of high crustal extension. The rift-phase greenstones,
typified by the Norseman-Wiluna Belt of Western Australia and the Abitibi Belt
of Canada, are distinctly better mineralized. The spatial separation between the
two associations, volcanic cumulate komatiite and subvolcanic komatiitic dunite,
is probably also related to the degree of tension in the crust (see Roberts 1970).
The absence of subvolcanic komatiitic dunites in the southern part of the Norse-
man-Wiluna Belt, for example, probably indicates a relatively high degree of ex-
tension so that all komatiitic magmas were erupted. In contrast, the presence of
both subvolcanic komatiitic dunites and komatiite lavas in the northern part of
the Norseman-Wiluna Belt, for example, probably indicates a somewhat lower
degree of extension or a smaller amount of supracrustal rocks, so that komatiitic
magmas were intruded and then erupted. This may result (1) from longitudinal
variations within a rift zone; (2) from lateral variations across a rift zone; or (3)
from temporal variations during evolution of a rift zone (Groves et al. 1984). In
58 C.M. LesherandD.1. Groves

any case, both environments were conducive to the rapid, continued eruption/ir-
ruption of komatiitic magma as manifested by dynamic lava conduits/magma
chambers.

Conclusions

Previous models for the formation of these deposits inferred derivation of

sulfides directly from the mantle or from low-level magma chambers, but there
are several field, physical, and thermodynamic constraints which suggest that
this may not be feasible: (1) the spatial concentration of mineralization within
certain parts of otherwise barren host units suggests that the latter were not uni-
formly saturated in sulfide at the time of emplacement; (2) there are physical
problems in transporting dense sulfides in low viscosity komatiite magmas,
although dispersed droplets may be supported during ascent; and (3) high-tem-
perature komatiitic melts generated by high-degree partial melting, or sequential
melting, of mantle source rocks are unlikely to be sulfide-saturated in the source
region or to remain sulfide-saturated during ascent.
These deposits are intimately associated with specific country rocks, par-
ticularly sulfidic sediments, and the distribution of such rocks appears to be
related to the distribution of the Fe - Ni - Cu sulfide mineralization. It is most
likely that the enhanced sulfur contents of the mineralized komatiitic rocks are
due to assimilation of siliceous, sulfidic, or ferric iron-rich country rocks during
emplacement. The distribution and texture of the mineralization was probably
influenced by (1) the mode of emplacement of the magma (volcanic vs sub-
volcanic) and degree of assimilation; (2) the relative volume of magma (lava
conduit vs magma chamber) and its capacity to dissolve sulfur; and (3) the timing
of sulfide separation and opportunity for segregation. For example, local assimi-
lation of sulfidic sediments by thermal erosion beneath narrow lava conduits,
such as interpreted for the Kambalda deposits, would result in sulfide separation
and segregation prior to crystallization of the host unit - producing stratiform
massive ores. In contrast, assimilation of sulfidic country rocks during intrusion
of a larger komatiitic dunite body may not induce sulfide saturation immediately
and sulfide separation could be delayed until late in the crystallization history -
producing strata-bound disseminated ores.
A fundamental aspect of this model is the coincidence of cumulate komatiitic
host units and sulfidic country rocks, the former representing a source of heat for
assimulation (a consequence of lava/magma replenishment) and of ore metals,
and the latter a source of sulfur or of components to induce sulfide separation.
Certain rift-phase greenstones appear to have formed in deeper water under con-
ditions of greater extension than other platform-phase greenstones and, there-
fore, represent the most favorable tectonic environment for both voluminous
komatiite eruption/irruption and accumulation of sulfidic sediments. Variations
in the degree of extension along or within the rift and/or in the tectonic timing of
emplacement may control the mode of emplacement and, therefore, the type of
deposit.
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 59

Acknowledgments. C.M.L. acknowledges the support of a University of Western Australia research

scholarship and a University of Alabama Faculty Research Grant (No. 1254). C.M.L. is very grateful
to Kambalda Nickel Operations, Western Mining Corporation, for supporting his research on the
Kambalda deposits. Both authors are grateful to the many mining companies in Western Australia
who granted access to deposits and extended many courtesies in the field, and to Mmes. S. Simpson
and S. Miller for their skillful typing of the manuscript. The paper has been improved as a result of
constructive comments by Prof. A.J. Naldrett, Dr. H.E. Huppert, and two anonymous referees.

References

Allegre CJ (1982) Genesis of komatiites in a wet plate. In: Arndt NT, Nisbet EG (eds) Komatiites.
Allen & Unwin, London, pp 495 - 500
Arndt NT (1976a) Ultramafic rocks of Munro Township: Economic and tectonic implications. In:
Strong DF (ed) Metallogeny and plate tectonics. Geol Assoc Can Spec Pap 14:617 - 657
Arndt NT (1~76b) Melting relations of ultramafic lavas (komatiites) at 1 atm and high pressure.
Carnegie Inst Washington Yearb 75:555 - 562
Arndt NT (1977) Ultrabasic magmas and high degree melting of the mantle. Contrib Mineral Petrol
64:205-221
Arndt NT (1983) The role of a thin, komatiite-rich oceanic crust in the Archean plate-tectonic
process. Geology 11 :372 - 375
Arndt NT, Brooks C (1980) Komatiites. Geology 8:155 -156
Arndt NT, Nisbet EG (eds) (1982) Komatiites. Allen & Unwin, London, 526 pp
Barley ME, Bickle MJ (1982) Komatiites in the Pilbara Block, Western Australia. In: Arndt NT,
Nisbet EG (eds) Komatiites. Allen & Unwin, London, pp 105 -116
Barnes SJ, Coats CJA, Naldrett AJ (1981) Petrogenesis of a proterozoic nickel sulfide-komatiite
association: The Katiniq sill, Ungava, Quebec. Econ Geol 77 :413 - 429
Bavinton OA (1981) The nature of sulfidic sediments at Kambalda and their broad relationships with
associated ultramafic rocks and nickel ores. Econ Geol 76:1606-1628
Bickle MJ (1978) Melting experiments on peridotitic komatiites. In: Progress in experimental
petrology. Natl Energy Res Counc Prog Rep 4 Ser D 11: 187 - 195
Bickle MJ, Ford CE, Nisbet EG (1977) The petrogenesis of peridotitic komatiites: Evidence from
high-pressure melting experiments. Earth Planet Sci Lett 37:97 -106
Billington LG (1984) Geological review of the Agnew nickel deposit, Western Australia. In:
Buchanan DL, Jones MJ (eds) Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall,
London, pp 43 - 54
Buchanan DL, Nolan J (1979) Solubility of sulfur and sulfide immiscibility in synthetic tholeiite melts
and their relevance to Bushveld-complex rocks. Can Miner 17:483-494
Burt DRL, Sheppy NR (1975) Mount Keith nickel sulphide deposit. In: Knight CL (ed) Economic
geology of Australia and Papua-New Guinea, I. Metals. Australas Inst Min Metall Monogr
5:159-268
Campbell IH, Naldrett AJ (1979) The influence of silicate:sulfide ratios on the geochemistry of
magmatic sulfides. Econ Geol 74:1503 -1505
Chimimba LR (1984) Geology and mineralogy at Trojan Mine, Zimbabwe. In: Buchanan DL, Jones
MJ (eds) Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall, London, pp 147 -155
Clift R, Grace JR, Weber ME (1978) Bubbles, drops and particles. Academic Press, London New
York, 380 pp
Clutten JM, Foster RP, Martin A (1981) Nickel mineralization in Zimbabwe. Episodes 2:10-15
Coad PR (1979) Nickel sulphide deposits associated with ultramafic rocks of the Abitibi Belt and
economic potential of mafic-ultramafic intrusions. Ontario Geol Surv Stud 20:84 pp
Dobrovinski IE, Esin OA, Barmin LN, Chuchmarev SK (1969) Physicochemical properties of
sulphide melts. Russ J Phys Chern 43:1769 -1771
Donaldson MJ (1983) Progressive alteration of barren and weakly mineralized archaean dunites: A
petrological, mineralogical and geochemical study of some intrusive dunites from Western
Australia. Ph D Thesis, Univ W Aust, 345 pp (unpublished)
60 C. M. Lesher and D.1. Groves

Donaldson MJ, Bromley GJ (1982) The Honeymoon Well nickel sulfide deposits, Western Australia.
Econ Geol 76:1550-1564
Donaldson MJ, Lesher CM (1982) Intrusive dunite and extrusive peridotite: Analogous products of
komatiitic magmas resulting from different cooling rates (abstract). International Geological Cor-
relation Program Projects 161 and 91, Nickel Suphide Field Conference III, Perth, Western
Australia, Abstr Vol, pp 31 - 32
Eckstrand OR (1975) The Dumont serpentinite: A model for control of nickeliferous opaque assem-
blages by alternation products in ultramafic rocks. Econ Geol 70:183 - 201
Gemuts I (1975) Miriam nickel prospect, Coolgardie area. In: Knight CL (ed) Economic geology of
Australia and Papus New Guinea. I. Metals. Australas Inst Min Metall Monogr 5:98 - 99
Green AH, Naldrett AJ (1981) The Langmuir volcanic peridotite-associated nickel deposits:
Canadian equivalents of the Western Australian occurrences. Econ Geol 76:1503 -1523
Gresham JJ (1986) Depositional environments of volcanic peridotite-associated nickel sulphide
deposits with special reference to the Kambalda dome, this Vol.
Gresham JJ, Loftus-Hills GD (1981) The geology of the Kambalda nickel field, Western Australia.
Econ Geol 76:1373 -1416
Groves 01, Batt WD (1984) Spatial and temporal variations of Archaean metallogenic associations in
terms of evolution of granitoid-greenstone terrains with particular emphasis on the Western
Australian Shield. In: KrOner A, Hanson GN, Goodwin AM (eds) Archean geochemistry.
Springer, Berlin Heidelberg New York, pp 73 - 98
Groves 01, Keays RR (1979) Mobilization of ore-forming elements during alteration of dunites, Mt.
Keith-Betheno, Western Australia. Can Miner 17:373-389
Groves 01, Lesher CM (eds) (1982) Regional geology and nickel deposits of the Norseman-Wiluna
Greenstone Belt, Western Australia. Univ W Aust Dep Geol Extension Serv Publ 7:234 pp
Groves 01, Hudson DR, Hack TBC (1974) Modification of iron-nickel sulfides during serpentiniza-
tion and talc-carbonate alteration at Black Swan, Western Australia. Econ Geol 69:1265 -1281
Groves 01, Barrett FM, McQueen KG (1976) Geochemistry and origin of cherty metasediments
within ultramafic flow sequences and their relationship to nickel mineralization. In: Glover JE.
Groves DI (eds) Archaean cherty metasediments: Their sedimentology, micropaleontology, bio-
geochemistry, and significance to mineralization. Univ W Aust Dep Geol Extension Serv Publ
2:57-69
Groves 01, Barrett FM, Binns RA, McQueen KG (1977) Spinel phases associated with meta-
morphosed volcanic-type iron-nickel sulfide ores from Western Australia. Econ Geol
72:1224-1244
Groves DI, Barrett FM, McQueen KG (1979) The relative roles of magmatic segregation, volcanic
exhalation and regional metamorphism in the generation of volcanic-associated nickel ores of
Western Australia. Can Miner 17:319-336
Groves DI, Barrett FM, Brotherton RH (1981) Exploration significance of chrome-spinels in miner-
alized ultramafic rocks and Ni-Cu ores. In: Villiers JPR de, Cawthorn GA (eds) ICAM 81: Proc
1st Int Conf Applied Mineralogy, Johannesburg. Geol Soc S Afr Spec Publ 7:21- 30
Groves 01, Lesher CM, Gee DG (1984) Tectonic setting of the sulphide nickel deposits of Western
Australia. In: Buchanan DL, Jones MJ (eds) Sulfide deposits in mafic and ultramafic rocks. Inst
Min Metall, London, pp 1 - 13
Hammerbeck ECI (1984) Aspects of nickel metallogeny of Southern Africa. In: Buchanan DL, Jones
MJ (eds) Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall, London, pp 135 -140
Haugthon DR, Roeder PL, Skinner BJ (1974) Solubility of sulfur in mafic magmas. Econ Geol
69:451-462
Helz RT (1977) Determination of the P- T dependence of the first appearance of FeS-rich liquid in
natural basalts to 20 kb (abstract). EOS 58:523
Huang W-L, Williams RJ (1980) Melting relations of the system Fe- S - Si - 0 to 32 kb with implica-
tions as to the nature of the mantle-core boundary (abstract). Lunar Planetary Science Conference
XI. Lunar Planet Sci Inst, Houston, pp 486 - 488
Huppert HE, Sparks RSJ, Turner JS, Arndt NT (1984) Emplacement and cooling of komatiite lavas.
Nature (London) 309:19-22
INAL Staff (1975) BHP/INAL nickel sulphide occurrences of the Widgiemooltha area. In: Knight
CL (ed) Economic geology of Australia and Papua New Guinea. I. Metals. Australas Inst Min
Metall Monogr 5:86-89
Controls on the Formation of Komatiite-Associated Nickel-Copper Sulfide Deposits 61

Keays RR, Ross JR, Woolrich P (1981) Precious metals in volcanic peridotite-associated nickel
sulfide deposits in Western Australia, II: Distribution within ores and host rocks at Kambalda.
Econ Geol 76:1645 -1674
Kilburn LC, Wilson HDB, Graham AR, Ogura Y, Coats CJA, Scoates RFJ (1969) Nickel sulfide ores
related to ultrabasic intrusions in Canada. In: Wilson HDB (ed) Magmatic ore deposits. Econ Geol
Monogr 4:267 - 293
Kullerud G (1963) The Fe- Ni - S system. Carnegie Inst Washington Yearb 62:175 -189
Lesher CM (1983) Localization and genesis of komatiite-hosted Fe - Ni - Cu sulphide mineralization
at Kambalda, Western Australia. Ph D Thesis, Univ W Aust, 318 pp (unpublished)
Lesher CM, Groves DI (1984) Geochemical and mineralogical criteria for the identification of miner-
alized komatiites in Archaean greenstone belts of Australia. Proc 27th Int Geol Congr. Petrology:
Igneous and Metamorphic Rocks 9:283 - 302
Lesher CM, Lee RF, Groves DI, Bickle MJ, Donaldson MJ (1981) Geochemistry of komatiites at
Kambalda: I. Chalcophile element depletion - a consequence of sulfide liquid separation from
komatiitic magmas. Econ Geol 76:1714-1728
Lesher CM, Donaldson MJ, Groves DI (1982) Nickel deposits and their host rocks in the Norseman-
Wiluna Belt. In: Groves DI, Lesher CM (eds) Regional geology and nickel deposits of the Norse-
man-Wiluna Belt, Western Australia. Univ W Aust Dep Geol Extension Serv Publ 7, B:l- 63
Lesher CM, Arndt NT, Groves DI (1984) Genesis of komatiite-associated nickel sulphide deposits at
Kambalda, Western Australia: A distal volcanic model. In: Buchanan DL, Jones MJ (eds)
Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall, London, pp 70- 80
MacLean WH (1969) Liquids phase relations in the FeS - FeO - Fe203 - Si0 2 system, and their
application in geology. Econ Geol64:865 - 884
Marston RJ, Kay BD (1980) The distribution, petrology, and genesis of nickel ores at the Juan
complex, Kambalda, Western Australia. Econ GeoI75:546-565
Marston RJ, Groves 01, Hudson DR, Ross JR (1981) Nickel sulphide deposits in Western Australia:
A review. Econ Geol 76:1330-1363
McCabe WL, Smith JC (1976) Unit operations of chemical engineering. McGraw-Hili, New York,
1028 pp
McQueen KG (1981a) Volcanic-associated nickel deposits from around the Widgiemooltha dome,
Western Australia. Econ Geol 76:1417 -1443
McQueen KG (1981b) The nature and metamorphic history of the Wannaway nickel deposit, Western
Australia. Econ Geol 76:1444-1468
Naldrett AJ (1966) The role of sulphurization in the genesis of iron-nickel sulphide deposits of the
Porcupine District, Ontario. Can Inst Min Metall Trans 69:147 -155
Naldrett AJ (1973) Nickel sulphide deposits - their classification and genesis with special emphasis
on deposits of volcanic association. Can Inst Min Metall Bull 66:45 - 63
Naldrett AJ (1979) Partitioning of Fe, Co, Ni, and Cu between sulfide liquid and basaltic melts and
the composition of Ni - Cu sulfide deposits - a reply and further discussion. Econ Geol
74:1520-1528
Naldrett AJ (1981) Nickel sulfide deposits: Classification, composition and genesis. In: Skinner BJ
(ed) Econ Geol75th Anniversary Vol. Econ Geol Publ Co, El Paso, pp 628-685
Naldrett AJ, Cabri LJ (1976) Ultramafic and related rocks: Their classification and genesis with
special reference to the concentration of nickel sulfides and platinum-group elements. Econ Geol
71:1131-1158
Naldrett AJ, Gasparini EL (1971) Archaean nickel sulphide deposits in Canada: Their classification,
geological setting and genesis with some suggestions as to exploration. In: Glover JE (ed) The
Archaean Rocks. Geol Soc Aust Spec Publ 3:331- 347
Naldrett AJ, Turner AR (1977) The geology and petrogenesis of a greenstone belt and related nickel
sulfide mineralization at Yakabindie, Western Australia. Precambrian Res 5:43 -103
Nesbitt RW (1971) The case for liquid immiscibility as a mechanism for nickel sulphide
mineralization in the Eastern Goldfields, Western Australia ~abstract). In: Glover JE (ed) The
Archaean Rocks. Geol Soc Aust Spec Publ 3:253
Nesbitt RW, Sun S-S, Purvis AC (1979) Komatiites: Geochemistry and genesis. Can Miner
17:165 -186
Nisbet EG (1982) The tectonic setting and petrogenesis of komatiites. In: Arndt NT, Nisbet EG (eds)
Komatiites. Allen & Unwin, London, pp 501 - 520
62 C. M. Lesher and D.1. Groves: Controls on the Formation of Komatiite-Associated

Nisbet EG, Chinner GA (1981) Controls on the eruption of mafic and ultramafic lavas, Ruth Well
Ni-Cu prospect, West Pilbara. Econ GeoI76:1719-1735
Nisbet EG, Walker D (1982) Komatiites and the structure of the Archaean mantle. Earth Planet Sci
Lett 60:105 -113
Page ML, Schmulian ML (1981) The proximal volcanic environment of the Scotia nickel deposits.
Econ GeoI76:1469-1479
Paterson HL, Donaldson MJ, Smith RN, Lenard MF, Gresham JJ, Boyack DJ, Keays RR (1984)
Nickeliferous sediments and sediment-associated nickel ores at Kambalda, Western Australia. In:
Buchanan DL, Jones MJ (eds) Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall,
London, pp 81 - 94
Porter DJ, McKay KG (1981) The nickel sulfide mineralization and metamorphic setting of the For-
restania area, Western Australia. Econ GeoI76:1524-1549
Prider RT (1970) Nickel in Western Australia. Nature (London) 226:691 - 693
Ringwood AE, Kesson SE (1976) Basaltic magmatism and the composition of the moon, Part II.
Siderophile and volatile elements in the moon, earth and chondrites: Implications for lunar origin.
Aust Natl Univ, Res School Earth Sci Publ1222
Roberts JL (1970) The intrusion of magma into brittle rocks. In: Newall G, Rast N (eds) Mechanisms
of igneous intrusion. Gallery Press, London, pp 278 - 294
Robinson DJ, Hutchinson RW (1982) Evidence for a volcanogenic-exhalative origin of a massive
nickel sulfide deposit at Redstone, Timmins, Ontario. In: Hutchinson RW, Spence CD, Franklin
JM (eds) Precambrian sulfide deposits. Geol Assoc Can Spec Pap 25:211- 254
Ross JR, Hopkins GMF (1975) Kambalda nickel sulfide deposits: In: Knight CL (ed) Economic
geology of Australia and Papau New Guinea. I. Metals. Australas Inst Min Metall Monogr
5:100-121
Ross JR, Keays RR (1979) Precious metals in volcanic-type nickel sulfide deposits in Western
Australia. I. Relationships with the composition of the ores and host rocks. Can Miner
17:417-436
Ross JR, Travis GA (1981) The nickel sulfide deposits of Western Australia in global perspective.
Econ Geol 76:1291-1329
Schmulian ML (1984) Windarra nickel deposits, Western Australia. In: Buchanan DL, Jones MJ
(eds) Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall, London, pp 95 -102
Seccombe PK, Groves DI, Binns RA, Smith JW (1978) A sulfur isotopic study to test a genetic model
for Fe-Ni sulphide mineralization at Mt Windarra, Western Australia. In: Robinson BW (ed)
Stable isotopes in the earth sciences. N Z Dep Sci Ind Res, Wellingston, DSIR Bull 220: 187 - 200
Seccombe PK, Groves DI, Marston RJ, Barrett FM (1981) Sulfide paragenesis and sulfur mobility in
Fe - Ni - Cu sulfide ores at Lunnon and Juan Main shoots, Kambalda: Textural and sulfur
isotopic evidence. Econ Geol 76:1675 -1685
Sheppy NR, Rowe J (1975) Nepean nickel deposit. In: Knight CL (ed) Economic geology of Australia
and Papau New New Guinea. I. Metals. Australas Inst Min Metall Monogr 5:91 - 98
Shima H, Naldrett AJ (1975) Solubility of sulfur in an ultramafic melt and the relevance of the system
Fe - S - O. Econ Geol 70:960 - 967
Stolz GW (1981) A petrographic and geochemical investigation of the archean volcanic succession in
the vicinity of the Scotia nickel deposit. Ph D Thesis, Univ Adelaide (unpublished)
Sun SS (1981) Chemical composition and origin of the earth's primitive mantle. Geochim Cosmochim
Acta 46:179-192
Travis GA (1975) Mount Clifford nickel deposit. In: Knight CL (ed) Economic geology of Australia
and Papua-New Guinea. I. Metals. Australas Inst Min Metall Monogr 5:144-146
Viljoen MJ, Bernasconi A, VanColler N, Kinloch E, Viljoen RP (1976) The geology of the Shangani
nickel deposit, Rhodesia. Econ Geol 71 :76 - 95
Weaver BL, Tarney J (1979) Thermal aspects of komatiite generation and greenstone belt models.
Nature (London) 279:689 - 692
Wentlandt RF (1982) Sulfide saturation of basalt and andesite melts at high pressures and tempera-
tures. Am Miner 67:877 - 885
Williams DAC (1979) The association of some nickel sulfide deposits with komatiitic volcanism in
Rhodesia. Can Miner 17:337 - 349
Woolrich P, Cowden A, Giorgetta NE (1981) The chemical and mineralogical variations in the nickel
mineralization associated with the Kambalda dome. Econ Geol 76:1629 -1644
Depositional Environments of Volcanic Peridotite-
Associated Nickel Sulphide Deposits with Special
Reference to the Kambalda Dome
J. J. GRESHAM!

Abstract

Volcanic peridotite-associated Fe - Ni sulphide deposits in the Archaean Yilgarn

Block of Western Australia display a strong volcanological control on the distri-
bution of individual ore shoots. Although considerably modified by tectonism
and amphibolite facies metamorphism, there are distinctive stratigraphic rela-
tionships within the host ultramafic komatiite sequence that distinguish ore en-
vironments from adjacent areas barren of Fe - Ni sulphides.
At the well documented Kambalda Dome deposits, contact ores which occur
at the base of the lowermost flow of the ultramafic sequence are commonly con-
fined within narrow (100 - 300 m), elongate embayments or troughs in the under-
lying tholeiitic footwall metabasalt. Hangingwall ores occur at the base of im-
mediately overlying flows and generally directly overlie the troughs and contact
ores.
Compared to adjacent barren areas, ore environments are characterized by
thicker more-magnesian basal ultramafic flows, fewer interflow metasediments
and better textural development and compositional differentiation in overlying
thin flow units. The ultramafic sequence overlying ore environments is generally
less ordered than that in adjacent non-ore environments where there is a more
consistent trend of diminishing magnesium content upward through the ultra-
mafic sequence.
The major ore-confining troughs are interpreted as original elongate linear
depressions within the footwall metabasalt, and some of the local ore-confining
structures probably represent original topographic or volcanological features.
Ore deposition in an active rift environment may account for the close spatial
relationship of contact and hangingwall ores, and the development of the distinc-
tive stratigraphic features of the ore environment. Analogies are drawn with
oceanic rift systems and recent submarine volcanism.

1 Western Mining Corporation, Kambalda Nickel Operations, Kambalda, Western Australia, 6442,
Australia

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et a\.)
© Springer-Verlag Berlin Heidelberg 1986
64 J. J. Gresham

Introduction

Volcanic peridotite or komatiite-associated (VP A) nickel deposits occur at the

base of komatiitic extrusive sequences (Groves and Hudson 1981, Naldrett 1981).
These deposits are usually small (several thousand to 5 million tonnes) and high
grade (1- 50/0 Ni) and are known to occur in the Archaean mafic and ultramafic
sequences (Greenstone belts) of Canada (Green and Naldrett 1981), Zimbabwe
(Williams 1979) and Western Australia (Marston et al. 1981). However from the
identified metal resource (Ross and Travis 1981) 52% of the nickel metal occur-
ring in this class of deposit occurs in the Australian deposits with 28% and 20%
being contained in the Canadian and southern African deposits respectively.
In Western Australia all the major deposits of this type are confined to the
Norseman-Wiluna belt (Fig. 1) of the Yilgarn Block (Gee et al. 1980) with almost
90% of the pre-mining reserve occurring between Kalgoorlie and Norseman
(Marston et al. 1981). VPA deposits of Western Australia represent a much more
significant component of Australia's ~0.8% nickel resource and production
than similar deposits do in southern Africa and Canada (Ross and Travis 1981).
The Western Australian volcanic peridotite-associated deposits constitute close
to 50% of the identified nickel metal in ores ~0.8% nickel in Australia and to
date account for more than 90% of the nickel metal produced by Australia.
The greatest concentration of nickel sulphide mineralisation in Western
Australia occurs in the Kambalda nickel field (Woodall and Travis 1969, Ross
and Hopkins 1975, Gresham and Loftus-Hills 1981, Lesher 1983) where pre-
mined reserves (January 1983) totalled 33.9 million tonnes at 3.78% Ni. Produc-
tion from ten separate mines from this field to June 1983 since mining com-
menced from Lunnon shoot (Fig. 2) in March 1967 has totalled 16.6 million
tonnes at 3.13% Ni.
Almost all the nickel mineralisation at Kambalda occurs associated with the
stratigraphically lowest Kambalda sequence (Gresham and Loftus-Hills 1981)
and the most intensely mineralized area of the field is the Kambalda Dome
(Fig. 2). Extensive diamond drilling, mine development and geological research
in this area has afforded detailed documentation of the nickel ores and their asso-
ciated host rocks. The features of the environment of the Kambalda Dome will
be discussed in detail in this paper and compared to features of other volcanic
peridotite-associated deposits of the Norseman-Wiluna belt to evaluate the
depositional environment of this important class of deposit.

Geological Features of Western Australian Volcanic Peridotite-Associated

Nickel Sulphide Deposits

The Norseman-Wiluna belt (Fig. 1) comprises sequences of mafic and ultramafic

rocks, felsic volcanics and sediments separated by granitoid bodies. The green-
stone rocks form linear belts, and are commonly dissected by major NNW trend-
ing strike faults. Fold styles are dominated by NNW trending axial surfaces but
these are thought to post-date an earlier, possibly major phase of recumbent
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 65

LEGEND
PROTEROZOIC

GJ Mefle granulite, gnel ... granitoid

D Granitoid end gno ... lc rock.

Volcanic Peridotite -Associated

Fe - NI Deposits

KAMBAlDA DOME

FOSTER

3 JAN

EDWIN" LANFRANCHI

S SPARGDVILLE 3

6 MI. EDWARDS

7 WIDGIE TOWNSITE

8 WIDGIE 3

9 REDROSS

10 WANNAWAY
11 CAR NIL VA EAST

N 12 BLAIR

13 NEPEAN
........................
... .,.
. ...............
.. ... ............... .......... .
-+ . . . . . . . . . . . . . . . . . . . . . ..
14 SCOTIA

... T ...................... + 15 WINDARRA

r
............. + ....... ..- ...
...... +++ ................ +++
+ ...... +++++ ........ .
......... +'+"+ ............... ++
.. ......... ++++++ ........... .. o 50 100 km
I
• ... ... ...... ... ... ... .. '+" " .. " ... E.p~r ••nc:•• ~ ........ ..

8o lttr.. JfIo OC •• '"

Fig. 1. Geological plan of the Norseman-Wiluna Belt showing location of major volcanic peridotite-
associated deposits. (After Marston et al. 1981)
66 J. J. Gresham

vvvvvvv
vvvvvvv
vvvvvvv
vvvvvvv
vvvvvvv
vvvvvvvv
vvvvvvvv
vvvvvvvvv
vvvvvvvv
vvvvvvvvvv
vvvvvvvvvv
vvvvvvvvvv
vvvvvvvvvvv
VVVVVVVVVVV
vvvvvvvvvvvv
vvvvvvvvvvvvvv
vvvvvvvvvvvvvvvv
vvvvvvvvvvvvvvvvvv
v v v V \~' V V V V V V V V V V V V
VVV\ 'vvvvvvvvvvvvv
VVV \g
I VV V V V VV V V V V VVV
VV V - I VV VV vvvv v vvvvv
v
LEGEND v
v
vvv
o FelsIc - intermediate intrusives
vvvv
vvvvvv
VVVVVVVV
[I] Felsic volcanics and sed iments
VVVVVVVVV
vvvvvvvvvvv
G Hangingwall basalts

o Ultramafic rocks
VVVVVVVVVVVV
vvvvvvvvvvvvvv
VVVVVVVVVVVVVVV
EEl Footwall basalt VVVVVVVVVVVVVVVV
~VVVVVVVVVVVVV
o Sedimentary beds VVVVVVVVVVVVVVVVVV
VVVVVVVVVVVVVVVVVVV
vvvvvvvvvvvvvvvvvvv
VVVVVVVVVVVVVVVV e"VVV
Nickel ore shoots vvvvvvvvvvvvvv ~ vvv
Nickel are occurrences , VVVVVVVVVVV \ O. VVVV
,VVVVVVVVVV V JVVVVV
VVVVVVVVVVVVVVVV
Fault VVVVVVVVVVVVVVVV
Inferred fault

Fig. 2. Geological plan of Kambalda Dome showing major nickel shoots in plan projection, (After
Gresham and Loftus-Hills '1981)
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 67

folding and nappe development (Archibald et al. 1978, Platt et al. 1978). The vol-
canic peridotite-associated deposits within the belt dominantly occur in the strati-
graphically lowest ultramafic formations (Gemuts and Theron 1975, Gresham
and Loftus-Hills 1981), a stratigraphic control also documented for similar
deposits in Zimbabwe (Williams 1979) and Canada (Green and Naldrett 1981).
Although previous geochronological data had indicated the VP A deposits are
concentrated in the 2.8 Ga Archaean sequences (De Laeter et al. 1981, Gee et al.
1980), recent Sm - Nd data (Claoue-Long et al. 1984) on Kambalda samples have
indicated ages of 3.2 Ga. This date has yet to be verified.
Volcanic peridotite-associated deposits are invariably located at or near the
base of komatiitic volcanic sequences, and footwall rocks usually comprise
tholeiitic basalts, although at Windarra (Fig. 1) sulphide and oxide iron forma-
tions occur in the footwall (Schmulian 1982). Sequences of tholeiitic and high-
magnesium basalts and tholeiitic intrusives overlie the ultramafic rocks. Ore-
associated ultramafic sequences can vary in thickness from 40 m at Nepean
(Hudson 1973) to over 1000m at Kambalda and can commonly be divided into a
lower sequence of thicker (10- 100 m) high magnesian (36 - 45070 volatile-free
MgO) flows and an upper sequence of thinner «10m) lower magnesium
(20- 32070 MgO) flows. Thin, albite-rich sulphidic inter flow sediments occur
within the ultramafic sequence and at Kambalda are restricted to the lower part
of the sequence.
The nickel sulphide mineralisation generally occurs at, or close to the base of
the first high-magnesium ultramafic flow unit and where it occurs directly on the
underlying footwall rocks is referred to as contact ore. Other ores may occur
within the basal part of the ultramafic succession either at the base of im-
mediately overlying flows or as disseminated and blebby sulphides within the
host units. These are referred to as hangingwall ores which at Kambalda display a
close spatial relationship to contact ores. Structurally displaced contact or hang-
ingwall ores are called offset ores (Woodall and Travis 1969).
The individual ore bodies or shoots are narrow, elongate bodies which seldom
exceed 300 m in width, but may be over 3 km long. Contact ores invariably occur
within embayments or depressions of highly variable morphology within the
footwall rocks. These embayments vary from open spoon-like depressions, as at
Wannaway (McQueen 1981b), Scotia (Page and Schmulian 1981) and Windarra
(Schmulian 1982) to strongly re-entrant troughs as at Kambalda (Gresham and
Loftus-Hills 1981). Ore thickness is generally less than 5 m although several ore
shoots contain thicker low-grade (1- 2% Ni) disseminated ore zones. A typical
ore profile may comprise a thin irregular layer of massive ( > 80% sulphides) ore,
overlain by matrix (40 - 80% sulphides) ore which in turn may be overlain by dis-
seminated (10 - 40% sulphides) ore.
Volcanic peridotite-associated primary nickel ores are typically dominated by
pyrrhotite-pentlandite-pyrite assemblages with lesser chalcopyrite, magnetite and
ferrochromite. Minor occurrences of millerite-rich ores have been recorded
(Keele and Nickel 1974). Nickel grades of massive ore, depending on mineralogy,
can vary from 5 - 35% Ni, but generally are in the range 10-16% Ni. Ni/Cu
varies between 10-16 and NilCo between 40- 65 (Marston et al. 1981) and both
vary systematically with nickel grade. Platinum group element concentrations
68 J. J. Gresham

generally show a strong positive correlation with nickel grade with typical con-
centrations in 100070 sulphides being, platinum 1000- 2500 ppb, palladium
1500-3500 ppb, iridium 150-200 ppb, osmium 200-1000 ppb, ruthenium
800-1500 ppb and rhodium 250- 800 ppb (Ross and Keays 1979, Keays et al.
1981, Groves and Hudson 1981, Donaldson 1984a). S/Se values of 5000-15000
and 6 34s values from 0 to + 3.5% (Groves and Hudson 1981) for all deposits
except Windarra are typical of values for magmatic sulphides.
The nickel ores and their associated host rocks have been subjected to poly-
phase deformation and greenschist to upper amphibolite facies metamorphism
(Binns et al. 1976, Barrett et al. 1977). Metamorphism is extremely heterogene-
ous and nickel deposits occur in both low- and high-strain environments.
Although the metamorphism has resulted in almost complete destruction of
original igneous mineralogy, igneous terminology is retained throughout this
paper. During metamorphism all ores reverted at least in part to Fe - Ni mono-
sulphide solid solution (Barrett et al. 1977) and present ore distribution and
textures also reflect the ductility of the sulphides during deformation.

Geology and Characteristics of Ore Environments at the Kambalda Dome

The stratigraphy of the Kambalda Dome has been well documented (Woodall
and Travis 1969, Ross and Hopkins 1975, Gresham and Loftus-Hills 1981) and is
summarised in Fig. 3. The footwall rocks are a thick sequence ( > 2000 m) of low
to moderate magnesium basalts (Redman 1982) overlain by a variably thick
(200 ~ 1000 m) series of extrusive ultramafic flows characterised by thick
(10 -100 m) high-magnesium (>40% volatile-free MgO) flows, (lower member),
at the base of the sequence with thinner (generally < 10m) lower magnesium
(20- 32% volatile-free MgO) flow units forming the upper part, (upper
member), of the ultramafic sequence. The lower member generally constitutes
<25% of the total ultramafic sequence. There is a general trend of diminishing
Mg content up through the ultramafic sequence, but reversals of this trend do
occur. Thin (generally <5 m) sedimentary horizons occur at the footwall basalt-
ultramafic contact and at interflow positions within the lower member. These are
finely-laminated, albite-rich, sulphidic rocks and cherty, carbonaceous and
chloritic varieties occur (Bavinton 1979). The ultramafic rocks are overlain by a
complex sequence of basalts with possible major mafic intrusives. A sequence of
pillowed, strongly variolitic basalts 60-100 m thick immediately overlies the
ultramafic rocks. Separating these basalts from a thick sequence of predo-
minantly high magnesium basalts is a thin ( < 20 m) laterally persistent sulphidic,
graphitic sedimentary horizon. The basalts are overlain by a thick, poorly
defined sequence of sediments and felsic volcanics. Intruding the rocks of the
Kambalda Dome are Archaean sodic-granitoids, rhyolitic and dacitic porphyries,
minor mafic dykes and Proterozoic dolerite dykes.
Four deformational events involving fold development with associated fault-
ing occurred in the Kambalda Dome area (Gresham and Loftus-Hills 1981). The
early F 1 structures, characterised by tight to isoclinal, recumbent and napped
 Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 69

Fel.ic 'IOIconlc - sedlmentary Thick _ _ alintercolol.d

sequence .... 'menl ...... f ..... wlcanoes.

... "" and pIIowod ftow-..ora

high tnOQr'ItSAm boIah
5111CIOUI

o
o ..

HOI1gingwo II basalt
UPPER MEMBER
>~m L"'9dY ,hoek mo ..... ftows.

Sediment-Intrusive com pie.

up 10 <10m
o 0 .'. 0 ... 0 Hongingwoll basalt
r H<!"-Or. ••• 0 .\ 0 ••• 0 ••• Or. LOWER MEMBER
~~V,!~I . . . . Eltffrrot'llNnf 60 -100m

.. . . . . . . i;)!;~
UPPER MEMBER
Mu~ ,pleIhln p!ct;lie flow unilS.u.;I$ 'unorder _ aI plume and
ore poorly differentiattel with wlOk rJdotitic multiple ",In unitt. WI
tel,turoJ deve.lopmenl. IrOnQ composllionot differentotion
rod good ''''urol _I_~

Multlph 'nin peridolilic flow uni

Poorly differenTiated witt'! weak
ItJ:tural .... elopment.

Pouibll Inlcker flows. of bulk

perldotlU, compositIOn.
~RMEM,*R
'Thin oil.,. per idolilie flow unil,
"'h inter flow sediments. SedIment
'::"::".::'.::',.: .. ::'.::',:: curs on structurally undisturbRd
footwall bosah-ullromaflc coni
0000 ..
Fine to mtd'lUm~ointd tt'deiiti(: bosch
o o ~ P:/.~\~ With pillowed moel oncI flow q, tnccios.
Q 0000
o o
Footwall basalt
~2.000m
~ 0
00000 lO'lfII·rnoderole moQnnllDn bosoll
With pi ad lOinH and How lOp
brecclGs.

Fig. 3. Stratigraphic column of the Kambalda Dome area with most intrusives removed for clarity.
Features of ore and non-ore environments are portrayed and described. (After Gresham and Loftus-
Hills 1981)

folds have been refolded by the dominant folding event, F 2 , with north-north-
west trending axial surfaces. These folds have been refolded by F3 with east-
northeast axial surfaces and the interference of these two folding events is con-
sidered to be chiefly responsible for the formation of the Kambalda Dome. F 4
structures are poorly developed upright folds with north trending axial surfaces.
Detailed metamorphic studies (Barrett et al. 1977, Bavinton 1979) indicate
metamorphic temperatures slightly in excess of 500 °C and pressures around
2.5 Kb. All rocks have been subjected to varying degrees of hydration, carbona-
tion and potassium metasomatism. Original igneous mineralogy is seldom
70 1.1. Gresham

preserved although rare occurrences of olivine have been recorded at Victor and
Durkin shoots (Fig. 2). The Kambalda Dome is an area of generally low strain
and original igneous textures are commonly well preserved.
Despite the metamorphic, metasomatic and structural modification of the
ores and associated host rocks, distinctive structural, stratigraphic and litho-
logical features at the Kambalda Dome which characterise ore environments can
be recognised. These features are displayed within prisms of rock that are
elongate along the length of an ore shoot and extend from the immediate foot-
wall upward through the ultramafic sequence stratigraphically above an ore
shoot. These features have been discussed in detail by Gresham and Loftus-Hills
(1981), are shown schematically in Fig. 3, and can be clearly recognized in the
cross-sections in Fig. 5.
Nickel sulphide mineralisation is closely correlated with troughs or embay-
ments within the footwall basalt. Troughs seldom exceed 300 m in width and
relief of the trough may vary from a few to several tens of metres. In areas
remote from ore the basalt-ultramafic contact, although only defined by rela-
tively widely spaced diamond drill holes, appears to be a relatively undisturbed
and planar sediment-covered surface. The cutoff from ore-bearing to non-ore
bearing contact is abrupt and generally defined by an irregularity in the basalt-
ultramafic contact.
Basal flow units associated with ore generally have thicknesses of 50 - 100m
and are essentially confined to the ore environments and thus form elongate,
rather narrow zones coincident with the associated troughs. Thin (10-20m)
laterally persistent flow units occur away the ore zones. The relationship between
the ore-associated basal flows and the flanking barren flows is not well under-
stood, but in many instances they appear to be stratigraphically continuous. The
chemistry of the basal flows has been subject to detailed work by Woolrich and
Giorgetta (1978) and by Lesher (1983) and typically they contain >400/0 volatile-
free MgO <45% SiO z and 7 - 9% total iron. Lesher indicated considerable
olivine enrichment of both flow types although ore-associated units are enriched
in MgO and depleted in TiO z, Al z0 3, CrZ03 and total FeO compared to barren
flanking flows. Woolrich and Giorgetta also indicated that ore-associated flows
are depleted in Zn and enriched in Ni compared to barren flanking flows. There
are generally two and possibly more, thick high-magnesium flows above the
basal unit and, typically these units are thicker directly above ore zones. The total
thickness of the lower member may exceed 200 m associated with ore zones, but
seldom exceeds 100 m on the barren flanks. Most ore zones are surrounded by a
zone of sediment-free contact (Fig. 5), and characterised by an absence of inter-
flow sediments in the overlying sequence although there may be partial overlap
of sediments across the ore zone. Remote from ore there may be numerous inter-
flow sediment horizons, but this number diminishes toward the ore zones.
The upper member of the ultramafic sequence is characterised by thin lower
magnesium (20 - 32% MgO volatile-free) flow units with rare thicker olivine
peridotite units. Remote from ore, particularly on the west flank of the Kam-
balda Dome, the upper member displays a generally regular trend of diminishing
magnesium content upward through the pile. Associated with ore the sequence is
much less ordered, but individual flow units associated with ore zones generally
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 71

display better compositional differentiation and textural development than those

remote from ore.

Nickel Deposits of the Kambalda Dome

The Kambalda Dome (Fig. 2), covers and area of approximately 48 km 2 and
comprises the richest known concentration of high grade nickel sulphide ore in
Western Australia. A pre-mined reserve of 22.4 million tonnes at 4.13070 Ni for
the Kambalda Dome has been defined (January 1983), representing 72% of the
original nickel metal resource of the Kambalda nickel field. To June 1983 pro-
duction from eight separate mines had amounted to 15.7 million tonnes at 3.15%
Ni, 95% of the nickel metal produced from the field. The ore reserve occurs
within 15 separate ore bodies or shoots and these shoots contain more than 240
ore surfaces which are defined as geologically and geometrically distinct, essen-
tially continuous zones of ore-grade mineralisation.
During the early stages of exploration and development of these deposits it
was recognised that a full range or spectrum of deposit size, quality and nature
existed and generalisations about the nature of the deposits were difficult to
make. Bavinton (1979), related ore quality to the degree of sediment association
and Marston and Kay (1980) and Woolrich et al. (1981) discussed grouping the
deposits based on nickel tenor.
In this study of ore bodies have been broadly subdivided into three groups or
classes (Fig. 4) based on their ore characteristics and associated stratigraphic
features which are summarised in Table 1.

Class I

These ore bodies occur on the eastern and northern flank of the Kambalda
Dome: Juan, Otter, Durkin, Durkin North, Gibb, Long, and Victor. They
comprise nearly 70% of the pre-mined nickel metal ore reserve of the Kamb~lda
Dome, are high grade and 96% of their contained metal occurs as contact ores.
The ore bodies range in size from greater than 5 million tonnes to about 100000
tonnes. No stratiform hangingwall ore occurs associated with these ore bodies,
the only hangingwall ore occurrences being strata-bound blebby ores within the
basal host unit. Individual ore surfaces vary from very large elongate bodies
containing more than 1 million tonnes of high grade ore to small elliptical
surfaces (e.g., at Juan) that may only contain a few tens of thousands of tonnes.
Mineralogy of the ores is dominantly pyrrhotite-pentlandite-pyrite although
Gibb and Otter shoots contain millerite-pyrite-magnetite assemblages. The ore
tenor of the shoots is variable but Gibb, Durkin, Otter, and Victor are high tenor
ores (i.e., Ni > 18% in 100% sulphides), Long a low to moderate tenor ore shoot
(Ni = 14-16% in 100% sulphides) and Juan contains both high and low tenor
ore surfaces that appear, on reconstruction of the original environment (Marston
and Kay 1980), to form two sub-parallel belts.
72 J. J. Gresham

o 2km
~!------~------~!

8'"

Fig. 4 •• Plan of Kambalda Dome showing classification of nickel deposits and location of geological
cross-sections in Fig. 5
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 73

Table 1. Characteristics of ore Shoots of the Kambalda Dome

Feature Class I Class II Class III

Shoots Juan, Otter, Durkin, Lunnon Wroth, Gellatly,
Durkin North, Gibb, Gordon, McMahon,
Long, Victor Loreto, Ken, Fisher,
Hunt

0/0 of total Kambalda 59.4 19.1 21.5

Dome ore tonnes
% of total Kambalda 67.8 15.6 16.6
Dome Ni metal tonnes
Average Ni grade% 4.71 3.37 3.22
% tonnage as contact 92.6 50.4 56.4
ore
% tonnage as hang- 3.3 42.5 38.2
ingwall ore
% tonnage as offset 4.1 7.1 5.4
ore
070 Ni metal as contact 95.6 64.1 62.6
ore
% Ni metal as hang- 1.3 29.5 31.4
ingwall ore
070 Ni metal as offset 3.1 6.4 6.0
ore
Ore body and ore char- Almost all ore occurs Ore tonnage almost Contact ore surfaces
acteristics as contact ore with equally distributed can be moderate to
the minor hanging- between contact and high grade, but indi-
wall occurrences hangingwall ore and vidual ore surfaces
comprising blebby each occurs as exten- seldom exceed
ores within the high- sive, once essentially 250000 tonnes. Tenor
magnesium basal continuous, ore sur- of contact ores is
flow units. Juan, faces. Contact ores highly variable and
Durkin and Long are are low tenor, but may be higher or
the major shoots and substantially higher lower than associated
individual ore sur- grade than the high hangingwall ores.
faces within these ore tenor hangingwall Hangingwall ore may
bodies may contain 1 ores. The contact be associated with the
million tonnes of ores at the north end second or third flow
high-grade ore. Otter, of the ore body are units
Durkin, Victor, Gibb overlain by a thick
and parts of Juan zone of low-grade
comprise high tenor ( < 1070 Ni) dissem-
ores whereas Long inated sulphides. The
and the greater part intensity of contact
of Juan are moderate mineralisation
tenor (i.e. massive diminishes from
sulphides 10070 -16070 north to south. The
Ni) hangingwall ores di-
rectly overlie the
contact ore. Minor
hangingwall ore
occurs associated
with the third flow
unit
74 J. J. Gresham

Table 1 (continued)

Feature Class I Class II Class III

Shoots Juan, Otter, Durkin, Lunnon Wroth, Gellatly,
Durkin North, Gibb, Gordon, McMahon,
Long, Victor Loreto, Ken, Fisher,
Hunt

Ore-confining features Most contact ores Almost all contact Almost all contact
occur within well-de- ore occurs within the ores occur in troughs
fined troughs or em- main Lunnon trough and embayments of
bayments, many with although low grade variable morphology.
strongly re-entrant ore does occur on the Definition of struc-
structures. Morpho- west flank in places tural confinement is
logy is highly vari- and associated with generally not as well
able, but generally sediments on the east developed as in the
the trough floor is flank. The trough has Class I and II depos-
several metres below been strongly modi- its and mineralisation
the barren flanks fied by post-deposi- may occur as irregu-
tional faulting and lar concentrations
these structures ap- along a broadly de-
pear to terminate the fined structural em-
hangingwall ores bayment. Hanging-
wall ores spatially re-
lated to contact ores,
but association not as
clearly defined as
Class II deposits
Stratigraphic features Ore associated with Contact ore associat- Contact ores asso-
thick (50-100 m) ed with thick (30 - 80 ciated with thick
high-magnesium m) flow that thins to basal flows. Contact
basal flows with thin, the south. Basal flow sediments generally
fractionated, flow has thick, strongly absent from major
tops. No contact fractionated flow troughs, but some
sediments within nor- top. No contact sedi- sediment-associated
mal ore environment ments occur on the contact ore occurs at
and interflow sedi- flanks up to the ore- Ken shoot. Hanging-
ments generally confining structures. wall ores associated
absent above the Hangingwall ore as- with thick flows and
troughs. Strongly un- sociated with 30 - 50 commonly associated
ordered upper mem- m thick flow also with sediments
ber with individual with thick spinifex-
flow units displaying textured picritic flow
strong textural top. Hangingwall
development and ores commonly asso-
compositional dif- ciated at margins
ferentiation with interflow sedi-
ments

The ore bodies are associated with thick (50-100m) high-magnesium flow
units and there is a total lack of contact sediments within the ore environment.
Interflow sediments are generally absent between flow units above the ore-confin-
ing troughs although they occur on the flanks away from the ore environment
(Fig. 5A). The thin flow units of the upper member above these ore bodies gener-
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 75

ally display good textural development and strong compositional differentiation,

but the sequence of flow units is compositionally unordered.

Class II

Lunnon shoot on the southeast flank of the Kambalda Dome is the only ore body
in this class although there are extensive zones of mineralisation south of Lunnon
shoot underneath Lake Lefroy which are thought to be plunge extensions of this
ore body. This ore body is characterised by extensive fault disrupted contact and
hangingwall ore surfaces with the hangingwall ore directly overlying the contact
ore. Ore tonnage is almost equally divided between contact and hangingwall
surfaces although the higher grade of contact ores results in nearly 650/0 of the
nickel metal occurring as contact ore.
The contact ores are low tenor (massive ores 8% Ni) and may be overlain by a
thick (up to 30 m) zone of low-grade disseminated mineralisation (0.5 - 0.8% Ni)
within the basal host unit (Ross and Hopkins 1975). The thickness of this zone
and the basal host unit, and the intensity of mineralisation diminishes to the
south (Ross 1974, Middleton 1980). The majority of the hangingwall ore occurs
at the base of the second flow unit occurring on top of the thick spinifex-
textured picritic flow-top of the basal flow unit although limited zones of weak
mineralisation are recorded at the base of the third flow. Hangingwall ores are
high tenor (massive ores up to 25% Ni).
The ores are associated with thick flows, the first flow being 70 - 80 m thick
to the north and thinning to 30 - 40 m to the south. Flows thin rapidly to
10 -15 m on the flanks of the trough (Fig. 5 B). Contact sediments are totally
lacking from the main Lunnon trough although there is the well-documented ore-
sediment relationship on the east flank (Middleton 1980, Paterson et al. 1984).
Hangingwall ores commonly terminate against or are associated at their margins
with stratigraphically equivalent interflow sediments and occurrences of
nickeliferous sediments are common.

Class III

These are the ore bodies of the western flank of the Kambalda Dome: Wroth,
Gellatly, Gordon, McMahon, Ken, Loreto, Fisher, and Hunt. They are smaller,
less continuous, lower grade, ore positions that comprise both contact and hang-
ingwall ores. Tenor of ores within and between shoots is variable. For example,
Ken shoot is composed predominantly of high tenor contact ores. Within Fisher
shoot work by Hancock (1978) indicated subparallel belts of high and moderate
tenor contact ores and at Hunt shoot, contact ores are moderate to low tenor and
the associated hangingwall ore, low tenor. Hangingwall ores are generally
associated with the second flow unit although at Fisher there is extensive
hangingwall mineralisation associated with the third unit.
Contact ores are associated with thick, high-magnesium flow units and sedi-
ments are generally absent from the ore environment apart from the occurrences
76 1.1. Gresham
Co)
.....
UI
o
o AI
o
m

SEA LEVEL

...... ..
............ .
_
.........
.........
...... . .

.........
...... ....
........
........ .... .. . ..
..
........
..............
...............
.........
.............. .
.

SOOm B.S.L.

;
0 250m
I I
D Lake Sediments

8 81
0 Felsic - Intermediate Intrusives

g Hangingwall Basalt

•
vvv

• 0 Undifferentiated Ultramafic Rocks

0 Picrite

D Peridotite

[f3 Olivine Peridotite

122m A.~·L.
1, 1 Nickel Ore

~ Cherty Sediments

...
~ Footwall Basalt

/ ' Fault
0, 250m
I /' Fig. SA, B
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 77

c c'
~ ~

............ .......... ..
.............
............................. ..
........................
........................ . . ...
.............
..................
.................... .
.............. .
............... ..
.

o 250m
I I

Fig_ 5 A-C. Geological cross-sections of the Kambalda Dome. A A - A / 547925 N. Victor Shoot;
B B-B' 545913N. Lunnon Shoot; C C-C/ 544572N. Hunt Shoot

at Ken shoot (Paterson et al. 1984). Hangingwall ores are commonly associated
with interflow sediments and at McMahon and Fisher there are extensive zones
of nickeliferous sediments.

Reconstruction of tbe Geological Environment of tbe Kambalda Dome

Although the rocks and ores of the Kambalda Dome have been strongly
deformed and metamorphosed, from the foregoing analysis of the various ore
shoots and from a simple structural analysis a generalised reconstruction of the
original environment can be made. Reconstructions of individual shoots have
been made by Marston and Kay (1980) for Juan shoot, Ross (1974) and
Middleton (1980) for Lunnon shoot and Hancock (1978) for Fisher shoot.
The Durkin and Juan shoot areas display a high degree of structural com-
plexity with considerable horizontal shortening involving large-scale thrusting
and recumbent folding in the Juan area. These structures are thought to be
caused by strong NW-SE compressive forces that may in part be generated by the
intrusion of the granite body occupying part of the core of the Kambalda Dome.
Considerable movement has been accommodated on the Juan Fault, a major
structural feature of the area. Durkin shoot also displays strong development of
78 J. J. Gresham

low angle thrusting (Hayden 1976). Marston and Kay (1980) reconstructed the
Juan-Durkin area and defined two sub-parallel belts of ore of differing tenor
converging to the northwest. Carrying this reconstruction a stage further, it is
suggested that the high tenor Durkin shoot may have once been continuous with
the similar tenor ore bodies Gibb and Victor on the east flank of the dome. Long
shoot is considered to be a continuation of the low-moderate tenor surfaces of
Juan shoot. Although not validated by detailed structural analysis, these ore
shoots appear to have once formed essentially linear belts of semi-continuous
mineralisation at least 8 -10 km long. Terminations to the north and south have
not been defined and the linear extent of these ore "belts" could be much greater
than 10 km (Fig. 8).
The other ore shoots of the Kambalda Dome are extremely structurally
complex in detail, but again generalised reconstructions can be made. Lunnon
shoot is 2.5 km long and appears to be continuous with mineralisation a further

CLASS m CLASS n CLASS I

LEFROY FAULT

o 5km
Fig. 6. Reconstruction of ore distribu-
~I--------------------~I
SCALE (apprOllJ tion pattern at the Kambalda Dome
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 79

1.5 km to the south. Extensions of Lunnon shoot to the north have been lost
through erosion or structural dislocation.
Hunt, Fisher, Ken and McMahon shoots form an essentially linear ore belt, in
places comprising sub-parallel high and low tenor belts (e.g., Hancock 1978) that
extend for 8 -10 km and again terminations to this trend have not been recorded.
The convergence of Hunt and Lunnon shoot at the south end of the dome is a
structural feature.
Therefore a simplistic reconstruction of the Kambalda Dome indicates four
major sub-parallel linear belts of ore 600 -1200 m apart that are at least 10 km in
length (Fig. 6). The ore in these belts may occur as large continuous ribbons over
2 km long (i.e., Lunnon, Long) or in discrete pods within the overall linear trend.
However, ore zones seldom exceed 300 m in width. The nature and quality of ore
is broadly asymmetric with the eastern belt containing essentially all contact ore,
the central belt comprising closely-spatially related, and essentially strati-
graphically continuous contact and hangingwall ore and the western belt less con-
tinuous contact and hangingwall ores.
In addition to the ore-distribution asymmetry across the dome there is a
broad asymmetry in the distribution and characteristics of the ultramafic rocks.
The features of the ultramafic rocks that characterise an ore environment (see
Sect. 3) are common to all types of ore shoots and this "prism" of rock is con-
sistent along the trend of the ore belts. However, the ultramafic sequence thins
from east to west (sympathetic with the decreasing ore quality in this direction)
and, in general, the ultramafic rocks to the west display a better defined and
more consistent trend of diminishing magnesium content upward through the
sequence. Combining the cross-sections of Fig. 5 a generalised cross-section

Geological Featur. .
•
1 Decreasing ore quality
2 Thinning ultramafic "equence
3 Poorer definition of ·ore-envlronment·
4 Gre.. ter ore-"edlment association

~ ' ~::[I
~-' .}-. -'i5-' . -... .

~
~
~ ,(.~., . 't;:n;.

Fig. 7. Cross-sectional reconstruction of the Kambalda Dome. No intrusives or structural detail

shown for clarity. (See Fig. 5 for legend)
80 J . J. Gresham

...
0
8
g,.,

552 ~ ON

BASALT OUTCROP

o'"
ORE BEARING CONTACT

CJ
o BARREN BASALT-ULTRAMAFIC
BASALT-ULTRAMAFIC
- - UMIT OF DATA

44000 N

§'"
Ie

Fig. 8. Plan projection of basalt contact, Kambalda Dome, showing nature of basalt-ultramaiic
contact
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 81

through the Kambalda Dome is shown in Fig. 7. The structural and time relation-
ships between the ore belts are uncertain. It is possible that each belt formed
quite independently of the others (i.e., in separate fault-bounded basins). How-
ever, Fig. 7 has been drawn without any structural features and intrusives are
also omitted.
Sediments occur on the basalt-ultramafic contact throughout the area of the
Kambalda Dome. Detailed work by Bavinton (1979) indicated limited chemical
and mineralogical variation with the sediments in the Dome area, the most signif-
icant variation being the greater of carbonaceous sediments on the west flank of
the Dome. A plan of the Kambalda Dome with the ultramafic rocks removed is
shown in Fig. 8. Contact sediment distribution is largely taken from Gresham
(1978) and is derived from broad-spaced diamond drill information. This plan in-
dicates the sediment-free nature of the contact close to the ore zones.

Comparison with Other Volcanic Peridotite-Associated Deposits

Other volcanic peridotite-associated deposits of the Norseman-Wiluna belt also

display features that are observed from the deposits of the Kambalda Dome.
Major deposits (Fig. 1) include Nepean (Sanders 1982), Carnilya East, Blair,
Spargoville 3 (Andrews 1975), Wannaway (McQueen 1981b) and Redross,
Widgie Townsite and Widgie 3 (McQueen 1981a), Mt. Edwards (Turner 1983),
Windarra (Schmulian 1982), Scotia (Page and Schmulian 1981) and other
deposits at Kambalda including Jan, Foster, Lanfranchi and Edwin (Gresham
and Loftus-Hills 1981). Many smaller deposits occur throughout the belt and
these are referred to in review articles (Ross and Travis 1981, Marston et al.
1981). The African deposits Shangani (Viljoen et al. 1976) and Damba-Silwane
(Williams 1979), and Langmuir (Green and Naldrett 1981) in Canada also display
features common with some of those observed at the Kambalda Dome. These in-
clude:
1. Ores occur in depressions or embayments in underlying rocks.
2. Ores zones generally form narrow elongate bodies with width: length generally
>3: 1.
3. Ores are directly associated with thicker zones of high magnesium rocks.
4. Ores and sulphidic sediments are commonly mutually exclusive.
5. The ultramafic sequence is stratigraphically more complex near ore.
Although the deposits of the Norseman-Wiluna belt are highly variable and com-
plex and it is difficult to categorise them, the deposits do display similarities to
those of the Kambalda Dome. The Carnilya East, Blair, Nepean, Spargoville 3,
Redross, Edwin, Lanfranchi, and Widgie 3 deposits consist almost entirely of
contact ore. Although none of the deposits attains the size of the major Class I
deposits of the Kambalda Dome they are generally the higher grade deposits of
the belt, have no direct association with sediments and generally occur in well-
defined footwall embayments. Scotia and Windarra are also primarily contact
ore positions, but have features which distinguish them from the other deposits.
82 J. J. Gresham

Table 2. Classification of volcanic peridotite-associated deposits of the

Norseman-Wiluna Belt according to classification of deposits of the
Kambalda Dome

Class I Class II Class III

Carnilya East Foster Jan.

Blair Mt. Edwards Many other smaller
Nepean Wannaway deposits
SparqoviIIe 3 Widgie townsite
Redross
Edwin-Lanfranchi
Widgie 3
(Scotia)
(Windarra)

Foster, Wannaway, Widgie Townsite, and Mt. Edwards are ore bodies with
extensive contact and hangingwall ores and can be related to the Kambalda
Dome Class II deposit. Jan and many of the smaller deposits of Kambalda and
Widgiemooltha Domes are characterised by small contact ore positions and asso-
ciated hangingwall positions and are analogous to the Class III deposits of the
Kambalda Dome.
The difficulty in categorising deposits cannot be over emphasised and it is
probable that the deposits represent a spectrum of depositional environments.
Previous workers, Lesher et at. (1982), Lesher (1983), Lesher et at. (1984) and
Groves et at. (1984), have discussed the classification of volcanic peridotite-asso-
ciated deposits as proximal and distal facies variants.

Aspects of Genesis of Volcanic Peridotite-Associated Deposits

Considerable detailed geological research and documentation of this class of ore

deposit have indicated a wide variety of field, geochemical, experimental and
theoretical evidence for an original magmatic origin for the nickel sulphide ores.
Important evidence for this origin includes:
1. Association of ores with the most magnesian basal units of the komatiite
sequence; localisation of ore within footwall embayments or troughs; strong
volcanological control as evidenced by original stratigraphic relationships
(Ross and Hopkins 1975, Gresham and Loftus-Hills 1981, Lesher 1983).
2. Relationships of massive/matrix/disseminated sulphides at base of host unit
and interpillow sulphides within underlying basalts (Woodall and Travis 1969,
Ewers and Hudson 1972, Hudson 1972, Naldrett 1973, Gresham and Loftus-
Hills 1981).
3. Distinctive ore geochemistry compatible with known or experimentally in-
ferred partitioning behaviour in mafic-ultramafic systems. These include
NilCu ratios, precious metal tenors and abundances, S/Se ratios and J34g
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 83

values (Rajamani and Naldrett 1978, Duke and Naldrett 1978, Ross and Keays
1979, Keays et al. 1981, Seccombe et al. 1981, Donaldson 1984a).
4. The presence of distinctive chalcophile ferrochromites (Ewers et al. 1976,
Groves et al. 1977, Woolrich et al. 1981, Donaldson 1984b) in the ores.
5. Chalcophile element depletion in the associated host komatiites at Kambalda
(Lesher et al. 1981).
Although there is general agreement as to the origin of these ores through
magmatic processes, the actual nature and mechanism of the emplacement of the
sulphides and their hosts is still uncertain. Early workers (Ross 1974) considered
the basal flows to consist of a picritic liquid (23 - 240/0 MgO) containing varying
amounts of olivine phenocrysts. Recent detailed work on the nature of the
komatiitic host (Lesher et al. 1981, Arndt 1982, Lesher 1983, Donaldson and
Lesher 1982) has indicated liquids containing up to 32% MgO and in situ crystal-
lisation of much or all of the olivine in these flows has been proposed. Combin-
ing this work with theoretical modelling of flow characteristics, Lesher et al.
(1984) have postulated thermal erosion of the associated sulphidic sediments to
induce sulphide segregation. Furthermore Huppert et al. (1984) have postulated
that the flows could even thermally erode the underlying footwall basalts hence
generating their own trough during flow passage. These concepts are reviewed in
the light of critical field evidence and based on the observed field relationships a
model for the depositional environment of these deposits is discussed.
A sedimentary origin for the sulphur in the Western Australian nickel
deposits was first suggested by Prider (1970) and this concept has been revived
recently (Lesher 1983, Lesher et al. 1984, Huppert et al. 1984) largely because of
theoretical problems in explaining the capability of low viscosity, phenocryst-
poor lavas to transport dense sulphides, and the large volume of sulphides
relative to the volume of host ultramafic units. The broadly antithetic relation-
ship between ore and sediments makes the concept of sediment assimilation
attractive, but there are many detailed stratigraphic and geochemical relation-
ships at the Kambalda Dome which are inconsistent with this concept.
1. The broad asymmetry of the quality of ore distribution, the variable tenor of
sub-parallel ore shoots and of contact and hangingwall ores of the same
shoots argue strongly against assimilation of sediments of relatively uniform
composition by distal-sourced lava flows.
2. Almost all contact ore occurs in well-defined embayments and there is
negligible ore outside these zones. The emplacement conditions and tempera-
tures of ore-associated lava flows were likely to be variable and a range of
gradational relationships between ore-bearing and barren contacts would be
anticipated. The sharp definition of the ore zones is contrary to the expected
relationships if sediment assimilation was important.
Although the contact ores generally are surrounded by a zone of sediment-
free basalt-ultramafic contact (Fig. 8), there are considerable areas of
sediment-free contact where there are no defined nickel sulphides. If the
sediment-assimilation model assumes that all the basalt was originally covered
with sediment, then there should be a more universal distribution of nickel ore
than is observed.
84 J. J. Gresham

Contact sediments are neither thicker nor more sulphur-rich on the east
and north flanks of the dome to account for the greater concentration of sul-
phides in the Class I deposits.
3. Hangingwall ores are invariably lower grade and sulphide abundance is less
than associated contact ores (Table 1). However, sulphur content of internal
sediments is higher (7.3070) than that of contact sediments (4.0%) (Bavinton
1981) and thicknesses of internal and contact sediments are comparable.
4. At some shoots of the Class I type deposits (e.g., Victor, Fig. 5 A) there are
2 - 3 flow units above the ore zone with no associated interflow sediments. On
the flanks of these shoots there are essentially continuous inter flow sediments.
The complete absence of stratiform hangingwall ores associated with this class
of deposit and the absence of interflow sediments directly above the contact
ores argue strongly against the concept of sediment assimilation.
5. At some volcanic peridotite-associated deposits, e.g., Mt. Edwards and Wan-
naway, there are extensive zones of massive sulphides associated with
sulphidic sediments with no apparent erosion or attenuation of the sediment
in the ore environment.
6. If all ore-bearing and barren contact shown in Fig. 8 was assumed to be sedi-
ment covered and subsequently assimilated by the ore-associated flow, there
would not be nearly enough sulphur in these sediments (4%) to generate the
amount of nickel sulphides observed. A simple calculation assuming a
sediment averaging 4 m in thickness and containing 4% sulphur would
indicate 28 km2 of contact sediment would have to be eroded to generate the
sulphides observed in the Class I deposits. In actual fact average contact
sediment thickness in the Juan area is approximately 1 m (Bavinton 1979).
7. In those shoots that have intimate massive ore-sediment relationships
(Paterson et al. 1984) and where some local assimilation of sediment is prob-
able, the ores have significantly higher zinc content (160 ppm) than normal
massive ores (40 ppm). This geochemical difference is also reflected at a shoot
scale where ores from shoots of the Class II and III deposits generally have
higher Zn and As content than those of the Class I deposits.
Although there may be some local assimilation of sediment, the detailed evi-
dence suggests this is an unlikely factor in the generation of the nickel ores and the
broadly antithetic ore-sediment relationships must be explained in another way.
Turning now to the origin of the troughs, Huppert et al. (1984) have
postulated that the high temperature komatiitic flows could thermally erode the
underlying basalt and create their own trough. This largely theoretical work con-
flicts with some of the detailed observations around the Kambalda Dome.
1. The very strong arguments developed against sediment assimilation.
2. Stratigraphic control of troughs as indicated by Harley (1980) and Lesher
(1983). In some areas stratigraphic continuity has been defined between
embayment contact (i.e., the one zone) and a unit boundary (e.g., flow-top
breccia) in the adjacent basalt. The depth of most embayments is comparable
to the thickness of most basalt flow units.
3. The many small elliptical depressions containing nickel ores (i.e., Ken, Juan
areas) that are considered primary topographic volcanic features. It is difficult
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 85

to conceive of these small scale features being generated through erosional

processes.
4. The strongly linear embayments at Kambalda contrast strongly with the
highly sinuous lunar rilles (e.g., Hadley rille) thought to have been generated
by thermal erosion (Hulme 1973, 1982, Wilson and Head 1983).
5. Complete back of geochemical contamination of basal flows within the
trough structures. These have Ti0 2 and Al 20 3 values lower than or equal to
the flanking flows (Lesher 1983).

Interpretation of Depositional Environment

The Norseman-Wiluna belt is characterised by thick sequences of volcanic and

sedimentary rocks. Accumulation of these sequences is considered to have been
rapid (Archibald et al. 1978) and deposition to have taken place in a rapidly sub-
siding, strongly rifting environment (Groves 1982) with the formation of many
discrete, fault-bounded elongate basins of deposition. Stratigraphic correlation
between different mafic-ultramafic sequences is uncertain although it is generally
thought that there is some stratigraphic control on the mineralisation, with the
volcanic peridotite-associated deposits occurring in the lower-most sequences
(Gemuts and Theron 1975, Gresham and Loftus-Hills 1981).
Ulramafic sequences hosting volcanic peridotite deposits have in almost all
cases been extruded onto tholeiitic basaltic sequences. At the Kambalda Dome
these basalts comprise medium to thick flows with, in places, well-developed
flow-top breccias, possible intrusive bodies, and abundant development of
pillowed sequences. The general lack of interflow sediments or wackes apart
from the discontinuous marker horizon 1oo-200m below the basalt-ultramafic
contact indicates rapid accumulation of these basalts which would have resulted
in the development of highly irregular topography on the ocean floor. The
almost complete lack of vesicles in the pillows of Kambalda indicates water
depths possibly exceeding, 5000 m (Jones 1969, Moore 1975) which would
indicate extrusion under high confining pressure. Rarely preserved pillows at
Carnilya East also lack vesicles whereas at Widgiemooltha there is development
of vesicles indicating deposition in shallower water.
The embayments confining the nickel ores are highly variable and obviously
no generalised statement can be made regarding their origin. Lesher (1983) has
argued that the embayments were largely topographic-volcanological features.
The marked continuity and linearity of these features at the Kambalda Dome
suggest a tectonic as well as a volcanogenic-topographic control. Abundant small
scale faulting is observed in modern ocean ridge environments (Ballard and
Moore 1977) and there is evidence at Kambalda for very early faulting (Richards
1983) and specifically at Lunnon Shoot for synvolcanic fault development (Ross
1974, Middleton 1980). Lack of suitable stratigraphic control elsewhere at
Kambalda precludes the definition of the universal occurrence of these struc-
tures. It is suggested that the broad distribution of the embayments and ore
shoots was controlled by the general extensional-rifting tectonics of the area. The
86 J. J. Gresham

Lefroy Fault, although obviously reactivated during later deformational events

may have been a major ore-controlling structure in the area. The troughs have
dimensions similar to modern oceanic rifts (Spiess et al. 1980, Francheteau and
Ballard 1983, Vinogradov and Udinstsev 1975). Along the trend of these struc-
tures a variety of topographic-volcanological embayments would occur.
The asymmetry of the Kambalda environment and ore quality suggests
diminishing thermal activity from east to west and it is significant that the major
Class I deposits are closest to the Lefroy fault. This relationship of nickel miner-
alisation to major north-northwest faults has been recorded by Gemuts and
Theron (1975) and other workers.
Continued basaltic volcanism and thermal activity along linear zones, now
occupied by the ores zones, inhibited sediment accumulation, and eruption of the
ultramafic lava and sulphides took place from thin fissures at intervals along
these structures. Fissures feeding these eruptions need to be no more than 1 m
wide (Wilson and Head 1983) and evidence from Lunnon shoot (Middleton 1980,
Lesher 1983) with the thinning basal flow unit and diminishing ore quality
indicates individual extrusion sites may be spaced at intervals of 2 - 4 km along
the linear trends of the ore belts. The lack of observed feeders in the footwall
basalts at Kambalda is not a problem if the fissures were extremely narrow as
they may be now unrecognisable or obliterated by the extensive felsic intrusives
in the area.
A detailed discussion of the nature of the basal ultramafic flow units is
beyond the scope of this paper, but the understanding of the magmatic, physical,
mineralogical and chemical characteristics of these flows is critical to the under-
standing of the origin of volcanic peridotite-associated nickel deposits. Stolz and
Nesbitt (1981) at Scotia and Lesher (1983) have documented clear evidence for in
situ growth of olivine in flows associated with these deposits. This has led to the
development of the concept that these flows originated from almost totally liquid
magma of 28-321170 MgO and the present magnesium rich content (40-45%
volatile free MgO) was achieved by a process of lava flow-through. The presence
of branching crescumulate olivine is evidence of in situ olivine crystallisation, but
this type of olivine may in fact represent only a minor component of the basal
flows. Flow-through and olivine enrichment as discussed by Lesher (1983)
requires the development of lateral facies variants of lower magnesium basal
rocks, examples of which are not observed over the 60 km of explored basalt-
ultramafic contact of the Kambalda field. The thin, laterally persistent, flanking
flows appear to be lateral equivalents of the ore flows, but they are only
marginally different in bulk chemical composition. In order to explain the slight
chemical differences between ore and non-ore flows, it is suggested that basal
flows contained appreciable quantities of intratelluric olivine and some riffling
of these olivine phenocrysts did occur as suggested by Naldrett and Campbell
(1982). The lateral persistence of the thin flanking flows is indicative of low
viscosity lavas and chemical profiles indicate that in situ fractionation occurred
in both ore-associated and flanking flows.
The presence of intratelluric olivine would also enhance the sulphide carrying
capacity of the flows. As indicated by Lesher et al. (1981), the ultramafic succes-
sion at Kambalda was in equilibrium with sulphide liquid at the magmatic stage.
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 87

Field evidence such as interpillow sulphides, sulphides in basalt flow-top breccias

and the occurrence of sulphide accumulations in topographic lows (i.e.,
ellipsoidal structures at Juan and Ken) and the along-trough ore relationships at
Lunnon are indicative of very early and rapid segregation of the sulphides.
Marston and Kay (1980) suggested that this took place during the early stages of
horizontal transport although segregation during vertical transport (Ross 1974,
Ross and Hopkins 1975) cannot be discounted.
The development of the well-defined ore "prism" (i.e., thicker high-mag-
nesium successive flows, spatially related hangingwall ores, lack of sediment
development and textural-compositional development within flows) is strongly
suggestive of continued reactivation and extrusion along the trend of the ore
shoots. The linearity and consistency of superposition of these features is dif-
ficult to reconcile with a single source, distal model.
Deposition of the basal high-magnesium flow units with inter flow sediment
accumulation was followed by continuous outpourings of thin (generally < 10m)
flows of low to moderate magnesium lavas and in turn by thick basaltic, felsic-
volcanic and sedimentary sequences.
The ore bodies of Kambalda Dome and other volcanic peridotite-associated
deposits display many enigmatic and variable geological features that are dif-
ficult to explain by any generalised genetic model. However, the broad strati-
graphic and ore relationships at the Kambalda Dome are consistent with a series
of sub-parallel extrusion sites that were periodically reactivated.
Sediment assimilation by ultramafic flows was not a major factor in the
generation of the nickel ores and thermal erosion of the underlying basalt to
generate the ore-confining embayments is not compatible with field relation-
ships.

Acknowledgements. Much of the basic geological information contained in this paper is the result of
geological work carried out by company mine and exploration geologists over the past 18 yrs. I would
like to thank many of these for stimulating discussions and useful comments. I would particularly like
to thank B. Thomson for his help in preparing cross-section 544572N and J. Ross for helpful discus-
sions and suggestions. I would also like to thank N. Archibald, A. Cowden, M. Elias, D. Groves,
G. Loftus-Hills, K. Cross, and R. Smith who provided constructive criticism of the manuscript.
Thanks are also extended to D. R. Hudson and an anonymous reviewer for their helpful comments.
I am indebted to Debra Woodman and Christine Wood for typing the manuscript and R. Perkin
and H. Bush for the preparation of the figures. This paper is published by permission of Western
Mining Corporation Limited.

References

Andrews PB (1975) Spargoville nickel deposits. In: Knight CL (ed) Economic geology of Australia
and Papua, New Guinea. I. Metals. Australas Inst Min Metall Mon 5:89 - 91
Archibald NJ, Bettenay LF, Binns RA, Groves 01, Gunthorpe RJ (1978) The evolution of Archean
greenstone terrains, Eastern Goldfields province, Western Australia. Precambrian Res 6:103 -131
Arndt NT (1982) Post-eruptive olivine and sulphide fractionation and accumulation in komatiite lava
flows (abs). IGCP Proj 161,91: Nickel Sulphide Field Conf III Abs Vol, p 22
Ballard RD, Moore JG (1977) Photographic atlas of the Mid-Atlantic ridge rift valley. Springer,
Berlin Heidelberg New York, 114 pp
88 J. J. Gresham

Barrett FM, Binns RA, Groves DI, Marston RJ, McQueen KJ (1977) Structural history and meta-
morphic modification of Archean volcanic-type nickel deposits, Yilgarn Block, Western Australia.
Econ Geol 72:1195 -1223
Bavinton OA (1979) Interflow sedimentary rocks from the Kambalda ultramafic sequence: Their
geochemistry, metamorphism and genesis. Ph D Thesis, Aust Nat! Univ, Canberra, 196 pp
(unpublished)
Bavinton OA (1981) The nature of sulfidic metasediments at Kambalda and their broad relationships
with associated ultramafic rocks and nickel ores. Econ Geol 76: 1606 - 1628
Binns RA, Gunthorpe RJ, Groves DI (1976) Metamorphic patterns and development of greenstone
belts in the eastern Yilgarn Block, Western Australia. In: Windley BF (ed) Correlation of the
Precambrian, vol II. Nauka, Moscow, pp 349-380
Claoue-Long JC, Thirlwell MF, Nesbitt RW (1984) Sm - Nd of Kambalda greenstones - revisited
and revised. Nature (London) 307, 5953:697 -701
Donaldson MJ (1984a) Geochemistry and mineralogy of the platinum group elements in Kambalda
nickel ores and concentrator products. WMC (KNO) Rep K/2781, 33 pp (unpublished)
Donaldson MJ (1984b) Progressive alteration of barren and weakly mineralised Archaean dunites
from Western Australia: A petrological, mineralogical and geochemical study of some komatiitic
dunites from the Eastern Goldfields Province. Ph D Thesis, Univ W Aust (unpublished)
Donaldson MJ, Lesher CM (1982) Intrusive dunite and extrusive peridotite: Analogous products of
komatiitic magmas resulting from different cooling rates (abs). IGCP Proj 91, 161: Nickel
Sulphide Field Conf III Abs Vol, pp 31 - 32
Duke JM, Naldrett AJ (1978) A numerical model of the fractionation of olivine and molten sulfide
from komatiite magma. Earth Planet Sci Lett 39:255 - 266
Ewers WE, Hudson DR (1972) An interpretative study of a nickel-iron sulfide ore intersection,
Lunnon Shoot, Kambalda, Western Australia. Econ Geol 67:1075 -1092
Ewers WE, Graham J, Hudson DR, Rolls JM (1976) Crystallization of chromite from nickel-iron
sulphide melts. Contrib Mineral Petrol 54:61 - 64
Francheteau J, Ballard RD (1983) The East Pacific rise near 21°N; 13°N, and 20 0 S: Inferences for
along strike variability of axial processes of the mid-ocean ridge. Earth Planet Sci Lett 64:93 -116
Gee RD, Williams IR, Wilde SA, Baxter JL (1980) Crustal development of the Yilgarn Block (abs).
2nd. Int Archaean Symp, Perth, Extended Abs, pp 79 - 80
Gemuts I, Theron A (1975) The Archaean between Coolgardie and Norseman-stratigraphy and
mineralization. In: Knight CL (ed) Economic geology of Australia and Papua New Guinea. I.
Metals. Australas Inst Min Metall, Melbourne, Mon 5, pp 66-74
Green AH, Naldrett AJ (1981) The Langmuir volcanic peridotite-associated nickel deposits:
Canadian equivalents of the Western Australian occurrences. Econ Geol 76:1503 -1523
Gresham 11 (1978) The Kambalda dome traverse study final report. WMC (KNO) Rep K12389, 32 pp
(unpublished)
Gresham 11, Loftus-Hills GD (1981) The geology of the Kambalda nickel field, Western Australia.
Econ Geol 76:1373 -1416
Groves DI (1982) The Archaean and earliest Proterozoic evolution and metallogeny of Australia. Rev
Bras Geocie 12(1- 3):135 -148
Groves DI, Hudson DR (1981) The nature and origin of Archaean stratabound volcanic-associated
nickel-iron copper sulphide deposits. In: Wolf KH (ed) Handbook of stratabound and stratiform
ore deposits, vol IX. Elsevier/North Holland Biomedical Press, Amsterdam New York, pp
305-410
Groves 01, Barrett FM, Binns RA, McQueen KG (1977) Spinel phases associated with meta-
morphosed volcanic-type iron-nickel sulfide ores from Western Australia. Econ Geol
72:1224-1244
Groves 01, Lesher CM, Gee RD (1984) Tectonic setting of the sulphide nickel deposits of the Western
Australian Shield. In: Buchanan DL, Jones MJ (eds) Sulphide deposits in mafic and ultramafic
rocks. Inst Min Metall, London, pp 1 - 13
Hancock SL (1978) K.N.O. Progress report - Fisher shoot. Rep K2385, 22 pp
Harley DN (1980) K.N.O. Progress report - Juan shoot. WMC (KNO) Rep K/2497 , 24 pp
(unpublished)
Hayden P (1976) Fault and contact movements at Durkin shoot and their effects on ore. WMC
(KNO) Rep K12261, 12 pp (unpublished)
Depositional Environments of Volcanic Peridotite-Associated Nickel Sulphide Deposits 89

Hudson DR (1972) Evaluation of genetic models for Australian sulphide nickel deposits. In: New-
castle Conf, 1972, Papers. Australas Inst Min Metall, Melbourne, pp 59-72
Hudson DR (1973) Genesis of Archaean ultramafic associated nickel-iron sulphides at Nepean,
Western Australia. Australas Inst Min Metall, W Aust Conf Perth, Pap, pp 99 -100
Hulme G (1973) Turbulent lava fow and the formation of lunar sinuous rilles. Mod Geo14:107 -117
Hulme G (1982) A review of lava flow processes related to the formation of lunar sinuous rilles.
Geophys Surv 5:243 - 279
Huppert HF, Sparks RSJ, Turner SJ, Arndt NT (1984) The emplacement and cooling of komatiite
lavas. Nature (London) 309:19-22
Jones JG (1969) Pillow lavas as depth indicators. Am J Sci 267:181-195
Keays RR, Ross JR, Woolrich P (1981) Precious metals in volcanic peridotite-associated nickel
sulphide deposits in Western Australia. II. Distribution within the ores and host rocks in
Kambalda. Econ Geol 76:1645 -1674
Keele RA, Nickel EH (1974) The geology of a primary millerite-bearing sulfide assemblage and super-
gene alteration at the Otter shoot, Kambalda, Western Australia. Econ GeoI69:1102-1117
Laeter JR, De, Libby WG, Trendall AF (1981) The older Precambrian geochronology of Western
Australia. In: Glover JE, Groves 01 (eds) Archaean geology. Geol Soc Aust Spec Pub17:145 -157
Lesher CM (1983) Localization and genesis of komatiite-associated Fe - Ni - Cu sulphide mineraliza-
tion at Kambalda, Western Australia. Ph D Thesis, Univ W Aust, 318 pp (unpublished)
Lesher CM, Lee RF, Groves 01, Bickle MJ, Donaldson MJ (1981) Geochemistry of komatiites from
Kambalda, Western Australia. I. Chalcophile element depletion - a consequence of sulfide liquid
separation from komatiitic magmas. Econ Geol 76:1714-1728
Lesher CM, Donaldson MJ, Groves 01 (1982) Nickel deposits and their host rocks in the Norseman-
Wiluna Belt. In: Groves DI, Lesher CM (eds) Regional geology and nickel deposits of the Norse-
man-Wiluna Belt, Western Australia. Univ West Aust, Dep Geol Ext Serv, Publ 7:B1- B63
Lesher CM, Arndt NT, Groves 01 (1984) Genesis of komatiite-associated nickel sulphide deposits at
Kambalda, Western Australia: A distal volcanic model. In: Buchanan DL, Jones MJ (eds)
Sulphide deposits in mafic and ultramafic rocks. Inst Min Metall, London, pp 70 - 80
McQueen KG (1981a) Volcanic-associated nickel deposits from around the Widgiemooltha Dome,
Western Australia. Econ Geol 76:1417 -1443
McQueen KG (1981b) The nature and metamorphic history of the Wannaway Nickel Deposit,
Western Australia. Econ Geol 76:1444-1468
Marston RJ, Kay BD (1980) The distribution, petrology and genesis of nickel ores at the Juan
complex, Kambalda, Western Australia. Econ GeoI75:546-565
Marston RJ, Groves 01, Hudson DR, Ross JR (1981) Nickel sulphide deposits in Western Australia:
A review. Econ GeoI76:1330-1363
Middleton CN (1980) K.N.O. Progress report - Lunnon shoot. WMC (KNO) Rep K/2517, 40 pp
(unpublished)
Moore JG (1975) Mechanism of formation of pillow lava. Am Sci 63, 3:269-277
Naldrett AJ (1973) Nickel sulphide deposits - their classification and genesis, with special emphasis
on deposits of volcanic association. Can Inst Min Metall Bull 66, 739:45 - 63
Naldrett AJ (1981) Nickel sulphide deposits: Classification, composition, and genesis. Econ Geol
75th Anniversary Vol, pp 628-685
Naldrett AJ, Campbell IH (1982) Physical and chemical constraints on genetic models for komatiite-
related Ni-sulphide deposits. In: Arndt NT, Nisbett EG (eds) Komatiites. Allen & Unwin, London,
pp 423-434
Page ML, Schmulian ML (1981) The proximal volcanic environment of the Scotia nickel deposits.
Econ Geol 76:1469 -1479
Paterson HL, Donaldson MJ, Smith RN, Lenard MF, Gresham JJ, Boyack DJ, Keays RR (1984)
Nickeliferous sediments and sediment-associated nickel ores at Kambalda, Western Australia.
Buchanan DL, Jones MJ (eds) Inst Min Metall, London, pp 81- 94
Platt JP, Allchurch PD, Rutland RWR (1978) Archaean tectonics in the Agnew supracrustal belt,
Western Australia. Precambrian Res 7:3 - 30
Prider RT (1970) Nickel in Western Australia. Nature (London) 226:691- 693
Rajamani V, Naldrett AJ (1978) Partitioning of Fe, Co, Ni, and Cu between sulfide liquid and
basaltic melts and the composition of Ni - Cu sulfide deposits. Econ Geol 73:82 - 93
90 J. J. Gresham: Depositional Environments

Redman BA (1982) Petrography, petrology and geochemistry of Archaean basic volcanism in the
Eastern Goldfields Province, Western Australia - stratigraphic control of gold mineralisation. M
Sc Thesis, Univ Melbourne, 304 pp (unpublished)
Richards T (1983) Structural geology of the Hunt shoot environment. WMC Rep (unpublished)
Ross JR (1974) Archean nickel sulphide mineralisation, Lunnon shoot, Kambalda. Ph D Thesis, Univ
Calif, Berkeley, 283 pp (unpublished)
Ross JR, Hopkins GMF (1975) Kambalda nickel sulphide deposits. In: Knight CL (ed) Economic
geology of Australia and Papua New Guinea. I. Metals. Australas Inst Min Metall, Melbourne,
Mon 5:100-121
Ross JR, Keays RR (1979) Precious metals in volcanic-type nickel sulfide deposits in Western
Australia, I; Relationship with the composition of the ores and their host rocks. Can Miner
17:417-435
Ross JR, Travis GA (1981) The nickel sulfide deposits of Western Australia in global perspective.
Econ Geol 76:1291-1329
Sanders TS (1982) Nepean nickel deposits. In: Groves 01, Lesher CM (eds) Regional geology and
nickel deposits of the Norseman-Wiluna Belt, Western Australia. Univ West Geol Aust Dep Ext
Serv Publ 7:C35 - C44
Schmulian ML (1982) Windarra nickel deposits. In: Groves 01, Lesher CM (eds) Regional geology
and nickel deposits of the Norseman-Wiluna Belt, Western Australia. Univ West Aust Dep Geol
Ext Serv Publ 7:C67 - C76
Seccombe PK, Groves 01, Marston RJ, Barrett FM (1981) Sulfide paragenesis and sulfur mobility in
Fe - Ni - Cu sulfide ores at Lunnon and Juan Main shoots Kambalda: Textural and sulfur isotopic
evidence. Econ Geol 76:1675 -1685
Spiess FN, McDonald KC, Atwater T, Ballard R, Carranza A, Cordoba D, Cox D, Diaz Garcia VM,
Francheteau J, Guerrero J, Hawkins J, Haymon R, Hessler R, Juteau T, Kastner M, Larson R,
Luyendyk B, MacDougall JD, Miller S, Normark W, Orcutt J, Rangin C (1980) East Pacific Rise;
Hot springs and geophysical experiments. Science 207,4438:1421-1433
Stolz GW, Nesbitt RW (1981) The komatiite nickel sulfide association at Scotia: A petrochemical
investigation of the ore environment. Econ Geol 76:1480-1502
Turner MStJ (1983) The geology and mining of the Mt. Edwards nickel orebody, Widgiemooltha.
WMC (KNO) Rep K12732, 22 pp (unpublished)
Viljoen MJ, Bernasconi A, Coller N, Van, Kinloch E, Viljoen RP (1976) The geology of the Shangani
nickel deposit, Rhodesia. Econ Geol 71 :76 - 95
Vinogradov AP, Udinstsev GB (1975) Rift zones of the world oceans. John Wiley & Sons, New York,
503 pp
Williams DAC (1979) The association of some nickel sulfide deposits with komatiitic volcanism in
Rhodesia. Can Miner 17:337 - 350
Wilson L, Head JW (1983) A comparison of volcanic eruption processes on Earth, Moon, Mars, 10
and Venus. Nature (London) 302:663 - 669
Woodall R, Travis GA (1969) The Kambalda nickel deposits, Western Australia. Proc 9th Commonw
Min Metall Congr, London (1969) 2:517 - 533
Woolrich P, Giorgetta N (1978) A study of the geochemistry and mineralogy of the basal ultramafic
unit at Kambalda. WMC (KNO) Rep K12380, 21 pp (unpublished)
Woolrich P, Cowden A, Giorgetta N (1981) The chemical and mineralogical variations in the nickel
mineralisation associated with the Kambalda Dome, Western Australia. Econ Geol 76:1629 -1644
Geochemistry of the Sudbury Igneous Complex:
A Model for the Complex and Its Ores
A.J. NALDREITI

Abstract

A recent interpretation of regional gravity and magnetic data (Gupta et al. 1984)
has indicated that the Sudbury Igneous Complex is underlain at depths of
5 - 8 km by a 60 x 40 km mass of mafic and ultramafic rock that is not part of
the exposed Complex.
The marginal rocks of the Complex are thought to have crystallized in situ.
The high Si0 2 and K20 and low CaO contents, and low Na20/K20 ratio of these
rocks in comparison with those of continental flood basalts are suggestive that
the magma responsible for the Complex experienced extensive contamination by
felsic country rocks. The REE profiles and major elements can be modelled if a
1 : 2 mixture of quartz monzonite and tonalite that form much of the basement at
Sudbury, is combined on a 1: 1 basis with a fairly primitive flood basalt
(MgNo = 0.61). The sub layer is more fractionated than the marginal unit of the
main mass, but its major element and REE concentrations also suggest signif-
icant assimilation of a similar contaminant. The high Sr initial isotope ratios of
both main mass and sublayer are consistent with the contamination hypothesis.
Mafic and ultramafic inclusions are restricted to variants of the sub layer that
are also mineralized. These inclusions have REE profiles with similar high La/Yb
ratios to those of the main mass and sublayer. The Fo content of olivines in the
inclusions indicates that they crystallized from liquids spanning the same range
of Mg Nos as is spanned by samples of the sub layer . It is proposed that the inclu-
sions have been derived from cumulate layers that formed as the sublayer
magmas fractionated.
The contamination that gave rise to the SiOrrich composition of the Com-
plex is believed to be the cause of the segregation of large amounts of sulfide. It is
suggested that the close association between sulfides, inclusions and sublayer
magmas is the consequence of the sulfides and ultramafic and mafic cumulates
settling together in hidden sills that are present peripheral to and beneath the
Complex. These are responsible in part for the gravity and aeromagnetic
anomalies. They were injected into the fractures in the rocks beneath the
Complex as offshoots from the main magma conduit. As they cooled and frac-
tionated, residual magma rose to the floor of the crater that now holds the

1 Department of Geology, University of Toronto, Toronto, M5S 1A1 Canada

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
92 A. J. Naldrett

Complex to form the presently exposed sublayer. Where magma from a deeper
sill cut and disrupted an overlying sill, it picked up sulfides and inclusions, and
carried them upwards to form the ore deposits.
The magma of the main mass of the Complex, cooling in part within the
central conduit, did so more slowly than the sublayer. It thus became con-
taminated more rapidly, but fractionated less rapidly. It was injected into its
present position as a series of pulses at essentially the same time as the sublayer,
possibly in response to structural adjustments taking place in the overlying
crater.

Introduction

The Ni - Cu ores of the Sudbury district are associated with the Sudbury Igneous
Complex (SIC), a layered intrusion ranging from quartz norite at the base,
through gabbro to a granophyric cap. The purpose of this paper is to describe the
geological setting at Sudbury, discuss the composition of the Complex, show
how the composition can account for the profusion of concentrations of
magmatic sulfide ore, and propose a model to account for the geology of the
Complex and some aspects of its setting.

Regional Geological Setting

The SIC has intruded beneath the Onaping formation (Fig. 1), a breccia com-
posed of fragments of country rocks and recrystallized glassy material set in a
matrix of glassy shards, and variably interpreted as an ignimbrite or the "fall-
back" breccia resulting from the impact of a meteorite. The SIC and strata over-
lying it are exposed as a series of concentric, crudely elliptical rings and dip
towards the centre, giving rise to the interpretation of the structure as a basin.
The basin is located at the contact between gneisses, remnants of greenstone belts
and intrusive quartz monzonites all of Archean age to the north, and rocks of the
Proterozoic Southern Province, which overlie the Archean basement unconform-
ably and thicken to the south.
In a recent analysis of regional gravity and magnetic data, Gupta et al. (1984)
have analyzed the combined residual gravity and magnetic anomaly that marks
the Sudbury region itself. After subtracting large-scale "super-regional" gravity
trends, they conclude that the broad + 20 to 30 mGal anomaly characterizing the
Sudbury region cannot be explained by the rocks of the Complex itself. Their
modelling indicates that a large (60 x 40 km) mass of mafic rock, with a density
similar to gabbro or gabbro-anorthsite (3.02 ± / - 0.03 tlm 3 ) underlies the
Complex at a depth of at least 5 km, extending beyond it, as is indicated in Fig. 2.
Their analysis of the regional magnetic data indicates that this can be explained if
some of the rocks of this body are partially serpentinized.
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 93
LEGEND
1---1 OLIVINE DIABASE _"'=__---,
b -,-.I GRA NOPHYRE
_ QUARTZ-RICH GABBRO
~ NOR ITE AND SUB -LAYER
c::J CHEL MSfORD fORMATIO N
1-- --=3 ONWATIN fORMATION
E::::J ONAP ING fORMATION
~ GRANITE AND GNE ISS
ITIIIJ QUARTZ ITE
1",,,:·:1 GREYWACKE,
AC ID AND BASIC
VOLCANIC , : L<VI'U,;.(JT
ROCKS
fAULTS
~A N D
SHEARING
8 LAKE

46·3 0'-

~ao-I °r;;;=~F5~~~~I
,. I I° MILES
h
! I

l
! , !

KILOMETERS 0 8 I I 16

Fig. 1. Geological map of the Sudbury district

LEGEND

~H CheimSford]~;
Ow Onwotln
o Onoping
... 0
~ oc
-~
~

ffi GronOPhyre~~8;
N Norite
.........
g~~
~-UE~---~~-----~~~~~'-'-----------=
DISTANCE (km) E'
P'l Hidden
LJ lgneous Body 5 15 25 35 55
Other
Mof ic 2.73
....E 2.73

~
Intrusives 5
[
B Gneiss IO~
L
~

0
B Gronite

o 15' SUBLAYER

Volcon ics

8 Sediments

" Brecciotion

Fig. 2. A northwest-southeast section across the Sudbury district showing the residual and
calculated gravity profiles, a two-dimensional reconstruction of the model used to calculate the
gravity profile, and a geological interpretation consistent with the reconstruction. (After Gupta et al.
1984)
94 A. J. Naldrett

Local Geological Setting

A detailed description of the local setting is beyond the scope of this paper. It is
well covered by a number of authors who have contributed to a recent book on
Sudbury geology (pye et al. 1984). Many aspects of the local setting suggest that
an explosion of unusually large intensity gave rise to a crater at Sudbury. These
include:
1. The basinal shape of the structure as interpreted from surface and under-
ground mapping and drilling.
2. The presence of an upturned collar around the basin, as seen particularly in
the Huronian rocks along the southern margin (Dressler 1984).
3. Evidence of shock metamorphism in the country rocks around the structure
(Dressler 1984).
4. The presence of Sudbury breccia (compared with the pseudotachylite of the
Vredefort and Ries structures) in the country rocks around the structure and
footwall breccia beneath the Complex.
5. Evidence of shock metamorphism in country rock inclusions in the Onaping
formation (Muir and Peredery 1984).
6. The 1800 m of Onaping formation itself, the lower part of which is variably
interpreted as a meteorite fall-back breccia or an ignimbrite (Peredery and
Morrison 1984, Muir 1984).
Opinions are divided between an extra-terrestrial and endogenic origin for the
structure. This author believes that meteorite impact is the more likely origin,
primarily because so many of the features observed at Sudbury are also found at
known impact sites. There are, however, many difficulties with such an origin
and these are summarized by Muir (1984).

Petrology of the Sudbury Igneous Complex

The main units of the Complex include (Fig. 3) (1) the sublayer; (2) the marginal
quartz-rich norite of the South Range and mafic norite of the North Range;

Units of
Sudbur

LOWER
1 - - - - - - 1 ZONE

NOT TO
Fig. 3. The main units of the Sudbury Igneous
SCALE Complex
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 95

(3) the South Range norite and Felsic norite; (4) the Quartz Gabbro; and (5) the
Granophyre and Plagioclase-rich Granophyre. All except the sublayer are
included within the main mass of the Complex.

Main Mass

The marginal unit of the main mass on the South Range is the quartz-rich norite.
In this, the quartz content increases very progressively towards the contact over
the outer 300m (Naldrett et al. 1970). This is unlikely to be due to contamination
since the increase in quartz occurs as much where the footwall is composed of
SiOrdeficient greenstone as it does where it is composed of granite. This in-
dicates that if contamination is involved, it is not in situ contamination. The in-
crease in quartz is accompanied by an equally progressive decrease in the average
Mg/(Mg + Fe) ratio of the pyroxenes (Naldrett et al. 1970). This decrease is due
to the pyroxenes becoming progressively more strongly zoned, with
Mg/(Mg + Fe) decreasing towards the edges of grains while the cores retain a
constant composition. These observations, coupled with a decrease in grain size
towards the margins (Naldrett et al.) indicate that the outer part of the quartz-
rich norite is a non-cumulate rock that crystallized essentially in situ.

Sublayer

The sublayer occurs discontinuously around the Complex. It can be divided

logically into those variants which occur close to the outer contact and those
which occur occupying dykes that radiate outwards from the Complex and are
known locally as offsets.
The contact sublayer consists of a suite of fine to medium grained norites and
gabbros that can be distinguished from the main mass felsic norite and quartz
gabbro by their lower quartz content in relation to pyroxene (Naldrett et al.
1972).
Sublayer rocks occurring within the offsets generally have the composition of
quartz diorite and are referred to as such. Grant and Bite (1984) recognize a
number of variants of quartz diorite which form a continuum grading from
hypersthene to biotite quartz diorite, a change that is attributed to varying
degrees of contamination.
Geological relationships between the sublayer and the main mass that bear on
the relative ages of the two are conflicting. Inclusions of marginal, quartz-rich
norite have been observed in sublayer and inclusions of sublayer have been
observed in norite of the main mass of the Complex. Contacts are never marked
by fine-grained chill zones, suggesting that whichever was the older at any par-
ticular location, it was still warm at the time of intrusion of the younger. On the
North Range, the distinction between main mass and sublayer is always clear
with the sublayer having the finer grain size and lower quartz content. This is not
always the case on the South Range, where a number of researchers have com-
mented on gradations that are visible between the two (Slaught 1951, Cochrane
96 A.J. Naldrett

1984). It would seem, therefore, that the introduction of the sublayer and main
mass was a complicated process with one preceeding the other and vice versa in
different localities.

Inclusions in the Sublayer

In some areas the sublayer is characterized by inclusions. These can be divided

into two groups, those of obviously local derivation, and those composed of
mafic and ultramafic rocks, many of which do not outcrop in the Sudbury area.
The latter group are those of concern here. Scribbins et al. (1984) have described
them as ranging from peridotite, through clino- and orthopyroxenites, to olivine
gabbro and norite. Most of the inclusions either display cataclastic or cumulate
textures. Olivine ranges in composition from FO s6 to Fo n , with the Fo content
decreasing as the plagioclase content decreases. Scribbins et al. conclude that the
inclusions are derived from layered intrusions that have fractionated at moderate
depths in the crust.

Geochemistry of the SIC

Major Elements
A selection of average and individual major element analyses of representative
material from the Sudbury Igneous Complex are listed in Table 1.

Main Mass. Judging from field and petrographic criteria, the quartz-rich norite
is the rock type that is likely to have been closest to the composition of the SIC
magma when it was emplaced along the South Range. The Mg/(Mg + Fe) atomic
ratio (henceforth referred to as MgNo) of 0.61 indicates that this is a reasonably
primitive rock.
Since the SIC is an example of continental tholeiitic magmatism, continental
flood basalts provide one of the best yardsticks with which to compare it. In
Fig. 4, the quartz-rich norite is compared with Keewanawan and Columbia River
flood basalts on the basis of their MgNos. The high Si0 2 and K20, low CaO and
low Na20/K20 ratio of the quartz-rich norite when compared with flood basalts
of similar MgNo is very obvious. The siliceous nature of the Complex has been
discussed in the past by Naldrett and Macdonald (1980) who suggested con-
tamination as the cause. Contamination of a relatively unfractionated Columbia
River basalt (column 4, Table 2) with 500/0 of a 1:2 mixture (column 3, Table 2)
respectively of Archean quartz monzonite (column 1) and tonalitic gneiss
(column 2) gives rise to the composition given in column 6. This is quite similar,
considering the arbitrary manner of choosing the end members, to the quartz-
rich norite (column 7). As discussed by Rao et al. (1983), any real process of con-
tamination is likely to be more complex than this, with crystallization accom-
panying assimilation. However, the conclusion on the basis of the major
elements is that contamination of the main mass SIC magma by felsic country
rocks is permissible, in fact, even likely.
Table 1. Comparison of major element content of rocks of the Sudbury Igneous Complex' Cl
(1)
0
(")
~
Offset rocks Main mass Flodd basalts Bushveld (1)

3
2 3 4 5 6 7 8 9 10 11 12 13 14 ...,fa"
'<
Quartz Typical Offset Quartz-rich South Felsic Quartz Grano- Keewan. Colum- Keewan. Colum- norite 0....,
poor Offset (NR) norite (SR) Range norite gabbro phyre basalt bia basalt bia So
(1)
Offset (SR) norite basalt basalt CIl
(SR) 0-
'"
0"

Si02 50.7 55.3 61.5 56.7 57.0 55.4 60.4 58.9 68.1 49.1 48.7 50.0 48.9 53.3 '<
'...,"
Ti0 2 1.56 0.94 0.73 0.57 1.34 0.49 0.65 1.62 0.65 1.89 2.73 0.99 0.96 0.19 ~
::>
(1)
AI20 3 16.3 14.3 14.9 16.3 16.4 17.8 15.6 13.5 13.7 16.80 16.4 16.3 17.0 19.2 0
FeO· 11.4 10.8 6.84 7.44 7.33 7.35 7.19 10.5 5.21 11.91 12.5 9.78 10.0 7.60 co
MnO 0.17 0.15 0.11 0.12 0.13 0.12 0.13 0.14 0.09 0.17 0.18 0.18 0.20 0.16
'"
(j
0
MgO 4.91 4.43 3.82 6.10 6.40 6.68 3.81 2.66 1.08 6.61 5.45 8.08 7.90 7.34 3
'0
CaO 7.92 6.71 4.70 6.66 7.28 7.75 6.05 5.48 2.15 10.03 8.7 11.1 11.1 10.0 (;)

Na20 2.76 2.61 3.15 2.41 2.50 3.12 3.13 3.31 2.76 3.34 2.52 2.49 2.95
..>0
~
K20 1.66 1.70 2.60 1.40 1.55 1.10 1.96 2.13 5.00 0.52 0.91 0.43 0.28 0.38 ~
Total 97.28 96.6 98.4 99.84 99.2 98.90 98.06 99.29 98.79 98.97 99.38 98.53 101.12 0
0-
~
Mg Atomic 0.43 0.42 0.50 0.59 0.61 0.62 0.47 0.31 0.27 0.50 0.44 0.60 0.59 0.63
(Mg+Fe) 0'
...,
So
(1)
Na20/ 1.76 1.54 1.21 1.56 2.27 1.59 1.47 0.66 5.31 3.67 5.86 8.89 7.76
(j
K20 wt.OJo 0

'0
3
• Explanation to Table 1: 1 Grant and Bite (t.v.) Table II, No.4 (Av of 4); 2 Grant and Bite (t.v.) Table II, No.2 (Av of 14); 3 Grant and Bite (t.v.) Table II, (;)
>0
No.5 (Av of 17); 4 Grant and Bite (t.v.) Table II, No.6 (Av of 31); 5 Collins (1934) p. 140, No. 92 W; 6 Collins (1934) p. 140, No. 89 W; 7 Grant and Bite III
::>
(t.v.) Table II, No.9 (Av of 31); 8 Grant and Bite (t.v.) Table II, No. 10 (Av of 17); 9 Collins (1934) p. 139, No. 56 W; 10 Transitional Basalt, Basaltic 0-
......
~
Volcanism, Study Project (1981), Table 1.2.2.19, No KEW 4; 11 Basalt, Basaltic Volcanism Study Project (1981), Table 1.2.3.4, No CP-4; 12 Olivine co

Basalt, Basaltic Volcanism Study Project (1981), Table 1.2.2.19, No KEW-8; 13 Olivine Basalt, Basaltic Volcanism Study Project (1981), Table 1.2.3.4, No ...,0
CP-3; 14 Naldrett, unpublished data. f.l

\0
.....
98 A. 1. Naldrett

• Keweenawan Basalt (BV. 1981) Fig. 4. Comparison of some

o Columbia River Basalt (BV. 1981)
.. Quartz-rich Norite major element concentrations in
(Grant & Bite. 1984) the quartz-rich norite with those
" Quartz-rich Norite in typical continental flood
(Collins. 1934)
basalts
55
Si02
(wt.%)
50

45~-------+--------~--------+---~

't ~"..
: .
00
K2 0 • ·.0 0 .-. 0: I
(wt.%) 0 L----'~~=-==f======t====-------+.------l

ot~.:J.
CaO • •
(wt.%) 0 •

8 ,

.
10 J
Na20 8
K20 6 .. ..
(wt.%) 4 o
~O
2

0.6 MgNo 0.5 0.4

Sublayer. The texture and field relations of the constituent units of the sublayer
indicate that these are also not cumulates, but are rocks that have solidified
essentially in situ, and thus represent magma compositions. The MgNos are
therefore important indices of the degree of fractionation of the rocks and of the
magmas. MgNos of North Range samples of contact (as opposed to offset) sub-
layer given by Rao et al. (1983) range from 0.42 to 0.50, and average 0.47; those
from the South Range vary from 0.37 to 0.58 and average 0.51. Thus the magmas
are relatively fractionated, considerably more so than the main mass as repre-
sented by the quartz-rich norite. At the same time, the MgNos indicate a variable
degree of fractionation for the sub layer.
Comparison of the major element compositions (Fig. 5) of sublayer with
those of flood basalts indicates that the sublayer is also enriched in Si0 2 and
K20, and low in CaO and Na20/K20 ratio for a given degree of fractionation as
indicated by the MgNo. Contamination by country rocks is therefore indicated
by the data.
The data of Rao et al. (1983) on the contact sublayer indicate a wide range of
MgNos, and they have called upon fractional crystallization in addition to assim-
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 99

Table 2. Compositions relating to the contamination of the Sudbury Igneous Complex magmas a

2 3 4 5 6 7 8 9
Major elements wt. 070

Si02 70.4 64.8 66.7 48.9 49.1 57.8 57.0 53.5 51.7
Ti0 2 0.46 0.78 0.67 0.96 1.89 0.82 1.34 1.59 1.31
Al20 3 15.6 16.2 16.0 17.0 16.80 16.5 16.4 16.6 14.7
FeO li 2.2 4.35 3.63 10.0 11.91 6.81 7.83 9.84 12.84
MgO 0.73 2.43 1.86 7.9 6.61 4.88 6.46 5.42 6.21
CaO 1.96 3.48 2.97 11.1 10.Q3 7.04 7.28 8.27 9.52
Na20 3.93 3.89 3.90 2.49 2.76 3.20 2.41 3.05 2.81
K20 3.33 1.80 2.31 0.28 0.52 1.30 1.55 0.97 0.81

Trace elements ppm

La 6.77 15.2 18.0

Ce 97.9 83.8 89 16.0 33.5 52.3 57.6 47.4 48.2
Nd 31. 37.2 35.1 26.8 17.8
Sm 4.55 6.57 5.9 2.72 4.80 4.31 4.52 5.08 4.65
Eu 0.89 1.66 1.37 1.03 1.58 1.20 1.26 1.53 1.96
Gd 2.88 4.79 4.15 3.18 d 4.8 d 3.67 4.13 4.64 4.60 d
Tb 0.59 0.83 0.56 0.76
Yb 0.696 1.20 1.03 2.20 2.35 1.62 1.88 2.02 1.43
Lu 0.33 0.36 0.27
Ba 807 992 930 180 555
Rb 166 75.9 106 2.3 10 54 34 21
Sr 290 753 599 338 284 469 362 602
MgNo C 0.477 0.585 0.497 0.56 0.611 0.495 0.463
K/Rb
Rb/Sr 0.18 0.007 0.035 0.115 0.094 0.035
Sc/Ba 0.64 1.88 0.85
Ce/Yb

a Explanation to Table 2: 1 Average "Granite" from Table XX.2.1. Trace elements are of Archean
quartz monzonite with very similar major element composition from Arth and Hanson (1975),
Table 2, analysis 29. 2 Average Gneiss from Table XX.2.1. Trace elements are of tonalitic gneiss of
similar major element composition from Arth and Hanson (1975), Table 2, analysis 22.3 Mixture
of 1 and 2 in the proportion 1: 2. 4 Relatively primitive continental flood basalt from Columbia
River Province, Basaltic Volcanism Study Project (1981), Tables 1.2.3.4 and 1.2.3.5 analysis No.
CP-3 (calculated on volatile free basis). 5 Moderately fractionated continental flood basalt from
Keewanawan Province, Basaltic Volcanism Study Project (1981), Tables 1.2.2.19, 1.2.2.22 and
1.2.2.23, analysis No. KEW-4 (calculated on volatile-free basis). 6 Calculated composition of South
Range initial magma combining columns 3 and 4 in the proportions 1: 1. 7 Composition of
marginal quartz-rich norite (Table XX.3.1, analysis 5) for comparison with column 6. 8 Calculated
composition of a relatively primitive sublayer magma, combining columns 3 and 5 in the propor-
tions 1: 3. 9 Composition of one of the least fractionated North Range sublayer samples from Rao
et al. (1983), Tables 1 and 3, analysis BVR-l1.
All analyses calculated on a volatile-free basis.
b FeO T = Fe calculated as FeO.
C MgNo = Mg/(Mg + Fe) atomic ratio.
d Chondrite normalized Gd for these two samples assumed to be the weighted average of two parts of
Tb and one part of Sm, all on a chondrite normalized basis.
100 A. J. Naldrett

• North Range Sublayer Fig. 5. Comparison of some ma-

o South Range Sublayer •• jor element concentrations in the

Si02
60 .
o
sublayer with those in typical
continental flood basalts. The
(wt.%) flood basalts fall within the out-
55 lined area (see Fig. 4)

50

3
..
o
..
o

10

*
8
CaO
(wt.%)
6

*
0.6 MgNo 0.5 0.4

ilation to explain the observed compositional variations. One sample (BVR-l1,

marked by a star in Fig. 5) is one of the more primitive of the North Range
samples. In its major element composition and Na20/K20 ratio of 3.47, it is not
unlike some of the more fractionated flood basalts. However, its trace element
composition is different, as is discussed below.
 Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 101

Trace Elements

Data on trace elements in the SIC are much less abundant in the literature than
those on major elements. Data exist for REE elements on most units of the Com-
plex, but that of Rao et al. (1983) on the sublayer are all that are available in the
way of modern determinations of other incompatible trace elements.
In Fig. 6, Kuo and Crocket's (1979) REE pattern for the quartz-rich norite is
compared with those for the two flood basalts samples of similar MgNo. The
much steeper slope of the Sudbury sample (La/Yb = 17) as compared to the
flood basalts (La/Yb = 3) is obvious.
The range of REE patterns for the sublayer of the SIC, together with the
averages for the North and South Ranges are compared with flood basalts of
similar MgNo in Fig. 7. The sublayer patterns, with their La/Yb ratios of 16, are
essentially identical in slope and magnitude to the average pattern for the quartz-
rich norite and, again, are very much steeper than those for the flood basalts
(La/Yb = 6).
The LREE enrichment of the Sudbury rocks could be attributed to contamina-
tion by LREE-enriched country rocks. One feature of the Sudbury patterns is the
absence of negative Eu anomalies. This would preclude contamination by typical
. granite or quartz monzonite with their characteristic negative Eu anomalies.
Tonalitic gneisses typically are enriched in LREE and show no Eu anomalies (Arth
and Hansen 1975), so that they are a possible contaminant. However, they cannot
account for the low Na20/K20 ratios of the Sudbury rocks.
A representative set of trace element data for the country rocks of the Sud-
bury area are not available. However, the major element chemistry of the Sud-
bury tonalites and quartz monzonites are identical to those studied by Arth and

100
90
80
70
60
50

.i!l. 40

«
~ 30

!
:€j 20
~
"o
o

'"uo 10
II: 9
~ 8
~ 7
:i: 6
5
4

3 Fig. 6. Comparison of REE data for the

quartz-rich norite with that for a
Keewanawan and a Columbia River
flood basalt of similar MgNo
 102 A. J. Naldrett
200
Fig. 7. Comparison of REE data for
the sublayer with that for relatively
fractioned Keewanawan and Colum-
bia River basalts

30

~
.t; 20
c
o

---
s::.
()

""uo 10 ~//~~~
,~

a: 9 Columbia River Basalt ' '~'''''~-

~ 8 Keweenawan Basalt
7
s::.
!: 6 5,,",' ....... 5,••"" ' ,...g. /./
5 North Range Sublayer Average

4 Tolal Variation of Sublayer

100
90
80
70
60

!
~ 20
c:

x,,', . .
o
.t::.
()

....
""o<J 10
a: 9
~ 8 Measured Sublayer / ""'" ..., / 0
.t::. 7
:: 6 Contaminanl
5
4

3 I Fig. 8. REE data relating to the mixing

I of Archean tonalitic gneiss and quartz
I monzonite with relatively un fractionated
I Columbia River basalt to produce a com-
position similar to the quartz-rich no rite
 Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 103

200 Fig. 9. REE data relating to the mixing

of Archean tonalitic gneiss and quartz
monzonite with a relatively fractionated
100
90 Keewanawan basalt to produce a compo-
80 sition similar to one of the less con-
70
taminated samples of the sublayer
60
50
:g, 40
os
~ 30
.,
..:

~ 20------.....
,g ...........
(.)

'"o<.> 10
II: 9
"*
.t::. 78 Measured Quartz-rich Norite
3: 6 Modelled Quartz- r ich Norite
5
4

2 LaCe

Hansen in northeastern Minnesota. REE data for a typical tonalitic gneiss and
quartz monzonite, and then for a 1 : 2 mixture respectively of quartz monzonite
and tonalite are given in Fig. 8. A linear mixture of this with relatively
unfractionated Columbia River basalt is compared with the Sudbury quartz-rich
no rite pattern in the same figure. It can be seen that the two are similar,
consistent with the close match that this mixture also produces for the major
elements.
In Fig. 9, one of the least contaminated contact sublayer samples is modelled
by mixing 25070 of the postulated granite-gneiss contaminant (column 3, Table 2)
with 75% of a relatively fractionated Keewanawan basalt (column 5, Table 2).
The REE pattern indicates a close match, except that the model composition does
not show the positive Eu anomaly of the sublayer sample. However, this positive
Eu anomaly is atypical of the sub layer - very few other sub layer samples show
it. Rao et al. emphasize that the majority of their sub layer samples are best
modelles as the consequence of the fractionation of the more primitive variants,
coupled with assimilation of country rocks. From the above discussion,
particularly that with reference to the REE profiles, even the more primitive
sublayer material seems to have been contaminated to some extent, although this
contamination has been less intense than that which affected the initial magma of
the main mass before the process described by Rao et al. (1983) started.

Isotopes
Krogh et al. (1984) have shown from Pb - U radiometric dating of zircons and
baddeleyite that the SIC has an age of 1.85 Ga. This has been used as a model age
104 A.J. Naldrett

.710

(SUdbUry Intru~ive Complex

.708 , ,/Ouartz-rich Norite
Bushveld-: r
I - South Range No~ite and
: K::.North Range Nonte
.706 I Sublayer
I
j,

.704
Great Dyke-o

.702

4.0 3.0 2.0 1.0 o

Age In 10' years
Fig. 10. Sr initial ratios for rocks of the SIC compared with those for other large layered intrusions
and a possible mantle growth line

to recalculate the initial Sr 87/S 86 ratios for the data of Gibbins and McNutt
(1975). The average and range for a number of units of the SIC are shown in
Fig. 10 compared on the basis of age with data for a number of other layered in-
trusions, and with a hypothetical mantle growth line. While Sudbury is not
unique in having high (Sr 87/Sr 86 )o ratios for its age, these values nevertheless
suggest either the incorporation of radiogenic strontium into the SIC magma,
through interaction with country rocks, or derivation of the magma from a
portion of the mantle that contained very radiogenic strontium. Values of the
sublayer are more variable, but are somewhat lower than those for the main mass
norites. This suggests that if contamination was involved, the main mass was
affected more consistently and to a greater degree than the sublayer .

Conclusions

Field relations and radiometric data indicate no substantial age difference

between any of the SIC rocks, including the sublayer, main mass norites and
granophyre. The unusually quartz-rich composition of the SIC magma relative to
its MgNo, and the LREE enrichment and the low Na20/K20 ratios shown by all
units are anomalous relative to other layered mafic complexes and to continental
flood basalt volcanism, and can be explained as the consequence of the con-
tamination of flood basalt magma by Archean basement rocks.

Relationship of Ore Deposits to the Rocks of the Complex

Several new descriptions and a summary of the general features of the ore
deposits are contained within the Sudbury volume (Pye et al. 1984). It is beyond
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 105

the scope of this paper to repeat them here. Although there is considerable varia-
tion in the characteristics of different ore deposits, there are a number of features
common to all of them. These include:
1. Embayments or Other Irregularities at the Base of the SIC. An increase in
sulfide content it usually observed at the lower contact throughout the Com-
plex, but it is where irregularities exist that the zone of sulfide thickens and
increases in intensity sufficiently to form ore.
2. The Presence of Sublayer. The spatial relationship of ore to sublayer is such
that the sulfides constituting the ore bodies appear to have settled out of
bodies of sublayer. Sulfides also occur within quartz-rich and mafic norites,
but except where they have been moved from their original position by
faulting, ore bodies are invariably associated with sublayer.
3. Ultramafic Inclusions Within Sub/ayer. Some sublayer is mineralized, other
varieties seem to be devoid of significant sulfide. There is little to distinguish
mineralized from unmineralized sublayer in so far as texture, mineralogy and
chemical composition of the rocks are concerned, except for the obvious
presence of sulfide. All types of sublayer contain inclusions, but both in the
contact and the offset environments, the mineralization is associated only
with sublayer that carries members of a suite of mafic and ultramafic inclu-
sions that Scribbins et al. (1984) conclude are derived from one or more
layered intrusions.

Relationship Between Inclusions, Sublayer, and Main Mass of the SIC

As discussed above, the REE profiles of the Sudbury rocks differ from those of
most layered intrusions and suggest that contamination by country rocks has
occurred on a large scale. The similarity in the profiles of the sublayer and the
main mass rocks is suggestive that the magmas responsible for them have a com-
mon heritage. REE profiles for inclusions have the same steep slopes (Fig. 11),
although the abundances are lower. Thus there is some evidence that the inclu-
sions were derived from a magma that had suffered similar contamination to the
remainder of the SIC rocks.
The MgNos of samples of main mass and sub layer are illustrated in Fig. 12
(contrary to the previous representation of MgNos, in Fig. 12 only 900/0 of the
total Fe has been included in calculating them, to allow for Fe in the magma in
the ferric state). Using Roeder and Emslie's (1970) relationship between the
Mg/Fe ratio of olivine and of the basaltic liquid that is in equilibrium with it,
MgNos of liquids in equilibrium with the olivine of the inclusions have been cal-
culated. It can be seen that they span the range of MgNos exhibited by the sub-
layer samples. This indicates that ignoring the Si0 2 contents of all of the mag-
mas, the range of olivine compositions observed in the inclusions is precisely that
which would be expected in cumulate rocks crystallizing from a magma evolving
along a compositional trend such as that recorded within the sublayer. Returning
to the question of the Si0 2 contents of the Sudbury magmas, although olivine is
not a common constituent of the sublayer, Pattison (1979) has reported it as
 106 A. J. Naldrett

100 Fig. 11. REE profile for a harzburgite in·

90
80 clusion in the sublayer compared with
70 that for other SIC rocks. (Rao, personal
60 communication, 1984)
50
8, 40
CD
Ci;
> 30
<
.~
i; 20
c
0
.r.
0

"""010
a: 9
G>
'0 87
.r.
;: 6
5
Peridotite Inclusion
4 70% olivine
30% plagioclase +
3 orthopyroxene

~ Calculated liquids in equilibrium with oliv ine

MAIN MASS SUBLAYER

=
.,.,." Quartz- rich IIIIIHID Ollset
No rite IZ2:I North Range Contact
~ South Range Contact

Fig. 12. MgNos of main mass and sub·

layer samples that are believed to ap·
proximate the compositions of liquids,
compared with the calculated MgNos of
liquids in equilibrium with olivines of the
0.6 0.5 0 .4 inclusions of the sublayer
MgNo

present at Whistle. The sublayer here is not obviously different in composition to

that elsewhere, which suggests that olivine may have played a role in the frac-
tionation of much of the sub layer , and only has disappeared from the liquidus
shortly before it was emplaced.
Thus two lines of evidence, REE profiles and olivine compositions, leads this
author to suggest that inclusions, sublayer, and main mass norite are all related,
and to propose the model given at the end of this paper.
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 107

Segregation of Sulfides

The SIC differs from other layered complexes in a number of significant ways,
including:
1. Evidence that the area into which it was intruded had been involved in a cata-
strophic explosion, in the view of this author probably the consequence of the
impact of a meteorite.
2. Rocks of the Complex are very siliceous when compared with other intrusions
on the basis of the Mg/(Mg + Fe) ratios of their pyroxenes, implying that the
magma responsible was unusually siliceous for its state of fractionation.
3. Other compositional data based on major and trace elements and isotopic
analyses that indicate that the high Si0 2 content is the consequence of country
rock assimilation.
4. An unusually large number of occurrences of very concentrated sulfide.
Irvine (1975) pointed out that assimilation of Si0 2-rich material by a mafic
magma could lower the solubility of sulfur within it, and suggested that this had
occurred at Sudbury. Naldrett and Macdonald (1980) used the 1200°C
isothermal section of the FeO - FeS - Si0 2 system, reproduced here as Fig. 13, to
illustrate the principle. Composition A lies in the field of homogeneous liquid,
but addition of Si0 2 to change its composition to B moves it into a two-liquid
field, and will result in the segregation of two immiscible liquids, one silicate-rich
(Y) and the other sulfide-rich (X). The evidence bearing on the Si0 2 -rich nature
of the SIC lends support to this hypothesis. The evidence indicating that exten-
sive country rock assimilation has occurred points to how the Si0 2 -rich composi-
tion has been achieved. The highly fractured ground underlying an explosion
site, coupled with the heating and melting resulting from the explosion would
have provided the ideal environment in which country rock assimilation could
have occurred.

Fig. 13. 1200°C isotherm of the

°
Fe - S - system illustrating how the
addition of Si0 2 to a homogeneous,
sulfide· rich liquid of composition A
will change its composition to point
Wustite + Liqu id B, where it will consist of two liquids,
'W~O~~'ii' Pm!~~IE!IM'V'
one silicate-rich (Y) and the other
sulfide-rich (X)
108 A. J. Naldrett

A Model for the Sudbury Igneous Complex

It is suggested that the catastrophic explosion that occurred in the Sudbury

region 1.85 Ga ago (this author believes that the weight of the available evidence
points to the impact of an extra-terrestrial projectile as the cause, although
impact is not essential to the development of this model) gave rise to the Onaping
formation, fractured the underlying crust, and triggered ascent of a magma
similar in its initial composition to continental flood basalt. Magma probably
rose through a centralized plumbing system in the lower crust, but some of it
spread out through numerous subsidiary channels as it entered the highly frac-
tured ground beneath the crater (Fig. 14). It is possible that these hidden, flank-
ing intrusions account in part for the buried mass of mafic and ultramafic rock
that the analysis of gravity and magnetic data by Gupta et al. (1984) has indicated
underlies and extends beyond the present Sudbury structure.
On its initial ascent, the magma fractionated to some degree, achieving an
MgNo of 0.65 or slightly less. Batches of magma that infiltrated flanking sub-

Ore Accumulation
Sublayer Magma
Main Mass of SIC
SIC Magma Pulse in
.
O naplng Main
Fm . Conduit
Footwall Rocks / /

Fig. 14. Cross-section illustrating the emplacement of the SIC. Magma rises up a vertical conduit into
a highly fractured zone beneath an explosion (possibly impact) crater. The central body of magma
cools somewhat slowly, assimilating much of the fractured country rocks. Subsequently it is
emplaced into the crater, beneath the low density Onaping formation, in response to structural read-
justments in the area. It is probable that this is a multi-stage process. Magma has also spread out as a
series of sills peripheral to the central conduit. ,Assimilation also occurs within these, although
possibly to a lesser extent than in the central conduit due to their cooling more rapidly. The magma of
the sills deposits sulfide and fractionates, giving rise to olivine and pyroxene-rich cumulates. Frac-
tionated magma from the flanking sills also works its way up to the crater floor, intruding to form the
sublayer of the contact and offsets. Where magma from a lower sill has cut and disrupted an overly-
ing one, it has picked up sulfides and inclusions, intruding with these in suspension within itself. Sub-
sequently both sulfides and inclusions settle out to give rise to the ore deposits with their typical
zonation of massive sulfides overlain by inclusion-bearing massive sulfide overlain in turn by sulfide-
rich sublayer rock
Geochemistry of the Sudbury Igneous Complex: A Model for the Complex and Its Ores 109

sidiary channels cooled, became contaminated through reaction with their sur-
roundings, and fractionated further, giving rise to olivine and pyroxene-rich
cumulates. The contamination, coupled with the cooling, caused sulfides to
precipitate at an early stage: these collected as massive bodies beneath the
cumulates in the flanking chambers of the plumbing system. The main body of
magma lost heat less quickly than that in the flanking chambers, and thus frac-
tionated less rapidly; because it cooled more slowly, it reacted with country rocks
to an even greater extent than was the case in the flanking bodies.
Successive injections of magma occurred along the unconformity between the
crater floor and the overlying, relatively unconsolidated, and thus low density,
Onaping formation to give rise to the main mass of the Complex.
The faster cooling, and thus more fractionated (consistent with the MgNos),
but less contaminated (consistent with the Sr initial ratios) magma of the flanking
bodies was also squeezed up to intersect the crater floor, where it spread out,
exploiting "troughs" and "embayments" in the floor filled with footwall breccia
and also infiltrating fractures in the basement rocks to form the offset dykes.
This magma is thus responsible for the sub layer. In some instances (Fig. 14), as it
rose up through the basement, it intersected and disrupted overlying bodies,
picking up sulfides and ultramafic cumulates in the process, thus accounting for
the very important association between the mineralization and the xenoliths.
Other batches of magma did not intersect overlying bodies and were emplaced
carrying only country rock inclusions and giving rise to the unmineralized sub-
layer that has been described. The fact that the sub layer intrusions were derived
from a number of bodies following parallel, but distinct, evolutionary trends
accounts for the greater variability of the sublayer in MgNo and Sr initial ratio in
comparison with the main mass.

References

Arth JG, Hanson GN (1975) Geochemistry and origin of the early Precambrian crust of Northeastern
Minnesota. Geochim Cosmochim Acta 39:325 - 362
Cochrane LB (1984) Ore deposits of the copper cliff offset. In: Pye EG, Naldrett AJ, Giblin PE (eds)
The geology and ore deposits of the Sudbury structure. Ont Geol Surv Spec Vol 1:347 - 360
Dressler BO (1984) The effects of the Sudbury event and the intrusion of the Sudbury Igneous
Complex on the footwall of the Sudbury structure. In: Pye EG, Naldrett AJ, Giblin PE (eds) The
geology and ore deposits of the Sudbury structure. Ont Geol Surv Spec Vol 1:97 - 138
Gibbins WA, McNutt RH (1975) The age of the Sudbury Nickel Irruptive and the Murray granite.
Can J Earth Sci 12:1970-1989
Grant RW, Bite A (1984) Sudbury quartz diorite offset dikes. In: Pye EG, Naldrett AJ, Giblin PE
(eds) The geology and ore deposits of the Sudbury Structure. Ont Geol Surv Spec Vol 1:275 - 300
Gupta VK, Grant FS, Card KD (1984) Gravity and magnetic characteristics of the Sudbury structure.
In: Pye EG, Naldrett AJ, Giblin PE (eds) The geology and the ore deposits of the Sudbury struc-
ture. Ont Geol Surv Spec Vol 1:381-410
Irvine TN (1975) Crystallization sequence of the Muskox intrusion and other layered intrusions - II.
Origin of chromitite layers and similar deposits of other magmatic ores. Geochim Cosmochim
Acta 3:991 -1020
Krogh TE, Davis DW, Corfu F (1984) Precise U - Pb zircon and baddeleyite ages for the Sudbury
Area. In: Pye EG, Naldrett AJ, Giblin PE (eds) The geology and ore deposits of the Sudbury struc-
ture. Ont Geol Surv Spec Vol 1:431- 447
110 A. J. Naldrett: Geochemistry of the Sudbury Igneous Complex

Kuo HY, Crocket JH (1979) Rare earth elements in the Sudbury nickel irruptive: Comparison with
layered gabbros and implications for nickel irruptive petrogenesis. Econ Geol 79:590- 605
Muir L (1984) The Sudbury structure. Consideration for models of an endrogenic origin. In: Pye EG,
Naldrett AJ, Giblin PE (eds) The geology and ore deposits of the Sudbury structure. Ont Geol Surv
Spec Vol 1:449 - 490
Muir TL, Peredery WV (1984) The Onaping formation in the geology and ore deposits of the Sudbury
structure. In: Pye EG, Naldrett AJ, Giblin PE (eds) Ont Geol Surv Spec Vol 1:139-210
Naldrett AJ, Macdonald AJ (1980) Tectonic settings of some Ni-Cu sulfide ores: importance in
genesis and exploration. Geol Assoc Can Spec Pap 20:633 - 657
Naldrett AJ, Bray JG, Gasparrini EC, Podolsky T, Rucklidge JG (1970) Cryptic variation and the
petrology of the Sudbury Nickel Irruptive. Econ Geol 65:122-155
Naldrett AJ, Greenman L, Hewins RH (1972) The main irruptive and the sublayer at Sudbury,
Ontario. Proc 24th Int Geol Congr, Montreal, Sect 4:206-214
Pattison EF (1979) The Sudbury sublayers: its characteristics and relationships with the main mass of
the Sudbury irruptive. Can Miner 17:257 - 274
Peredery WV, Morrison GG (1984) Discussion of the origin of the Sudbury structure. In: Pye EG,
Naldrett AJ, Giblin PE (OOs) The geology and ore deposits of the Sudbury structure. Ont Geol Surv
Spec Vol 1:491- 512
Pye EG, Naldrett AJ, Giblin PE (eds) (1984) The geology and ore deposits of the Sudbury structure.
Geol Surv Spec Vol 1:603 pp
Rao BV, Naldrett AJ, Evensen NM, Dressler BO (1983) Contamination and the genesis of the
Sudbury ores. In: Pye EG (00) Geoscience research grant program, summary of research,
1982 -1983. Ont Geol Surv Misc Pap 113:139 -151
Roeder PL, Emslie RF (1970) Olivine-liquid equilibrium. Contrib Mineral Petrol 29:275 - 289
Scribbins B, Rae DR, Naldrett AJ (1984) Mafic and ultramafic inclusions in the sublayer of the
Sudbury Igneous Complex. Can Miner 22, 1:67 -75
Slaught WH (1951) A petrographic study of the copper cliff offset in the Sudbury district. M Sc
Thesis, McGill Univ (unpublished)
Sulfide Petrology and Genesis of Copper-Nickel Ore
Deposits
v. V. DISTLER, A. D. GENKIN, and O. A. DYUZHIKOV 1

Abstract

Some general problems of the formation of copper-nickel ores are discussed on

the basis of new data from the Norilsk deposits. These deposits, together with the
deposits of greenstone belts and some others, are classified as the deposits con-
nected with ore-bearing volcanic-intrusive complexes of intercontinental rift
zones. Massive sulfide ores, sulfide-bearing intrusions, and ultramafic vo1canites
are shown to have been formed as the result of successive emplacements of
magmatic melts formed in the process of deep differentiation of mantle magmas.
The factors determining the zonality of distribution of sulfide mineral assem-
blages in sulfide-bearing intrusions of the bodies of massive sulfides and the
formation conditions of platin-bearing horizons of layered intrusions are dis-
cussed.
In the recent decade, studies of copper-nickel deposits have been undertaken
within the International Program of Geological Correlation. A number of basic
works aimed at investigating copper-nickel deposits were carried out in the
USSR. These studies have contributed significantly to a better understanding of
geological and petrological regularities of sulfide ore formation.
A concept of the magmatic origin of copper-nickel ores as products of the
evolution of ultrabasite-basite sulfide-silicate magmatic systems is emerging.
Investigations of ancient copper-nickel ore deposits in greenstone belts of the
Archean cratons were especially important in this respect. Presently, the results
obtained, as well as the achievements in studies of evolution of structures and
endogenous ore formation through the Earth's history (Kazanskii 1983, Smirnov
1982), allow us to consider the copper-nickel ores as a specific type of magmatic
product formed in the course of definite structural transformations of the
continental crust.
Two groups of magmatic assemblages, the formation of which is accom-
panied by copper-nickel ore formation persist in all metallogenic epochs. The
first group includes highly differentiated volcanic-intrusive magmatic assem-
blages occurring in the zones of near-fault depressions or intercontinental rift
structures; the second group comprises magmatic assemblages forming rhythmi-

I Institute of the Geology of Ore Deposits, Petrography, Mineralogy and Geochemistry of the USSR
Academy of Sciences, Moscow 109017, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
112 V. V. Distler et al.

cally layered stratified plutons occurring in the regional deep faults pertaining to
cratonic activity. These two groups show essential differences in ore-genetic pro-
cesses and the scale of ore matter concentration.
Volcanic-intrusive assemblages give rise to the deposits in the Archean green-
stone belts of Western Australia, the Rhodesian and Canadian shields, and also
the Proterozoic belt of Pechenga-Imandra-Varzuga and the early Mesozoic deep-
seated Norilsk block of the Siberian platform. Here the high degree of deep dif-
ferentiation of initial magmas that is responsible for the wide diversity of types of
melts and for separating of the great volumes of essentially sulfide magma in-
truding as an independent subphase is characteristic.
Evidence of the separate introduction of sulfide melts could be presented in
more detail for the deposits of the Norilsk ore district. The main facts are as
follows: (1) The formation of massive sulfide ores is separated from the emplace-
ment of sulfide-bearing intrusives during the time in which magma cools to such
a state that the intrusives react during tectonic deformation as solid bodies. The
tectonic movements lead to the generation of cavities of different sizes that were
filled by sulfide liquid with the formation of veins. The part of such a vein filling,
formed by the moving apart of rock blocks and the fabric of the vein contacts are
shown on Fig. 1 A, B. (2) The majority of the bodies of massive sulfide ores are
separated from the intrusives by unmineralized sedimentary rock seams of dif-
ferent thicknesses. The formation of massive sulfide ores is accompanied by their
own aureole of metamorphic transformations that are superimposed on contact-
metamorphic rocks formed by the emplacement of intrusive. Recrystallization
and remelting of hornfels and metasomatic rocks (Fig. 1 C) are the main types of
contact transformations connected with massive sulfide ores. (3) The bodies of
massive sulfide ores are localized not only on the lower endocontact of intrusives
but, under favorable tectonic conditions, also in the rocks of the roof of the in-
trusives or intersect the sulfide-bearing intrusives. (4) The absence of correlation
of the zonality of sulfide mineralization of layered intrusives and the zonality of
massive sulfide ores. (5) The disseminated ore mineralization accompanying the
massive ores is superimposed on sedimentary-metamorphic rocks and the rocks
of intrusives as well.
It is suggested that magmatic rocks and sulfide ores originate from ultra-
mafic-mafic melts, that are the result of partial melting of garnet peridotites the
composition of which is equivalent to that of the integrated differentiated sul-
fide-rich magmas. Involvement of large masses of sulfide-rich mantle substrate
and the fractionation of the parent ultramafic-mafic melt in the conditions of
high permeability of the Earth's crust contributed to ore formation. The Norilsk
ore basin is a good example of the volcanic-intrusive nickel-rich association
activated in the connection with the development of the Yenisei-Khatanga global
rift of Early Mesozoic age.

Fig. lA - C. The relation of sulfide massive ores with enclosing rocks. A Vein massive ores in a
completely differentiated intrusion. The vein was formed by moving apart the blocks of gabbro-
dolerites (shown by arrows). Polished hand specimen; natural size. B Contact of massive ore with
gabbro-dolerite. Thin section. Magnification 20 x. 1 Massive ore; 2 reaction rim of magnetite with
pyroxene relics; 3 clinopyroxene; 4 plagioclase. C Granophyre formed by remelting of sandstones on
contact with massive sulfide ores. Thin section; magnification 20 x
Sulfide Petrology and Genesis of Copper-Nickel Ore Deposits 113

Fig.lA-C
 r r r r r r r r
r
r r

.e
E

~
0.
Ci \ 01" ~ tl
.. '"
:l: O~ () 0 ~~~~{ ~
~l~
..0

0
~

-'"
.. >
.....
0
u 0 o tl ~ 0
vJ O~ 0~ ~ O~
D 0 ~
~ '() i.\~
« 0 () <::l ~
oc
~ ~ 0
0 ~~
::;

.. d V
r 0 -- -- -- --
~
"t: 6
:::>
<'> 0 Sulfidep oor ultra malic mell
~

Fig. 2. Genetic model of magmatism and sulfide copper-nickel ore formation in the Norilsk region.
1-4 Late Paleozoic-Early Mesozoic volcanogenic rocks: 1 Tholeiite basalts of later magmatism;
2 hyperbasite-basite basalts (comagmatic to differentiated intrusives); 3 tholeiitic basalts of enhanced
alkalinity in the early stage of magmatism; 4 alkaline-basite basalts; 5 Deep sills of hyperbasite-
basites; 6 - 9 sulfide-bearing intrusives and massive ores: 6leukocratic differentiates (Phase I); 7 dif-
ferentiates with ore inclusion (Phase II); 8 massive ores; 9 melanocratic differentiates (Phase III);
10 deep faults controlling the alkaline-basite and hyperbasite-basite volcanogenic rock development
Sulfide Petrology and Genesis of Copper-Nickel Ore Deposits 115

The volcanic-intrusive assemblage of the Norilsk region includes the ultra-

mafic-mafic lavas and later emplaced ore-bearing layered intrusives; these are the
result of deep differentiation of sulfide-rich juvenile melts (Fig. 2). This process
is responsible for the segregation of immiscible sulfide liquids and for the
emergence of several groups of intrusives differing in their composition, degree
of inter-chamber differentiation, and content of ore. The highest concentration
of sulfide liquid results in the formation of intrusions with disseminated ores and
of massive ores due to the independent intrusion of sulfide liquid. Also there are
grounds to believe that in such structural conditions a much greater part (by
volume) of ultrabasite-basite melts does not ascend beyond the upper mantle, or
at best reaches only the basement of the consolidated crust. This may be attribut-
ed to the density relationship among the lower horizon of the crust, mantle, and
ultrabasite-basite melts, and also to the relationship between the lithostatic load
and the pressure of the rising ultrabasite-basite astenolith (Cox 1980). The melt
spreading in subhorizontal directions and its solidification make the crust heavier
and thicker, and lead the border with the mantle to become diffuse. This type of
deep structure is, evidently, one of the most essential features defining the areas
where volcanic-intrusive nickel-rich associations develop. It also should be
stressed that the location of copper-nickel deposits developed under hypabyssal
subvolcanic conditions is controled by systems of long, deep-reaching faults and
related regional depressions; the latter coincide with the areas of ultrabasit~­
basite and earlier alkaline-basite magmatism.
It is suggested that prechamber splitting of melts does not occur in magmatic
assemblages of rhythmically layered plutons due to the inherent structural condi-
tions of such magma intrusions. As a result, the bulk of such sulfides is brought
into the ore-deposition zone with the silicate melt. Ore deposition in areas asso-
ciated with layered plutons is a process of interrelated consolidation of two
immiscible melts - sulfide and silicate.
The formation of copper-nickel ores is controlled by two types of evolution
of the sulfide melts: (1) evolution of isolated sulfide liquid without essential inter-
action with the enclosing environment; (2) evolution of sulfide liquid in close
interaction with the silicate melt.
The major purpose of research in sulfide petrology of magmatic ore forma-
tion is to ascertain the regularities governing the formation of copper-nickel ores,
the stability of their parageneses, factors contributing to the zonality of mineral-
ization in the general process of differentiation of sulfide and sulfide-silicate
systems.
Studies of ore formation associated with the evolution of isolated sulfide
liquid, interpretation of the paragenetic relationships of the ore-forming
minerals - pyrrhotite, chalcopyrite, pentlandite, cubanite, and magnetite
groups - are based on fundamental data provided by investigations of sulfide
systems undertaken by Kullerud and Donnay (1970), Naldrett et al. (1967), Craig
and Kullerud (1968), Yund and Kullerud (1966), Cabri (1973), and others.
The interpretation of the crystallization of ores on the basis of monosulfide
solid solutions forming on the liquidus of a sulfide liquid followed by subsolidus
transformations at all temperatures from the liquidus downward give an
adequate general description of the ore-forming mineral relationships. These con-
 116 v. V. Distler et al.

cepts are still more convincing since the bulk composition of sulfide aggregates in
many deposits indicates the stability of principal ore-forming components. God-
levsky, developing this notion of the formation of copper-nickel ores, came to
the conclusion of the inert behavior of sulfur in natural sulfide melts (Godlevskii
1967). The study of the zonality of ore bodies in the Norilsk deposits (Genkin et
al. 1981) led us, however, to the conclusion that parageneses that may be ascribed
to the inert-sulfur facies are not present at the crystallization of sulfide melts
composing them. All ore bodies display cryptic sulfide layering, manifesting
through successive changes of parageneses, reflecting the changing fugacity of
sulfur in a melt in the process of its crystallization differentiation. Early paragen-
eses are represented by products of low-sulfur monosulfide solid solutions, while
the later ones - by high-sulfur. The changing sulfur content of parageneses also
can be seen in the nature of fractionation of ore-forming metals in pyrrhotites
with various Me/S ratios and pentlandites of varied compositions (Fig. 3).
One more event contributing to the fractionation of components in copper-
nickel ore formation should be mentioned. Interpreting the zonal origin of the

Ni in Pn ( "10) A Ni in Pn(Ofo ) B

35

30 30

25 Ni in Pot Ofo) Ni in Po("Io)

,,~ 0.5 1.0 1.5 0.5 1.0

4B2 \'0,@
4B.O . ,,~ .1
(j)
47.B "~
\@.. ........ ••
-
-
- 0-
...... -
-
® c 2
""'-. +.......
.......00 --..--
_.-+ _0 -
0)
• 3
47.0 ....... @
00"0.........
47.4 @
--6-
®
L Me in Po (at %) -A-- ®
Fig. 3A, B. Nickel distribution between pyrrhotite (Po) and pentlandite (Pn), also in pyrrhotites with
various iron contents (IMe in Po, at. %) in stratified intrusives of various ore deposits (A) and in ores
of the Pechenga deposit (B), Circled numbers: 1 Maksut area (Eastern Kazakhstan); 2 - 5 intrusives
of the Norilsk ore region: 2 Lower Norilsk; 3 Zubovsk; 4 Talnakh; 4a massive ores of the Talnakh
intrusive; 5 Manturovsky; 6 Karik-Yarvi area of the East-Pechenga region. Numbers 1- 3 ore types
of the Pechenga ore region: 1 disseminated in peridotites; 2 brecciated; 3 massive
Sulfide Petrology and Genesis of Copper-Nickel Ore Deposits 117

ore bodies of the Norilsk deposits that are composed of pyrrhotites, cubanites,
and talnakhites, we have revealed that the most convincing proof of the phase
relationships of sulfides may be obtained if one assumes the existence of two
liquids in a sulfide melt - rich in copper and rich in iron and nickel.
We have obtained direct evidence of the possibility of such liquation events in
sulfide liquids. During the study of sulfide mineralization of oceanic basalts,
layered sulfide globules, consisting of sulfide solid solutions [iron-nickel (Mss)
and iron-copper (Iss)] were estabilished (Distler et al. 1983). Phase relations
(Fig. 4) witness that both solid solutions are the products of chilled crystallization
of two sulfide liquids with a distinct phase boundary between them. Distinct frac-
tionation of components (Table 1) takes place in these liquids with limited solu-
tion of copper in Mss and nickel in Iss. The chemical composition of one of the
liquids corresponds to nickel-pyrrhotite solid solution with the formula
[FeO.60Nio.19· (Co, CU)o.od SO.80; the second with the formula of the solid solution
(CUO.87Nio.13)1.oFe2.0S3.08 is close to the proportions of the components in
cubanite. It must be stressed that we are not aware of experimental data repro-
ducing the liquation in a sulfide melt of such composition. On the basis of the
analysis of natural sulfide assemblages, it could be assumed that the layering
takes place with relatively low concentration (fugacity?) of sulfur in sulfide
melts. In the Norilsk intrusions, we have observed, in particular, the layering of
troilite-mooihoekite globules. In zonal bodies of massive ores, the separation of
great sulfide masses occurs with mooihoekite (talnakhite, putoranite) as the prin-
cipal ore-forming minerals in one of them and troilite in the other.
The fact that the specific features of evolution resulting in these sulfide para-
geneses have been disclosed in occurrences of Mesozoic magmatism cannot be
overlooked. Variations in the composition of coexisting sulfides in Precambrian
deposits are more pronounced if different deposits are compared than is true
within anyone of them. Moreover, there are cases where the correlation in the
distribution of elements between such coexisting minerals such as pyrrhotites of
various sulfur content and pentlandites is not observed. One of the reasons here
is that ore metamorphism affects more strongly the pyrrhotite composition than
those at other coexisting phases. Similar data have been identified for deposits in
the Pechenga area (Fig. 3).
Examining the specific nature of ore formation associated with the evolution
of a sulfide liquid coexisting with a silicate melt, two stages in the formation of
the sulfide melt composition should be outlined. The first pertains to the equilib-
rium distribution of components between two liquids, which may be attributed to
the changed relationship of the sulfur and oxygen fugacity in the course of the
differentiation of the melts. The second, includes the consolidation of isolated
sulfide liquid coexisting with the solid-phase products of the silicate-melt crystal-
lization.
Comparative analysis made both for the Precambrian and Phanerozoic
deposits has shown that the interrelations between sulfide and silicate
parageneses are characterized primarily by the stratified distribution of sulfide
mineralization being in accord with the general layering of the massifs.
Cryptic layering becomes pronounced in the regular changes of the composi-
tion of rock-forming minerals - olivine, clino- and orthopyroxene, and plagio-
 118 v. V. Distler et al.

5
Fig. 4. Characteristic X-ray scans of a layered sulfide globule consisting of quenched copper-iron (Iss)
and iron-nickel (Mss) sulfide liquids. The boundary between the two quenched liquids and inflowing
of copper-iron liquid along the border with silicate matrix are most pronounced. I Reflected light
image; 2 copper Ka scanning image; 3 nickel Ka scanning image; 4 iron Ka scanning image; 5 cobalt
Ka scanning image; 6 sulfur Ka scanning image

clase - and is a well-known and very important feature of stratified intrusives.

Sulfide mineralization has been established also to feature cryptic layering,
reflected in the changed composition of coexisting pyrrhotites, pentlandites, and
chalcopyrites (Distler et al. 1979). Clearcut types of sulfide parageneses peculiar
to basites and ultrabasites have been identified. Parageneses of sulfur-deficient
sulfides are mostly developed in ultrabasites, while parageneses of sulfur-rich sul-
Sulfide Petrology and Genesis of Copper-Nickel Ore Deposits 119

Table 1. Chemical composition of sulfide solid solutions of layered globules from oceanic basalts a
(X-ray microprobe analysis, mass 010)

Phase Fe Ni Co Cu S Total

43.45 14.61 0.28 0.30 41.49 100.13

41.01 2.68 0.14 20.14 36.27 100.24

a Sample 6 -1, 112 -114, hole 505B (Deep Sea Drilling Project, Costa Rica Rift). Chromite-bearing
olivine basalt.

fides - in basites. Thus, the typical minerals of the former are troilite and sulfur-
deficient minerals of the chalcopyrite group (such as talnakhite and mooihoekite)
or hexagonal pyrrhotite with a low sulfur content, associated with chalcopyrite
and cubanite. Pyrrhotites with a high sulfur content and stoichiometric chalco-
pyrites are common for basite parageneses (Genkin et al. 1981).
In basite and ultrabasite horizons the distribution of ore components is
strongly dependent on temperature; for instance, that of nickel between sulfides
and olivines or orthopyroxenes as well as coexisting sulfides, corresponding
to general temperature parameters of basite and ultrabasite melts. In genetic
related series of sulfide-bearing rocks, the value of KI1~L-OL = (XNiSI XFeS)sL:
(XNiSio.sO/XFeSio.S02)OL increases from ultrabasites to the gabbroids. In com-
pletely differentiated intrusions of the Norilsk ore district (Distler and Genkin
1980) the value in sulfide-bearing pic rites is KI1~L-Ol = 5.4 and in sulfide-
bearing gabbroids accordingly 17.1. The value KI1~L-Ol does not depend on
the whole rock concentration of nickel in the melt. In differentiated melanocratic
intrusions (Distler and Genkin 1980) for the picrite-troctolite horizon, the value
of the distribution coefficient is also 5.5; although the whole rock concentration
of nickel in these rocks is nearly one order lower than is its content in picrites of
completely differentiated intrusions. The differences in whole rock concentra-
tions are reflected only in the nickel content of olivine. With equal magnesium
content of olivines (80 - 820/0 Fo), the nickel contents in olivines of completely
differentiated intrusions reaches 0.3 mass% and in differentiated melanocratic
intrusions 0.10-0.16 mass%. The value Kt;kL-Ol depends on the temperature of
the equilibrium of the sulfide liquid with coexisting melt. For the sulfide-bearing
picrites the value Ko = 5.4 corresponds to 1300° -1400°C and the value 17.1 cor-
responds to 1150°C. The latter value is very near the Ko value, established ex-
perimentally for Mss- Fo under 1100°C of 11.3 (Clark and Naldrett 1972). The
value increases up to 25.4 under 900°C and up to 39.9 under 700°C. The distri-
bution of nickel in coexisting sulfides and clino- and orthopyroxenes reflects the
same tendency established for the sulfides and olivines.
We stress one more aspect of sulfide-silicate interaction in magmatic melts. It
concerns the effect of the sulfide liquid on the succession of crystallizing phases
peculiar of the silicate melt. Many researches have shown that gravitational
events facilitate the parallel concentration of sulfides and olivines with the
emergent sulfide-rich peridotites in the bottom part of the intrusives. In the pro-
cess of sulfide-silicate interaction in magmas, there exist definite influences of the
immiscible sulfide liquid on the evolution pecularities of the silicate melt. The
120 v. V. Distler et al.
presence in the melt of S 2- serves as the factor of depolymerization and of
increase in the degree of ionization of the melt by means of connecting of part of
the petrogenic components, first of all iron, in the Me - S groups. This prevents
the formation of Me- 0- Si5bonds and, accordingly, results in the isolation of
the solid phases and favors more perfect differentiation of silicate melts in the
presence of sulfide liquid.
The depolymerizing role of sulfide liquid in sulfide-silicate melts also is felt in
the process of formation of mafic pegmatoids rich in sulfides that form isolated
discordant bodies that cut the intrusive stratification or individual.horizons of
the gabbroids of the pegmatoid or taxitic structure, localized nevertheless
concordantly with the general stratification of the massifs. Taxitic sulfide-rich
rt>cks of the completely differentiated Norilsk intrusives, ore-bearing pegmatoid
gabbroids of the bottom part of the Monchegorsk pluton, and others are
examples of such horizons. These are easily melted residues from fractional crys-
tallization of melts. However, the very fact of their emergence and abnormal
localization, infringing on the gravitational regularity of layered sequence forma-
tion, is indicative of specific interaction of components in magmas. Apart from
the sulfide liquid, no less important for the formation of these rocks is the fluid
phase of melts coexisting with sulfides. Parallel accumulation of sulfide and fluid
phases in products of crystallizing melts emerges as a decisive factor controlling
rock localization and inner structure.
The interaction of sulfide and fluid phases with silicate melts is also essential
for formation of some other very interesting pecularities of layered plutons, in-
cluding chromite-rich pegmatoid gabbros belonging to the upper gabbro series of
the Norilsk intrusives. Since the development of those rocks is of general signif-
icance, let us describe their inner structure, composition, and the specifics of ore
mineralization in more detail. Figure 5 shows the location of the pegmatoid gab-
bro horizon in the intrusive profile and the principal features of its inner struc-
ture. The major specific feature of the series is a greatly diverse petrography of
the rocks making it up. Three basic horizons may be identified: (1) contact gab-
bro-dolerites; (2) taxitic chromite-rich anorthositic gabbros; (3) prismatic-
granular gabbro-dolerites and diorites. Most interesting are the taxitic chromite-
rich gabbros in which the upper level of chromite, sulfide, and platinum mineral-
ization becomes established. The following zones are found in this horizon: Zone
I includes predominantly even granular (pegmatoid in places) chromite-rich an-
orthositic gabbros; Zone II - taxitic chromite-rich anorthositic gabbros with ir-
regular areas of troctolites, picrites, plagioperidotites, and clinopyroxenites. The
ultrabasic rock varieties are often rimmed with essentially anorthositic rocks. In
this zone, schlierenlike chromitite isolations are found, highest development of
sulfide and other ore minerals is observed, and a complex of secondary minerals,
including prehnites, biotites, talc, chlorites, carbonates, anhydrites, and others
are also widespread. Zone III has the least thickness within the horizon and is
composed mainly of even granular gabbros. It acts as a transitional zone to a
horizon made up of sulfide-free, chromite-free gabbro-dolerites and quartz-con-
taining gabbro-dolerites (Fig. 5).
Sulfide mineralization is represented by the common assemblage of minerals
such as pyrrhotite, pentlandite, chalcopyrite, and also assemblages of later ore
 Vl
rK rK rK Contact gabDro-dolerite E.
~
Q.
'"c r. " ;' .;
/
'"'"0
" !l
e:o /
Contact gabbro - dolente (r.). rT S I 3
..Q '\ Zone I
, '-, ",,''
..0 ..... ,. , g
no ," I ,
0> upper taxllic gabbro -dolero te Evengrained and pegmatold chromitebearing anorthite
\]j I , , / ,
(rTsl. prIsmatIc graIned gabbro Q.
=
'"
0..
'" I I '/1
I ",
0..
:::> gabbro - dolerIte ('i7/ 1 I,, 1\" '\ \ , og
/,' I , II/ &l
(;;'

' '"
"
1/ "
,
I
o-,
()
I \ " II' o
I ,
, -,/",\ ,'I I
'" ';'
'"'"'"
~ II... ' I , \ Cl!A...4t-tt1 Z
:I OIivinetree,ollvlOebear Ing. /,1 I o·
,.,.
Zone II !!.
olivine gabbro- dolente
Tax ilic anorthite gabbro with irregular putches of
...o
11I1I1I1IIIIII1 '"~
'" picrites.troclol ites. plagioperid olites,cli nopyrox enites '"
no
Q;
E (shaded). anorthosites (An). schlierlike segregations of
'"~.
., E
chromitites (Cr)
:- I Olivine - biotit-;-gabbro- doleriteJIl.fiftffttHJ~ ~ , I "
--- -- rOo. ~""G
P cI rote gabbro- dolerite.
, ,' I
tro ctolite, plaglopend olrte
, " ,\ "I,
1111RIBI: I , ","..! "
//,,/t ...
Zone III
~
,I , ,"I,
Evengrained and taxitic gabbro
rT" (T"
~ Lower taxi tic olivine gabbro- I E
..0 => , \ I , \', I ,"
0>
'" - doiellie (rTH l. evengrained
rTH .= --
rp rp ~~ \t \lj ~ Prismatic grained gabbro- dolerite.quartzbearing
0.. '"
'"•~ .,'"f':: gabbro (r p).contact gabb ro - II: I
o ., ....
--' '" dolerite (r. I rK \7! '\7; gabbro-dolerite
I. I Ore mineralization ~

Fig. 5. Profile and inner structure of the Upper gabbro series of the Talnakh intrusive (description see text)
122 v. V. Distler et al.

minerals pyrite, millerite, argentopentlandite, galena, sphalerite, arsenides,

and sulfoarsenides of nickel, associated with prehnite, biotite, chlorite, and
carbonate. Minerals of platinum metals - sperrylite, stibiopalladinite, merens-
kyte, mayakite, as well as platinum-palladium-containing nickelite and
maucherite, palladium-ruthenium-containing pentlandite, palladium-containing
millerite, and others are found in intergrowths with the sulfides and chromian-
spinels.
All specific features inherent in the structure and composition of the Upper
gabbro series and its ore mineralization emanate from the fluid-magmatic con-
vection that is essential for both rock formation and mineralization. Redistribu-
tion of petrogenetic and ore components against the background of normal dif-
ferentiation in a cooling melt in evidently the major factor responsible for the ab-
normal location of chromitites and ultrabasites in the upper horizons of a
stratified intrusive and the colocation of disequilibrium assemblages of primary
and secondary sulfides as well.
The rock types and the ore-bearing nature of the Upper gabbro series in the
Norilsk intrusives are very similar to those of the troctolite-anorthosite zone I
comprising the J - M Reef with sulfide, chromite, and platinum mineralization in
the intrusive Stillwater complex (Todd et al. 1982). The Upper gabbro series
differs from the above troctolite-anorthosite zone mainly by the morphology of
the rock bodies of definite petrographic composition. In the Norilsk intrusives,
these are multiple isolations of irregular shape; in the Stillwater, these are
continuous, regularly interbedded horizons. However, if we take into account
that the layering in the Stillwater complex develops within a rather large (up to
several kilometers thick) magmatic body with the cooling rate much smaller than
in the ribbon bodies of the Norilsk intrusives (hundreds of meters thick), then
these differences would look quite natural. What is important is that in both
cases similar complexes of rocks are developed that crystallized from melts of
various compositions, often accompanied by an abnormal distribution of
components within them. The latter is manifested, for instance, in the chrome
accumulation in the basite melt that favors the crystallization of chromite-rich
anorthosites. Later, mineral assemblages, comprising phases with volatile
components, are widespread in these horizons too. Such parageneses provide a
typomorphic character to these rocks since, in other horizons of intrusives, they
are either absent or poorly developed.
Specifics discussed above enable the conclusion that the origin of these rocks
may be associated with two interacting processes during the consolidation of
magmatic melts. The first is common differentiation of melts by separating solid
phases peculiar of the formation of layered intrusives. The second is the fluid-
magmatic convection responsible for the changed sequence of melt differentia-
tion and redistribution of components in a melt. The redistribution involves com-
ponents accumulated in residual liquids and components constituent of early
solid phases that dissolve partially while reacting with the residual melt. The
effect of such a melt on the olivine + chrome-spinel assemblage causes the
migration of its components to the upper horizons of the crystallizing intrusive
and the forming of local bodies of ultrabasite melts. The model proposed does
not require additional injections of hypothetical ultrabasite melt to explain
Sulfide Petrology and Genesis of Copper-Nickel Ore Deposits 123

the origin of rocks and mineralization of the horizons in question (Todd et al.
1982).
In conclusion, it should be stressed that joint analysis of the formation of
intrusive rocks and ores of the copper-nickel deposits bears out a close relation-
ship of ore formation with the whole complex of events attending the develop-
ment of sulfide-silicate magmatic melts from their origin in the zones of magma
formation to the final stages of their consolidation under hypabyssal-subvolcanic
conditions.

References

Cabri LJ (1973) New data on phase relations in the Cu - Fe - S system. Econ Geol 68:443 - 454
Clark T, Naldrett AJ (1972) The distribution of Fe and Ni between synthetic olivine and sulfide at
900°C. Econ Geol 67:939 - 952
Cox KG (1980) A model for flood basalt vulcanism. J Petrol 21 :629 - 650
Craig JR, Kullerud G (1968) Phase relations in the Cu - Fe - Ni system and their application to
magmatic ore deposits. In: Magmatic ore deposits. Econ Geol Monogr 4:344 - 358
Distler VV, Genkin AD (1980) Deposits of sulfide copper-nickel ores of the USSR and their connec-
tion with cratonnal volcanism. Proc 6th IAGOD Symp. Schweizer bart, Stuttgart, pp 275 - 295
Distler VV, Smirnov AV, Grokhovskaya TL, Filimonova AA, Muravitskaya GN (1979) Stratifica-
tion, cryptic layering and formation conditions of sulfide ore mineralization of differentiated trapp
intrusions. In: Formation conditions of magmatic ore deposits. Nauka, Moscow, pp 211- 269 (in
Russian)
Distler VV, Pertsev NN, Boronikhin VA (1983) Sulfide petrology of basalts from deep sea drilling
project holes 504B and 505B. Initial Rep Deep Sea Drill Proj 69:607 - 617
Genkin AD, Distler VV, Gladishew GD, Filimonova AA et al. (1981) Copper-nickel sulfide ores of
the Norilsk deposits. Nauka, Moscow, 234 pp (in Russian)
Godlevskii MN (1967) On differential mobility of the components in the formation of sulfide copper-
nickel ores. Geol Rudn Mest 9:17-31 (in Russian)
Kazanskii VI (1983) The evolution of the earth's crust and endogenic ore formation. In: Deep struc-
ture of ore districts, ore fields, and deposits. Nauka, Moscow, pp 5 - 24 (in Russian)
Kullerud G, Donnay G (1970) Chalcopyrite solid solution. Carnegie Inst Washington Yearb 69
Naldrett AJ, Craig JR, Kullerud G (1967) The central portion of the Fe - Ni - S system and its
bearing on pentlandite exsolution in iron-nickel sulfide ores. Econ Geol 62:826 - 847
Smirnov VI (1982) Endogenous ore formation in geological history. Geol Rudn Mest 4:3 - 20 (in
Russian)
Todd SG, Keith DW, Le Roy LW et al. (1982) The J - M platinum-palladium reef of the Stillwater
complex, Montana I. Stratigraphy and petrology. Econ Geol 77:1454-1480
Yund RA, Kullerud G (1966) Thermal stability of assemblages in the Cu-Fe-S system. J Petrol
7:454-488
Types and Distinctive Features of Ore-Bearing
Formations of Copper-Nickel Deposits
M. N. GODLEVSKyt and A. P. LIKHACHEV 1

Abstract

Ore-bearing formations and copper-nickel deposits are recognized as follows:

(1) Duluth-type gabbro-troctolite formation; (2) Norilsk-type gabbro-dolerite
formation; (3) Bushveld and Monchegorsk-type gabbro-norite-pyroxenite-peri-
dotite formation; (4) Pechenga-type gabbro-pyroxenite-peridotite formation; (5)
Kambalda- and Allarechensk-type pyroxenite-peridotite formation; (6) Mount-
keith-type peridotite-dunite; and (7) regenerated Subdury-type diorite-norite
formation. Nickel-bearing structures are noted in the greater thicknesses of the
Earth's crust and in downwarped Moho boundaries in weakly eroded provinces.
There is a clear relationship between magmatic thicknesses of continental
depressions and their ore potential: the thickness of volcanogenic units in ore-
bearing structures is over 3 km, whereas that of barren structures is less than
2 - 3 km. In undislocated areas, the relation between the morphology of intrusive
bodies and their ore potential is observed. Ore-bearing magmatic bodies are
typically elongated, band-chonolitelike bodies with flat roofs and downwarped
bottoms; this is attributed to the excess of density of ore-bearing magmas over
that of environment (at the expense of sulfide load) and gravity field effect
during intrusion of the bodies that brought about one-way movement of
magmatic masses.
Copper-nickel deposits are associated with mafic-ultramafic magmatic
formations that are divided into three series: (1) (Fig. 1 A, Table 1) mafic sulfide-
free and slightly sulfide-bearing series (MgO ~ 8 mass 070 ); (2) mesomafic sulfide-
bearing (MgO 8-33 mass%); and (3) ultramafic sulfide-free series (MgO >33
mass%). Each of the series is subdivided into magmatic groups that correspond
to magma varieties and produce magmatic rock complexes or geological forma-
tions that host various copper-nickel deposits.
The following ore-bearing (sulfide-nickel-bearing) formations are recognized
for copper-nickel deposits (Fig. 2): (1) Duluth-type gabbro-troctolite formation;
(2) Norilsk-type gabbro-dolerite formations; (3) Bushveld- and Monchegorsk-
type gabbro-norite-pyroxenite-peridotite formation; (4) Pechenga-type gabbro-
pyroxenite-peridotite formation; (5) Kambalda- and Allarechensk-type pyrox-

I Central Research Institute of Geological Exploration for Base and Precious Metals, Moscow,
USSR

Geology and Metallogeny of Copper Deposits

(ed. byG. H. Friedrich et aI.)
© Springer-Verlag Berlin Heidelberg 1986
 Types and Distinctive Features of Ore-Bearing Formations of Copper-Nickel Deposits 125

KM

100 2

200

300

A
400~------,-------~------~------.-----~~
800

Fig. 1 A-E. Types and conditions of formation of mafic-ultramafic magmas. AI, 2 Lines showing
solidus and liquidus of pyrolite with 0.1 weightOJo H 20 (After Wyllie 1971); 3 line showing solidus of
pyrrhotite (Sharp 1969, Kennedy and Ryzhenko 1973); 4, 5 inferred lines of mantle sulfide, solidus
and liquidus. I The area of melting of relatively low-temperature sulfide-free mafic magmas; II the
area of melting of sulfide-bearing magmas: IIo of relatively low-temperature magmas with Cu-rich
sulfides (Ni: Cu = 1 : 3 -4), lIb medium-temperature magmas that involve the whole sulfide fraction
of initial matter (Ni: Cu = 1: 1.2 - 2.5), lIe high-temperature magmas with Cu-poor sulfides
(Ni: Cu = 1 ~ 10 : 1); III the area of melting of sulfide-free high-temperature ultramafic magmas. To
the right are magma generation diagrams: for deep-seated mantle (B) and for asthenolith diapir (C).
Below (D and E) the state of iron-nickel-copper sulfide liquid is shown (white) in the oxide-silicate
melt of the Norilsk picritic gabbro-dolerite as observed in the experiments: D at 1500°C, E) at
1250°C, with 350 and 150 magnification, respectively

enite-peridotite formation; (6) Mountkeith-type peridotite-dunite, and regener-

ated Sudbury-type (Likhachev 1978b) diorite-norite formations (Table 1).
All sulfide-nickel-bearing formations and copper-nickel deposits now occur
within cratons (Fig. 3). They are located in superimposed continental depressions
filled with Mesozoic sedimentary and volcanogenic mafic-mesomafic trap rocks
126 M. N. Godlevsky and A. P. Likhachev

Table 1. Classification and setting of nickel-bearing complexes and copper-nickel deposits

Magmatic Magmatic Ore-bearing Morphology and Types of

series groups formations setting of formations deposits and
ore occurrences,
and their
examples

Mesomafic Weak magnesian Gabbro-troctolite Sill-shaped intrusions Kureika-type;

sulfide- group formation with in craton cover ore occurrences
bearing (MgO ~=~g Cu-rich (relative of Kureika dis-
MgO massOJo)a with to nickel) tric, Siberian
8-33 Cu-rich (relative platinum-nickel- Platform
massOJo to nickel) sul- copper ores
fides (Ni: Cu Large stratified Duluth-type;
1:2.5-10) massifs in craton Duluth deposits
basements (USA)

Moderate mag- Gabbro-dolerite Craton-associated N orilsk -type;

nesian group formation with elongated chonolith- Norilsk de-
(MgO 110~31 g) Cu-bearing pla- and bandlike gently posits, Insizwa
with Cu-bearing tinum-nickel- dipping bodies of (SAR)
sulfides copper ores craton cover
(Ni:Cu =
1: 1.2-2.5)

Magnesian group Gabbro-norite- Craton basement Monchegorsk-

(MgO It~:) pyroxenite- (infrastructures) Bushveld types;
with Cu-poor peridotite forma- associated large Bushveld and
sulfides (Ni: Cu tion with Cu- stratified intrusive Monchegorsk,
=1-2:1) poor platinum- complexes Great Dyke,
copper-nickel Stillwater, and
ores other deposits

Higher magne- Gabbro-pyrox- Sags (suprastruc- Pechenga-type;

sian group enite-peridotite tures) associated Pechenga de-
(MgO 210~N) formation with small interforma- posits; Cape
with moderate moderate Cu- tional bodies Smith-Weikhem
Cu-poor sulfides poor platinum- Bay deposits in
(Ni: Cu = 2: 1) copper-nickel Canada
ores

High magnesian Pyroxenite-peri- Greenstone belts Kambalda-type;

group dotite formation (suprastructures) greenstone belt
(MgO ~g=~~) with Cu-poor associated conform- associated de-
with Cu-poor platinum-copper ably dipping minor posits in
sulfides (Ni: Cu nickel ores bodies Western Aus-
= 2-20: 1) tralia, Canada,
Africa

Craton basement Allarechensk-

(infrastructures) type;
associated cross- Allarechensk
minor bodies deposits

a The numerator indicates variation limits of MgO content weighted mean; the denominator
indicates variation limits of MgO content in magmatic differentiates.
Types and Distinctive Features of Ore-Bearing Formations of Copper-Nickel Deposits 127

Table 1 (continued)

Magmatic Magmatic Ore-bearing Morphology and Types of

series groups formations setting of formations deposits and
ore occurrences,
and their
examples

Mesomafic Ultra magnesian Peridotite-dunite Greenstone belts Mount Keith-

sulfide- group formation with (suprastructures) type; Mount
bearing (MgO ~~:::~!) platinum-copper- associated cross-cut- Keith, Six-Mile
MgO with sulfides nickel ores very ting (dyke-shaped) deposits
8-33 very poor in Cu poor in Cu bodies (Western Aus-
massll!o (Ni : Cu = tralia)
20+60 : 1)

Mafic Hybrid low Diorite-norite Craton basement Sudbery type;

initially magnesian formation with associated large and Sudbery lopo-
sulfide- (MgO ri.t-
~2) ores poor and small bodies lith deposits,
free with sulfides very poor in Cu Elkinsk ore
(MgO ";;8 poor and very occurrence in
mass%) poor in Cu the Voronezh
(Ni:Cu = crystalline
1-10:1) massif

~
~0-540
2(}-6(J:I
1-"['.5-3

00:1 ~5-2"
~
5-20:1

20.1

10·1 20-25
/- .1S
15-20 2-5:r
/-40
10-15
H
2-30
MgOy'6-10
/-20
( :/ /Vl:Cu 12,5-/0
Pt:Pd ;:3,3

1-"5

NO
.:f 10 rS 20 30 MgO mas X

Fig. 2. The relationship of Ni - Cu ratio and magnesian content of the ore-bearing formations.
Diagram symbols: The numerator indicates variation of MgO limits of the mean weighted contents.
The denominator indicates alteration of the contents during magmatic differentiates; Ni: Cu and
Pt: Pd in sulfides
128 M. N. Godlevsky and A. P. Likhachev

.. + . + .. .. .
+ + . . . . . + .. +

! II
Fig. 3. Environments of nickel-bearing massifs and related copper-nickel deposits: I in superimposed
continental depressions of Mesozoic craton cover (Norilsk-type); II in Archean and Proterozoic
supracrustal terrains (deposits related to greenstone belts, the Pechanga structure); III in infrastruc-
tures: Archean granite-gneiss complexes of shield basement (Allarechensk-type); IV Archean and
Proterozoic large intrusive massifs (Bushveld-, Monchegorsk-types)

(Norilsk and Insizwa deposits); in supracrustal formations developed in Archean

and Proterozoic sedimentary-volcanogenic troughs and depressions super-
imposed upon the shield basement (deposits related to greenstone belts of
Western Australia, Canada, Africa, and the Pechenga structure of the Kola
Peninsula); in infrastructures, viz. Archean granite-gneiss complexes of the
shield base with sulfide-bearing ultramafic bodies (such as the Allarechensk
deposit); in large Archean- and Proterozoic-layered intrusive massifs related to
the craton basement, and in old cover (such as the Bushveld and Monchegorsk).
The fact that sulfide-nickel complexes and copper-nickel deposits are invari-
ably associated with continental cratons is attributed to peculiar features of
nickel-bearing magmatism: sulfide-bearing magmas generated at deep-seated
mantle horizons, with the thick continental crust playing the role of a heat screen
to account for the accumulation of heat energy and melting of sulfide-bearing
mantle substance at the expense of temperature increase (Fig. 1 B). This results in
generation of high-temperature melts were heavy sulfide liquid is present in a dis-
persed state (Fig. 1 D); this provides the manner of sulfide transport from the
deep-seated mantle into the upper-horizons of the Earth's crust in the less dense
oxide-silicate melt (thermal magma formation inherent in continental mag-
matism). In other cases, e.g., in decompressional magma formation (Fig. 1 C)
typical of oceanic, island-arc, and continental-rift type magmatism, sulfides of
initial matter, being incorporated into the melt, occur as relatively large segrega-
tions (Fig. 1 E). The main bulk of the segregations plunges into the deep-seated
mantle without reaching the upper horizons of the Earth's crust and its surface
(Likhachev 1978a, Godlevsky and Likhachev 1981). Temperature-dependent
variation of the mode of occurrence of sulfide liquid in oxide-silicate melt has
been demonstrated experimentally (Fig. 1 D) (Godlevsky and Likhachev 1981).
Most copper-nickel deposits are linked to effusive-intrusive complexes devel-
oped in superimposed continental depressions filled with sedimentary-volcano-
genic rock units. Some of the units are practically undislocated (Norilsk area)
Types and Distinctive Features of Ore-Bearing Formations of Copper-Nickel Deposits 129

others are slightly dislocated (Pechenga structure), still others are intensely dis-
located (Archean greenstone belts). Nickel-bearing magmatism produces bodies
of various shapes, dimensions, and ore contents depending on the erosional level
(Likhachev 1983).
In relatively weakly eroded structures, there is a relation between the
magmatic thicknesses and their ore potential. In cases in which effusive
magmatic complexes are less than 2 - 3 km-thick, there are practically no ore-
bearing magmatic bodies or copper-nickel deposits (traps of India, South
America, the southern Tunguska basin), whereas 3.5 -10 km-thick magmatic
complexes are ore-bearing sags in the Norilsk area and the Pechenga sag.
The relation between the thickness and ore potential observed in magmatic
formations is attributed to various melting degrees and transport to the surface
of initial matter that was incorporated into the melt as easily fusible to infusible
fractions (in all cases stage-by-stage). Sulfide-free and weakly sulfide-bearing
mafic magmas (MgO ~8 mass%) that slightly dissolve sulfur «0.1 mass%) are
formed first and transported toward the Earth's surface, then sulfide-bearing
mesomafic magmas (MgO 8 - 33 mass%) producing copper-nickel deposits are
formed (Fig. 1 A) (Likhachev 1973, 1978a). The content of sulfides in the initial
mantle material is estimated to be < 1 mass% (Likhachev, Krivtsov 1975). There-
fore, the production of sulfide ore-bearing magma (> 1 mass% sulfides) (God-
levsky and Likhachev 1983) requires a preliminary sulfide concentration that can
be performed through melting and transport to sulfide-free and slightly sulfide-
bearing mafic fractions from the initial material. As has been demonstrated by
experiments (Naldrett and Richardson 1967, Shamaraki and Clark 1973, Sharp
1969, Kennedy and Ryzhenko 1973) solidus temperature for silicates should be
lower than for sulfides (Naldrett 1973, Naldrett and Turner 1977, Likhachev
1973) (Fig. 1 A). The accumulation of sulfide matter and ascent of sulfide-
bearing magma to the surface depends on the volumes of transported mafic
material (no matter what their fractionation is: whether direct melting or dif-
ferentiation of more magnesian magma). This has a bearing on the thickness of
magmatic rock sequences in nickel-bearing sags. With the mafic magmas produc-
ing less than 2 - 3 km-thick magmatic rock sequences, conditions required for
generation, accumulation, and ascent of sulfide ore-bearing products to the
upper crust are not achieved. In this case, magmatism is restricted by intrusions
and outflow onto the surface mainly of sulfide-free and slightly sulfide-bearing
mafic magmas.
Nickel-bearing magmatic areas are noted for the higher thickness of the
Earth's crust (30-45 km), the downwarping of the Moho boundary, and a flat
bottom of sags. In contrast, nickel-free rift troughs are characterized by the
uplifted Moho boundary and steplike, en bloc structures.
The contrast is very conspicuous in the Norilsk area as well as in adjacent ter-
ritories with their typical sags, swell-like uplifts, and extended deep faults.
These sags are filled with basaltic effusive-intrusive complexes in thicknesses
of over 3 km. The downwarped structures in geophysical fields are outlined at the
levels of the craton cover boundary, granite-gneiss basement, and upper mantle
(Fig. 4). This suggests that they formed in the deep-seated mantle and that by
their nickel-bearing troughs are actually compensating structures that evolved as
 130 M. N. Godlevsky and A. P. Likhachev

N Yil arn Block Cool ardia SS IV

o~~~~~~~~~u==-~ = 5,0-6,2 _ v
6,0-5,2 +
+ +
6,~ - 6,~
eo + + + + +
+
+ + +
" " " 7,2-7,5 " " "
" " " " "
~~~ __ ~ ____ ~ __ ~ ____ ~ ____ ~ __ ~ L ____ ----~----~-- __- - - -__ ~~

o 200 400 600 800 loookm 0 so . 100 150km

Drummond 1981; Mathur at al. 1977 Crats et al. 1979

NW Norilsk SE NW Baikal SE
0 v v v + +
+
- 5,7-6,0 _ + + + +
+ +
20 6,1'+ 6,3 + + + + +
+ +

" " " "

6,5 - 6, 6 " "
40 " " " " I
6,1 I
/(",60 /
0 SO 100 ISO/rm 0 so lookm
Colubkov 1979 Logatchav 1984

Fig. 4. Seismic sections of nickel-bearing areas

a result of mantle material subtraction and subsequent formation of a depression

in the mantle horizon of magma generation. On the other hand, nickel-free struc-
tures with a slightly uplifted Moho boundary formed as a result of crustal tension
in the period of asthenolithic diapir rise acompanied by sulfide-free decompres-
sional magmatism.
There is a close relation between the morphology and ore contents of mag-
matic bodies. Thus, intrusive bodies of the Norilsk area are grouped according to
the morphology and ore contents into three categories (Fig. 5): barren, sill-
shaped bodies; weak ore-bearing, platelike bodies with swells and pinches; ore-
bearing, elongated band- and chonolithlike bodies with flat roofs and down-
warped bases.
Barren intrusions consist of undifferentiated and slightly differentiated
diabase and gabbro-diabase magmatic massifs that are comagmatites of the asso-
ciated basaltic cover that is composed of sulfide-free mafic magmas (MgO ~ 8
mass%) and their magnesian differentiates (MgO 15 - 20 mass%). The intrusions
are spread throughout the craton-cover section, giving rise to interlayered sub-
horizontal bodies in thicknesses of several centimeters to 250 m. The bodies ex-
tend for hundreds and thousands of square kilometers. Leukocratic gabbro and
taxitic-poikilophytic gabbro-dolerite sill-shaped intrusive bodies also are barren.
These bodies making up the frontal parts of ore-bearing intrusions are differ-
entiates of magmatic columns that lost their magnesian and sulfide components
Types and Distinctive Features of Ore-Bearing Formations of Copper-Nickel Deposits 131

?if"
"
;.

"
-/
;.

" "
;.

;.
;.
;.

.>
" ;.
;.

oa C>
b c

Fig. Sa-c. Morphological varieties of magmatic bodies in the Norilsk area: a ore-bearing; b weak
ore-bearing; c barren. Black patches at the base show areas of gravity tension in magmatic chambers:
a with magma density exceeding the density of the environment; b with magma density less than that
of the environment; c with the density identical for magma and environment

in the process of crystallization - gravity differentiation as the magmatic masses

rose to the upper crust (Likhachev 1965, Godlevsky and Likhachev 1979, 1983).
Slightly ore-bearing intrusions are of the Lower Talhakn type, being differen-
tiated gabbro-dolerite magmatic massifs localized in sediments, ranging in age
from the Permian Tunguss series downward to the Lower Devonian, and, in
thickness, from several meters to 150 - 350 m. Gradual transitions between dif-
ferentiates (from diorite and olivine-bearing gabbro-dolerite at the top to picritic
and taxitic gabbro-dolerite at the bottom) are typical. The total higher-magnesia
content (MgO weighted mean up to 18 mass%), low-chromium content
(Cr203 ~ 0.01 massOJo), and rather poor disseminated sulfide chalcopyrite-pent-
landite-pyrrhotite mineralization (1- 3 massOJo, Ni: Cu = 1: 1) can be traced
throughout the massif cross-section, the lower differentiates being somewhat
enriched.
Ore-bearing intrusions of the Norilsk type are strongly differentiated (ranging
from olivine-free at the top to picritic and taxitic gabbro-dolerite at the bottom)
massifs with copper-nickel sulfides (3 to 10 massOJo) concentrated at the bottom
of magmatic bodies as well as at exocontacts as disseminated or massive mineral-
ization. They occur at the same horizons as intrusive bodies of two other varieties
(barren and weak ore-bearing) displaying echelon structures; magnesium-,
chromium-, and sulfide-poor intervals alternate with intervals enriched with
these components. The latter make up copper-nickel deposits (Godlevsky and
Likhachev 1979, 1983).
The previously mentioned differences in morphology of the intrusive bodies
are not controlled by tectonics. This is indicated by the fact that weak ore-
132 M. N. Godlevsky and A. P. Likhachev

bearing and ore-bearing intrusions identical in time and emplacement differ in

morphology.
The reasons explaining the link between the morphologies of the intrusive
bodies and their ore potential are revealed by analysis of the generation and
intrusion of the magmas.
The pattern of magma transport into the Earth's crust and its surface
observed in the Norilsk area (from less to more magnesian varieties) suggests
fractionated melting of the initial matter and stage-by-stage generation of
magmas. This process starts from sulfide-free mafic (MgO ~ 8 mass%) magma
that gave rise to basaltic sequenes and comagmatic barren intrusions to sulfide-
bearing mesomafic varieties (MgO 8 -15 mass%) which account for weak
sulfide-bearing (1 % - 3% sulfides) and ore-bearing (3% -10% sulfides) intru-
sive bodies and related copper-nickel deposits. The main bulk of the sulfides
(over 90%) throughout the existence of magma (from the place of its generation
within the mantle to crystallization in the Earth's crust) occurred in the form of
liquid with -4 g cm -3 density, thinly dispersed in the oxide-silicate melt with
about 2.7 g cm -3 density (Godlevsky and Likhachev 1983).
The basic physical factor determining the form of the intruding magmatic
masses consisted in the differences of density values that largely depended on the
amount of sulfide liquid in the magma.
As is shown by experimental data and direct measurements taken in present-
day volcano lavas, the density of sulfide-free mafic magmas varied from 2.65 to
2.70 g cm -3. Weak sulfide-bearing magmas in the oxide-silicate portion pos-
sessed a somewhat higher density, and the presence of 1% - 3% sulfide liquid ac-
counted for about a 2.75 - 2.8 g cm - 3 density increase. Sulfide-bearing meso-
mafic magma that produced magmatic bodies approximated the density of the
oxide-silicate portion of weak sulfide-bearing magma. However, owing to the
large volume of 4 g cm -3 - density sulfide liquid (3% -10%), its total density
was high: 2.8 - 3.2 g cm -3. The amount if sulfide liquid in the intruding magmas
was determined in terms of weighted mean content of observed sulfide matter in
the magmatic mass of ore-bearing intrusions and dividing it into liquation (dis-
solving in silicate melt - -0.5 mass%) and segregation (surplus, unsolving frac-
tions). The accumulation of up to 10 mass% sulfides during the magma ascent
can probably be attributed to the descent of the sulfides, in the form of large
drops, from the cooled upper part of the magma column to its deeper and high-
temperature levels where sulfides became suspensed because of dispersion effects
(Godlevky and Likhachev 1983).
The influence of the density of magmas on the morphology of magmatic
bodies is attributed to the effect of gravity.
Mindlin (Roberts 1972) showed that tangential stress in the cyclindrical cavity
that occurs below the Earth's surface has minimum, though positive, values at the
uppermost point ¢J = nl2, and maximum values at the lowermost point ¢J = 3 nl2.
The same stress is maintained while the cavity is filled with fluid having a lower
density than the environment. On the other hand, the relation is reversed if the
density of the fluid filling the cavity is greater than that of the environment.
Since fracturing tends to occur at points ¢J = nl2 and ¢J = 3 nl2, the gravity
effect is found to limit such ruptures at the point ¢J = nl2 if the magma is lighter
Types and Distinctive Features of Ore-Bearing Formations of Copper-Nickel Deposits 133

than the environment, and at the point f/J = 3 nl2 if the magma is more dense
than the host rocks. Similarly, as fracturing related to shearing also tends to
occur at the same points, the chamber with a relatively lighter magma will be
fractured at the point where f/J = 3 nl2, and the chamber with a heavier magma,
at the point where f/J = nl2.
So the upper portion of the chamber with the relatively dense magma (higher
density than the environment) will be compressed, while the lower part will be
extended. This latter phenomenon accounts for the increase of the magmatic
volume downward and formation of magmatic bodies with flat roofs and down-
warped bottoms. On the other hand, in the chamber where density is less than
that of the host rocks, the extension of the environment and expansion of the
chamber will occur in its upper portion, thus, giving rise to convex bodies. The
density of the magma and host rocks being identical, the stress at the roof and at
the base of the chamber is balanced. In such cases, at the flanks of the chamber,
at points of minimum tangential stress, the negative difference between the
density of magma and host rocks and positive hydrostatic pressure are bound to
occur, thereby providing lateral increase of the chamber volume. Sill-shaped
bodies owe their formation to this phenomenon.
The revealed pattern of relations between the densities of the fluid and of its
environment, as well as gravity field affecting the stress and shape of forming
chambers, explain fairly well the differences in the morphology of intrusive
bodies in the Norilsk area (Fig. 5).
Sulfide-free magmas by their density (2.65 - 2.7 g em - 3) were equivalent in den-
sity to their host rocks (2.5 - 2.8 g em -3), thereby giving rise to environments most
in the energy needed to produce sill-shaped bodies. Ore-bearing magmas exceeded
the density of the environment (2.8 - 3.2 g em -3) through effect of the sulfide load,
so their extension was determined by the influence of gravity that provided "a chan-
nel" in the lower part of the chamber and one-way movement of magmatic masses,
including the formation of elongated chonoliths with downwarped bottoms and flat
roofs. On the other hand, weak ore-bearing magmas, characterized by intermediate
density, accordingly gave rise to magmatic bodies of intermediate morphology.
Roberts (1972) demonstrated that the movement of magma was provoked by
minimum tangential stress and positive hydrostatic pressure at the frontal part of
the magmatic column. The latter acted as a wedge, the transverse stress of which
affected host rocks five times as much as the longitudinal, stress produced by
magmatic pressure.
The aforesaid suggests the following conclusions:
1. Sulfide-nickel-bearing formations are associated with continental thermal
magmatism that provided for the ascent of heavy sulfide liquid out of the deep-
seated mantle to the upper parts of the Earth's crust and formation of copper-
nickel deposits.
2. Nickel-bearing structures are characterized by higher thicknesses of the
Earth's crust and downwarped Moho boundaries in contrast to nickel-free struc-
tures that are generated as a result of the tension in the Earth's crust caused by
mantle diapirs and decompressional magmatism.
3. There is a clear relation between the thickness of magmatic complexes asso-
ciated with continental depressions, and ore potential, viz., in ore-bearing struc-
134 M. N. Godlevsky and A. P. Likhachev: Types and Distinctive Features

tures the thickness of volcanogenic units is over 3 km, whereas in barren struc-
tures, it is under 2 or 3 km, this being attributed to the variable degree of melting
of the initial mantle matter: in barren structures magmatism produces mainly
sulfide-free mafic magmas (MgO ~ 8 mass%), whereas in ore-bearing structures
it produces mafic and sulfide-bearing mesomafic magmas that give rise to
copper-nickel deposits.
4. Ore-bearing magmatic bodies in undislocated areas are elongated, band-,
chonolith-like bodies with flat roofs and a downwarped bottom, which is attri-
buted to the excess of ore-bearing magma density greater than that of environ-
ment, and gravity fields affecting their intrusion and providing one-way move-
ment of magmatic masses.

References
Drummond BJ (1981) Crustal structure of the Precambrian terrains of North-West Australia from
seismic refraction data. BMR J Aust Geol Geophys 6:123 -135
Godlevsky MN, Likhachev AP (1979) Generation and crystallization of ore-bearing magmas produc-
ing copper-nickel deposits. In: Basic parameters of endogenetic ore formation, vol I. Nauka,
Novosibirsk, pp 109-118 (in Russian)
Godlesvky MN, Likhachev AP (1981) Formative conditions and evolution of ore-bearing ultrabasic
magmas. West. All-Union Miner Soc Iss 6:646-655 (in Russian)
Godlevsky MN, Likhachev AP (1983) Copper-nickel ore formation in the Norilsk area. In: Genetic
model for endogenetic ore formations, vol I. Nauka, Novosibirsk, pp 47 - 54 (in Russian)
Kennedy JR, Ryzhenko BN (1973) The influence of the pressure on eutetics in Fe- FeS systems.
Geochimia 9 (in Russian)
Kratz KO et al. (1978) The Earth crust of the eastern Baltic Shield. Nauka, 232 pp (in Russian)
Likhachev AP (1965) The role of leucocratic gabbro in the formation of Norilsk differentiated intru-
sions. Izv Akad Nauk SSSR Geol Ser 10:75 - 88 (in Russian)
Likhachev AP (1973) On the nature of magmatic deposits. SOy Geol 5:33 - 47 (in Russian)
Likhachev AP (1977) Magmatism and nickel potential in North-Central Siberia. SOy GeoI2:30-45
(in Russian)
Likhachev AP (1978a) Formative conditions of ore-bearing and barren mafic and ultramafic
magmas. Rep Akad Nauk 238, 2:447 - 450 (in Russian)
Likhachev AP (1978b) On the genesis of the Sudbury copper-nickel deposits. SOy Geol 6:60 -71 (in
Russian)
Likhachev AP (1983) Geology and classification of copper-nickel deposits. West. All-Union Miner
Soc Iss 1:14-27 (in Russian)
Likhachev AP, Krivtsov AI (1975) Sources and possible ways of ore matter concentration. SOy Geol
5:69 -79 (in Russian)
Logatchev NA (1984) The Baikal rift system. Episodes 7,1:38-42
Naldrett AG (1973) Nickel sulfide deposits. Their classification and genesis, with special emphasis on
deposits of volcanic association. Trans Can Inst Min Metall 76: 183 - 201
Naldrett AG, Richardson SW (1967) Effect of water on the melting of pyrrhotite-magnetite
assemblages. Carnegie Inst Washington Yearb 66:429-431
Naldrett AG, Turner AR (1977) The geology and petrogenesis of a greenstone belt and related nickel
sulfide mineralization of Yakabindie, Western Australia. Precambrian Res 5:43 -104
Roberts J (1972) Magma intrusion in loose rocks. In: Mechanism of magma intrusion. Mir, Moscow,
pp 230-283
Shamazaki H, Clark LA (1973) Liquids relations in the FeS-FeO-Si0 2 -H 20 system and
geological implications. Econ Geol 68, 1:79 - 96
Sharp WE (1969) Melting curves of sphalerite, galena, and pyrrhotite and the decomposition curve of
pyrite between 30 - 65 kilobars. J Geophys Res 74, 6
Wyllie PI (1971) Experimental limits for melting in the Earth's crust and upper mantle. Geophys
Monogr Ser 14:279- 301
On the Role of Metamorphism in the Formation of
Nickel-Copper Sulfide Deposits in the Kola Peninsula
Yu. N. YAKOVLEV and A. K. YAKOVLEVA 1

Abstract

Three stages are apparent in the formation of the nickel-copper sulfide deposits,
viz. magmatic, regional metamorphic, and late metamorphic. In the magmatic
stage, syngenetic ores are simple and constant in the composition formed. In the
regional metamorphic stage, epigenetic ores appear that are distinguished by
diversities in mineral composition, typical of the specific tectonic zones formed
during such metamorphism. Late metamorphic alteration is confined to local
tectonic zones and is characterized by the inconstancy of the mineral assem-
blages. In the process of metamorphism, the chemical composition of the miner-
alized rocks experiences less alteration that results in a reduction of the geo-
chemical contrast of their separate varieties. All metamorphic alteration events
took place within the primary ore bodies and did not result in the formation of
new ore shoots. Thus, the Ni - eu sulfide deposits of the region can be classed as
metamorphosed magmatic deposits rather than metamorphic deposits.

Introduction

Nickel-copper sulfide mineralization of the Kola Peninsula occurs in the massifs

of mafites and ultramafites belonging to five magmatic groups (Gorbunov et al.
1982): websterite-gabbro-norite, gabbro-lherzolite-pyroxenite, ultrabasite, peri-
dotite-pyroxenite-gabbro-norite and gabbro-wehrlite. The first two groups
(Smirnov 1982) belong to the Kola upper Archean metallogenic stage and the
remainder to the Karelian (Lower Proterozoic) stage. Several nickel provinces are
distinguished (Fig. 1) and in each, the massifs of one group are well-developed.
Thus, ore-bearing intrusions of the Kola stage are concentrated in the
Lovnoozero province. The early Karelian is represented by massifs of ultrabasite
in the Allarechka province and peridotite-pyroxenite-gabbro-norite in the East-
Pechenga (Luostari massif), Monchegorsk (Monchegorsk pluton), and Imandra-
Varzuga (Fyodorova and Panskie tundra massifs) provinces. Late Karelian

1 Institute of Geology, Kola Branch of the USSR Academy of Sciences, Apatity, Murmansk Region,
184200, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G.H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
 w
~I 0\
-
lZ2J2
c=JJ
[1 1:114
~5
~6
[ZIJ7
CJ8
1-:- -:- -:-19
I @ 11 0
~II
I" "1 12
-<
~ /J ~
:z
k--1/4 -<
~
o
25 0 25 50l(M S-
, , ! , o
<
~
::l
0-
Fig. 1. Geological map of the Kola Peninsula. (After Yakovlev et al. 1981) 1 Rifes terrigenous complex (sandstones, conglomerates, schists); 2 - 4 ~
Karelian volcanogenic-sedimentary complex (2 South-Pechenga and Tominga series; 3 Pechenga and Varzuga series; 4 Strelna series); 5 -7Kola- ~
Belomorian amphibolite-gneiss-crystalline schist complex (5 Keivy schist series; 6 Lebyazhka, Voronyetundra series and their analogs; 7 Vol- -<
~
chelambino, Olenegorsk, Annam series, and their analogs); 8 oldest complex of primary sialic crust (plagiogranites, tonalites, plagiogneisses); ...
o
9-13 intrusive formations (9 nepheline alkaline syenites; 10 alkaline gabbroids; 11 alkaline granitoids; 12 mafic and ultramafic rocks; 13 grani- <
toids); 14 most important fractures. Circled numbers: nickel areas (1 Pechenga; 2 Allarechka; 3 Lovnoozero; 4 East-Pechenga; 5 Monchegorsk; ~
~

6 Imandra-Varzuga)
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 137

examples include intrusions of gabbro-wehrlite in the Pechenga and Imandra-

Varzuga provinces.
Detailed geological characteristics of the provinces, the structure of the
deposits, the petrology of the mafite-ultramafites, and the ore mineralogy are
available in a number of general references (Gorbunov 1968, Gorbunov et al.
1978, Kozlov 1973, Yakovlev et al. 1981). Thus, only brief information on these
questions is given in the present paper.
The Pechenga province lies in the NW part of the Pechenga-Varzuga tectonic
and metallogenic zone which extends throughout the Kola Peninsula (Fig. 1).
The province is made up of volcanogenic-sedimentary rocks of Pechenga strata
including four volcanogenic sheets divided by masses of tuffogenic-sedimentary
rocks. Nickel-bearing mafic-ultramafic intrusions are concentrated in the fourth
tuffogenic-sedimentary mass that is designated as the productive mass and fre-
quently contains pyrite-pyrrhotite mineralization.
The massifs have blanket-, lenticular, and, occasionally, phacolithlike shapes
outlining the contours of large folds in the enclosing rocks. They dip SE - SWat
30° - 60°. The massifs extend for 100 - 6000 m along the strike, have a
100-1500 m downward dip, and are from 10-15 m to 200- 300 m thick. Major
deposits are associated with differentiated massifs in which the following rock
types occur from bottom to top: amphibole-chlorite (near contact), serpentinite
(after peridotite and olivinite), serpentinized peridotite (wehrlite), pyroxenite,
and gabbro. At the western flank of the area, serpentinite and serpentinized peri-
dotite compose up to 700/0 of the massif section, at the eastern flank gabbro is
predominant. The mineralization is confined throughout to the lower parts of the
massifs and is represented by low-grade disseminated and high-grade im-
pregnated ores in the serpentinite, by breccia and massive ores in the interstratal
tectonic zones extending along the lower contacts of the massifs with the tuf-
fogenic-sedimentary rocks, and by stringer-type impregnated ores in the exo-
contact phyllite (Fig. 2a, b).
The Allarechka province is made up of Archean amphibolite-gneiss masses.
Block-faulted structures and a wide development of domelike structures are com-
mon. Numerous small massifs are known in the area. They occur roughly con-
cordantly with the enclosing gneisses and amphibolites, frequently containing
disseminated pyrrhotite mineralization. The massifs are located in the marginal
parts of the domelike structures and in the interstices between them, frequently in
groups of 5 -10 massifs and in several stages. They are 100 - 2000 m along the
strike and 5 - 200 m thick. Most of these were traced for 100- 200 m and in-
dividual massifs for up to 1000m. The massifs are mainly of blanket, lenticular,
and occasionally bandlike shapes. Many massifs are not markedly differentiated:
levels of olivinites, peridotites (harzburgites), and pyroxenites are distinguished
in them. As a rule, they are cut by the veins of granite pegmatites up to 20 - 25 m
thick. Economic mineralization was contained in only several massifs
(Allarechka and Vostock deposits, Severnoe and Runnijoki ore showings).
Disseminated ores are predominant in the ultramafites of lower and medium
parts of the sections; massive and breccia ores are concentrated in the tectonic
zones. At the Allarechka deposit, such a zone extends along the western flank
and contains mainly massive ores and at the Vostock deposit, the ore zone ex-
138 Yu. N. Yakovlev and A. K. Yakovleva

s
lent IX

'"" '\....
,
\., .... -.... '-
"~, ..... '-
'\...
,, I.. "\...
~
'-
,
~

'\...
'- ......
\...

\.:.', , 'l....
'- '" .....
....
~' ... '-
'-
'" "-
"\...

"I
1 ",
- 'f
~\-....

~\
-,- I
I \ I

b .
1"( ,/,lf I(~ '<12 ~3 1.11. .n.14 @I,s 11116 [Zj 7
a.
~ 4 ~g
~4~ 1'•';..:'''110
••

Fig. 2a, b. Transversal geological sections of the Flangovoye deposit (a) and the Eastern ore cluster
(b) (Gorbunov et al. 1978). 1 Effusive diabase; 2 gabbro-diabase; 3 rocks and ores of differentiated
intrusives; 3 gabbro; 4 pyroxenite; 5 serpentinite (0 ore-free; b with low-grade mineralization); 6 high-
grade impregnated ore; 7 breccia and massive ores; 8 tuffogenic-sedimentary rocks (0 ore-free; b min-
eralized); 9 tectonic dislocations; 10 quarternary sediments

tends along the lower contacts of the massifs and contains principally breccia
ores (Fig. 3 a, b); stringer-type mineralization is frequently marked in zones of the
exo-contacts of the high-grade ores.
The Lovnoozero province lies in the SW part of the belt of Lapland
granulites. Two complexes of rocks took part in its formation: (1) granulite (it is
predominant in the southern part of the area) and (2) gneiss (it is predominant in
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 139

s N

b
.lYw

\
\
\
\
\
\
\

c
Fig. 3 a-c. Transversal geological sections: a Allarechka deposit; b Vostock deposit; c Lovnoozero
deposit (Yakovlev and Yakovleva 1974, Zak et at. 1972, Gorbunov et at. 1978, respectively). 1 Gneiss
and granite gneiss; 2 feld-spathic amphibolite; 3 ore-free peridotite; 4 peridotite with low-grade min-
eralization (ore type I); 5 peridotite with high-grade mineralization (ore type II); 6 breccia and
massive ores (ore types III and IV); 7 tectonic dislocations; 8 quaternary sediments; 9 hypersthenic
plagiogneiss and granulitelike rocks; 10 norite and gabbro-norite with low-grade sulfide mineraliza-
tion; 11-13 mineralized norite and gabbro-norite with nickel and copper content (total): 110.3 -1.0
wtDJo; 121.0- 2.5 wtDJo; 13 more than 2.5 wtDJo
140 Yu.N. YakovlevandA.K. Yakovleva

the northern part). A great number of small massifs of mafites and about a dozen
ultramafites were defined there. All massifs are of lenticular-, cigar-, and band-
like shapes. They occur concordantly and rarely discordantly with enclosing
rocks. The massifs are 100-1000 m (sometimes up to 3000m) in strike length;
and 10-100 m (rarely 300-600m) thick. They were traced for 100-2oom
(seldom for 500 - 600 m) in a downward dip. The massifs of mafites and ultra-
mafites are frequently cut by veins of granite pegmatites with a thickness of
0.5 -100 m. Mafic rocks are represented by norites and gabbro-norites; ultra-
mafic - by pyroxenites (websterites) and by peridotites (lherzolites). The signs of
differentiation are marked both in the mafic massifs and the ultramafic ones.
Sulfide mineralization is mainly associated with mafic massifs (Lovnoozero
deposit and a number of small ore showings) and rarely with ultramafic (Suein-
lagash ore showing). Disseminated ores develop everywhere. They are irregularly
distributed in the mafites (Fig. 3 c) and tend to reach the levels of olivine-free
rocks close to the flat wall of the ultramafic massifs.
The Monchegorsk province is in the central part of the Kola Peninsula
(Fig. 1) at the site of the sharp turn of the Pechenga-Varzuga structural zone.
Nickel-copper sulfide mineralization developed in the large Monchegorsk pluton
(about 55 km2 in area) that consists of a peridotite-pyroxenite-gabbro-norite
formation. The pluton is interformational and occurs between Archean gneisses
and Proterozoic volcanogenic-sedimentary rocks. In plan, the pluton is arched;
its sublatitudinal branch includes the Sopcha and Nyud-Poaz massifs, and the
submeridional branch includes the Nittis-Kumuzhiya-Travyanaya massifs
(NKT). The NKT and the Sopcha massifs are composed mainly of ultramafites
forming a layered series, and the Nyud-Poaz massifs are also composed of layer-
ed mafites.
Nickel-copper sulfide ores are represented by several morphological types;
the major types are disseminated, nestlike impregnated, and veined. Dis-
seminated ores are confined to the levels of the olivine-bearing rocks in the upper
parts of the massifs ("beds" of Sopcha and Nyud) or of feldspathic rocks in the
lower parts (bottom deposits of Sopcha and NKT massifs). Nestlike impregnated
ores are localized in areas of nonuniform structure (the so-called critical level of
Nyud). Veins of massive sulfide ores (often with magnetite) developed in the
median parts of the NKT and Sopcha massifs; they are spatially associated with
diorite-pegmatite.
The knowledge available on the East-Pechenga and Imandra-Varzuga
provinces is very slight and is not dealt with, but it is noteworthy that in the East
Pechenga province well-defined massifs of mafites-ultramafites are divided into
three groups, viz. peridotite-pyroxenite-gabbro-norite (Luostari), ultrabasite (a
number of small massifs), and gabbro-wehrlite (Nyasyukki dikes), whereas in the
second province the massifs of mafites-ultramafites of two groups were establish-
ed, viz. peridotite-pyroxenite-gabbro-norite (Fyodorova and Panskie tundras)
and gabbro-wehrlite (small massifs in the tuffogenic-sedimentary rocks).
Nickel-bearing massifs of the first three provinces have been metamorphosed
together with the enclosing rocks over a wide range of P - T conditions corre-
sponding to greenschist (Pechenga), amphibolite (Allarechka), and granulite
(Lovnoozero) facies of regional metamorphism. This corresponds to a range of
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 141

temperature from 3()()O - 350 °C to 700 0 -715 °C and of pressure from 2 - 3 to

5 - 7 kb_ The rocks of large massifs of peridotite-pyroxenite-gabbro-norite
formation (Monchegorsk and other provinces) are the least altered, although
autometamorphic alteration and other alterations related to contacts, veins, and
dislocations are also present. Thus, magmatic, metamorphic, and hydrothermal
processes are reflected in the formation of Ni - Cu deposits with the relative
importance ranging widely through the different tectonic zones (Fig.4). The
relationship between syngenetic and epigenetic mineralization types also differs
in the same manner. The majority of workers correlate the development of the
epigenetic ores with the regressive metamorphism or with the transition from the
progressive to the regressive substages of metamorphism.
Concepts of the formation of Ni - Cu epigenetic ores from Pechenga,
Allarechka, and Lovnoozero provinces in the process of regional metamorphism
have developed over a long period. Some workers believe that under meta-
morphism, primary magmatic sulfides were altered and redeposited in tectonic
zones as breccia and massive ores (Kozlov 1960, Likhachov 1974 and other
authors). The others think that primary magmatic sulfides were absent or were
very small in amount and the major part of the ores formed in the process of
metamorphism at the expense of Ni and Fe liberation from primary silicates with
the sulfur and copper supplied from the enclosing rocks (Predovsky et al. 1968,
Zak et al. 1972). Some researchers, basing their ideas mainly on experimental
data, connect the formation of epigenetic ores with sulfurization processes under
hydrothermal conditions (Arutyunyan et al. 1978).
On the whole, the majority of supporters of the metamorphogenetic forma-
tion of the Ni - Cu ores believe that their positional connection with mafites-
ultramafites is paragenetic. On the one hand, these authors believe that these
rocks were the metal source and that sulfur came in from the enclosing rocks or
was supplied by fluids from deep sources. On the other hand, these rocks
provided a favorable environment for ore deposition.
Some of the distinctive features of the metamorphic ores are as follows:
higher Cu and S content, variability of mineral composition, and poor correla-
tion between ore components. We shall consider several of these features, desig-
nating the following ore types: low-grade disseminated and high-grade impregna-
tions in mafites and ultramafites (I and 11); breccia and massive ores in tectonic
zones (III and IV); stringer-type ores impregnated in exo-contact metasomatic
and enclosing rocks (V and VI). Ore types I and II are always predominant; ore
types III and IV are significantly less abundant, and ore types V and VI are rare.
The low-grade disseminated ores and parts of the high-grade impregnated and
massive ores are purely syngenetic, whereas the remainder are epigenetic
(Gorbunov et al. 1973 and others). Besides, data on mineral and chemical ore
composition are represented by an average for a specific ore type or deposit and
are calculated on the basis of a great number of ordinary determinations, i.e.,
quantitative mineralogical calculations of thin and polished sections (tens of
samples) and the results of sampling (many tens and low hundreds of ordinary
samples).
 NJ,co
~ Cast. clustr;,. AUarechka ./ Ni/EMe Vostok
;::, <3 ~ ~ Flan~ovoe --~ ~
~1~1.~j~ I> N!:!.IMe ~cu
1M 7 '< 'x //
""'. / '\.
~

II '''':i/Co __o
/
~
~ I /j'x=x=><=o
~
//
/ ~ /1')(
07 0 Ni/tiJ //' ----.,
d
'x..."O--x->e--ONils 1/<
d'/

f-l20~a2-

I II 11I I II JI[
~8~~/ NKT
~1~i~l~ ?

~
-~ Ni/CU
~ ~x, "'<l:: ?!!
o----c.... ~ ~ EMe/S ><
"><,"I - Nl'CO ~x~ _~ ~
N£'''''
'-x=x~" " ....... 0 I"" x.......xx_ ~---x
)Ci i
--0 0-' - - - ~)('0Ni/S 0-' - - """"--0
o--:.c_~'x£,Me/S ~~~ 8.
-)(~
o-x__ --o-- ?>
Ni/S x-x-o
~
][ I
><
I DI I II DI I 1/l m ~
Fig. 4. The relationship between major components in Ni - eu ores (I low-grade disseminated; II high-grade impregnated; III breccia
and massive) 1
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 143

Mineral Composition of the Ores

The paragenesis of the mineral assemblages during deposit formation shows a

common sequence (trend) everywhere. Qualitatively each assemblage is present
in each area, but there are conspicuous differences in the lmportance of each
assemblage in specific areas. Three stages are defined, viz. magmatic, regional
metamorphic, and late metamorphic (hydrothermal), with each having its sub-
stages (Yakovlev and Yakovleva 1974, Yakovlev et al. 1981).
The magmatic stage involves: (1) the crystallization of primary silicates and
oxides (olivine, pyroxenes, plagioclase, titanomagnetite, chrome-spinel - early
magmatic substages); (2) early amphiboles and micas, monosulfides of syn-
genetic ores (troilite, hexagonal pyrrhotite, pentlandite, chalcopyrite, cubanite -
late magmatic substages); and (3) then a partial transformation of both
(serpentine, talc, early magnetite - autometamorphic substage).
The regional metamorphic stage is characterized by an intensive replacement
of the primary silicates by serpentine, Ca-amphibole, Fe - Mg micas and
chlorites, and by the appearance of late generations of major sulfides and
numerous subordinate ore minerals (regional metamorphic substages proper).
Subsequently, Fe-Mg amphiboles (anthophyllite, cummingtonite), biotite, talc,
scattered disulfides, and rare ore minerals (substage of dislocation meta-
morphism) developed at the contacts with enclosing rocks, in veins of granitic
pegmatites near mineralized fractures, and in zones of foliation.
The assemblages of the late metamorphic stage are confined to tectonic
breccia zones; these are carbonates, late serpentines, chlorites, and talc, second-
ary sulfides, rare sulfates, hydroxides, and oxides. They are characterized by
intensive alteration of all rocks and ores in specific localities.
This scheme is typical of the Kola Peninsula as a whole, but in specific areas
the mineral assemblages of separate metamorphic events are noticeably different
(Table 1). All metamorphic transformations are more pronounced in the ultra-
mafites than in mafites.
The extent of metamorphic alteration decreases from Pechenga through
Allarechka to Lovnoozero. Alteration affects the entire bulk of the massifs, but
neocrystallization is generally restricted, affecting up to 10- 15 volume % of the
ultramafites and 5 - 7 volume % of the mafites. Mineralized rocks preserve both
primary structures and textures and the main features of the mineral composi-
tion, which are characterized by constant assemblages and properties of the main
minerals determined by their parent intrusives and the total chemical composi-
tion of the ore.
With increasing metamorphic grade, the following change in main assem-
blages takes place: actinolite-chlorite-serpentine is typical of the Pechenga area,
phlogopite-magnesio-hornblende is typical of the Allarechka area, and biotite-
hornblende (pargasitic, tschermakitic) of the Lovnoozero area.
The metamorphic transformations affect the massifs, as a whole, both ore-
free and ore-bearing, but these are more intensively developed (especially amphi-
bolization) in their endo-contact zones (up to 1 - 2 m thick). All stages of succes-
sive transformation of primary sulfide disseminated ore up to the formation of
"negative" (Gorbunov 1968) and subgraphic textures are markedly developed in
144 Yu.N. YakovlevandA.K. Yakovleva

Table 1. The sequence of metamorphogenic mineral formation in ore impregnations, NW Kola

Peninsula

Areas Major silicates in the ultramafites Major silicates in the mafites

A utometamorphism
All areas Serpentine, talc, hastingsitic Talc, hastingsitic,
hornblende hornblende, serpentine
Regional metamorphism
Regional-metamorphic substage
Pechenga Serpentine, chlorite, actinolite Chlorite, actinolite
Allarechka Magnesiohornblende and actino-
litic hornblende, phlogopite,
chlorite
Lovnoozero Pargasitic and edenitic Tschermakitic hornblende,
hornblende, phlogopite biotite
Substage of dislocational metamorphism
All areas Anthophyllite, cummingtonite,
phlogopite, actinolite, talc
Late metamorphism
All areas Carbonates, chlorites, talc,
serpentines, micas, actinolite

the ore-bearing massifs. As a result, in many ultramafic intrusives, the magmatic

structures and textures are virtually replaced by metamorphic ones and by recrys-
tallization and partial redeposition of the syngenetic sulfides. However, there are
relics of the primary mineralization as "shadow textures" , droplike and emulsion
inclusions of early ore minerals even in the highly altered rocks. As a rule, sulfide
redistribution is virtually confined to the volume of the original ore bodies and,
on the whole, the mineralization grade of individual sites in the massifs does not
depend on the intensity of metamorphism of the rocks composing them.
The total change of mineral composition during metamorphism is con-
sidered, taking as an example the impregnated ores from the northwestern part
of the Kola Peninsula (Table 2). These ores are the most abundant of all deposits
and are most affected by metamorphic processes. Only the most abundant
primary and secondary minerals that define the composition of the impregnated
ores are considered. For comparison, Table 2 shows the composition of the least
altered impregnated ores from the Monchegorsk pluton (Sopcha and NKT
deposits). The mineral compositions of the ores are compared in accordance with
increasing P - T conditions of metamorphism, that is from the Pechenga area to
Lovnoozero. These data indicate the following features.
The number of individual primary silicates in the ores defined by each group
of intrusives differs widely and irregularly, but their relative volume increases
considerably with increasing metamorphic grade. The number of primary ore
minerals (oxides, sulfides) and their relative abundance ranges only slightly and
irregularly. Note that Fe and Ti oxides are equally abundant in all areas, but Ti
oxides are predominant in the Lovnoozero area. The Allarechka ores are dis-
Table 2. Mineral composition of the disseminated and impregnated ores a o
::;
5-
...
Minerals Number of mineral species Mineral content (vol07o)
~
o
p A L,u L,rn S NKT P A L,u L,rn S NKT o-,
"::
Primary silicates 6 2 4 3 4 4 12 20 55 65 95 95
Primary oxides 4 4 4 4 4 4 2 1 2 2 1 1 ~
Early sulfides 5 7 6 6 5 5 4 7 5 5 3 3 o...
"C
=.
Total 15 13 14 13 13 13 18 28 62 72 99 99 3
s·
Secondary silicates 20 17 10 8 9 9 70 60 30 20 ...5-
Secondary oxides 4 3 3 3 2 2 6 4 1 1 '"I1
o
Late and secondary sulfides 10 10 6 6 6 6 6 8 5 7
8
~
Total 34 30 19 17 17 17 82 72 36 28 o·
::;
o
-,
Sum total 49 43 33 30 30 30 Z
('i'
i'<"
a The areas are: P Pechenga; A Allarechka; L, u Lovnoozero - ultramafites; L, rn Lovnoozero - mafites; S Sopcha, peridotite and pyroxenite layers; ~
NKT Nittis-Kumuzhia-Travyanaya massif, bentic deposit. (;
o
:g
......
Ul
E..
::n
p.
'"o
.g
o
~.
'"

~
v.
146 Yu.N. Yakovlev and A.K. Yakovleva

tinguished by the volume of primary ore minerals, since they contain less oxides
and more sulfides. On the whole, because of the insignificant role of primary
sulfides, the total volume % of primary minerals increases regularly from
Pechenga to Lovnoozero which latter ores are similar to the ores of the Monche-
gorsk pluton.
The number of secondary silicates in the ores decreases systematically and
their amount decreases more abruptly from low- to high-metamorphic grade.
Principal changes involve layered hydrated silicates (serpentines;- chlorites, talc),
whereas the volume of amphiboles and, to a lesser extent, micas, increases
slightly. The number of secondary oxides is virtually constant, whereas their
volume essentially decreases mainly at the expense of secondary magnetite with
increasing metamorphic grade. The number of species of late and secondary
sulfides gradually decreases, and their content differs irregularly with increasing
metamorphic grade.
A gradual reduction occurs in the total number of mineral species with in-
creasing metamorphic grade up to the level of practically nonmetamorphosed ore
of the Monchegorsk pluton. The important changes are due to variations in the
number and volume of silicates and, in the first place, to the intensity of develop-
ment of secondary hydrated minerals, viz. serpentines, amphiboles, micas, talc.
The ratio of primary to secondary silicates defines the degree of metamorphic ore
transformation, whereas the number of mineral species of sulfides and oxides
and their volume differ insignificantly and, on the whole, irregularly.
Thus, maximum changes in mineral composition are typical of ore meta-
morphosed at the greenschist facies and minimum changes take place at the
granulite facies.

Chemical Composition of the Ores

The chemical composition of the ores (Table 3) is based on the concentration of

major economic elements (Kozlov 1973, Yakovlev and Yakovleva 1974,
Yakovlev et al. 1980). Massive and breccia ores are grouped together for all
deposits, but it should be noted that breccia ores are dominant (do not constitute
less than 80% by volume) in the Pechenga and Lovnoozero areas as well as at the
Vostok deposit, but massive ores are predominant at Allarechka and in the
Monchegorsk pluton. There are no data on the sulfur contents of the Pechenga
ores.
These data indicate the following features. The Ni and Cu contents of mono-
type-impregnated ores (I and II) from the Pechenga, Allarechka, and Lovno-
ozero areas differ dramatically (2.3 - 4.4 times), whereas the Ni/Cu ratio has a
narrower range (2.1 - 2.2 times). In contrast, the Co content is more constant,
but the NilCo ratio is quite varied (3.3 -4.5 times) because of the differences in
Ni content. A successive increase in the NilCu ratio from low-grade disseminated
ores to high-grade impregnated and further to breccia (massive) ores is apparent
at the majority of deposits. The exceptions are the Eastern ore cluster of
Pechenga and the deposits of the Monchegorsk piuton in which this parameter
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 147

Table 3. The chemical composition of the ores

Ore Content (wtOJo) Ratio

types
Ni Cu Co S Nil Nil Nil Nil LMel
Cu Co l:Me S S

The Eastern ore cluster (the Pechenga area)

0.51 0.20 0.016 2.55 31.9 0.70

II 1.78 0.75 0.036 2.37 49.4 0.69
III 1.96 0.93 0.035 2.11 56.0 0.67

The Flangovoye deposit (the Pechenga area)

I 0.65 0.31 0.018 2.10 36.1 0.66

II 2.39 1.13 0.034 2.11 70.3 0.67
III 2.25 1.01 0.043 2.23 52.3 0.68

The Allarechka deposit

I 0.67 0.30 0.013 1.48 2.23 51.5 0.68 0.45 0.66

II 4.80 1.72 0.043 13.85 2.79 67.6 0.73 0.35 0.48
III 13.17 3.95 0.16 36.95 3.33 82.3 0.76 0.36 0.47

The Vostock deposit

I 0.83 0.50 0.023 4.13 1.66 36.1 0.61 0.20 0.33

II 3.13 1.75 0.043 8.68 1.79 72.8 0.64 0.36 0.57
III 7.37 3.10 0.097 18.60 2.38 76.0 0.70 0.40 0.57

The Lovnoozero deposit (mafites)

I 0.62 0.30 0.019 4.06 2.04 32.4 0.66 0.15 0.23

II 1.71 0.79 0.054 11.04 2.15 31.6 0.67 0.15 0.23
III 2.68 0.48 0.114 25.86 5.58 23.5 0.82 0.10 0.13

The Sueinlagash ore showing (ultramafites)

0.33 0.28 0.021 3.62 1.18 15.7 0.52 0.09 0.18

II 1.09 0.82 0.067 13.53 1.33 16.3 0.55 0.08 0.15
III 1.58 0.43 0.093 20.50 3.67 17.0 0.75 0.08 0.10

The Sopcha deposit

I 0.53 0.21 0.Q28 1.43 2.52 18.9 0.69 0.37 0.54

III 3.19 2.75 0.127 19.79 1.59 31.0 0.60 0.16 0.31

The NKT deposit

I 0.32 0.14 0.020 1.10 2.28 16.0 0.64 0.29 0.44

III 5.1 2.6 0.180 31.92 1.96 28.3 0.65 0.16 0.25
148 Yu.N. YakovlevandA.K. Yakovleva

diminishes. It shows a maximum range in the ores from the Lovnoozero area
(2.7 -3.1 times), less in the Allarechka ores (1.4-1.5 times), slightly less in the
ores from the Monchegorsk pluton (1.2 -1.6 times), and least in the Pechenga
ores (1.1 -1.2 times). The Ni/Co ratio generally differs similarly, that is, it
increases toward the high-grade ores, except for the Lovnoozero area and the
Flangovoye deposit; it decreases in the same direction in the ore of the former
and ranges irregularly in the ore of the latter.
The NilS and IMe/S ratios are important geochemical characteristics. At all
deposits, except for Vostock, the NilS value decreases regularly from low-grade
disseminated ores to breccia ones. Maximum differences in the ratios are charac-
teristic of the ores of the Monchegorsk pluton and the Allarechka area (1.8 - 2.3
and 1.3 - 2.0 times, respectively) and minimum differences of the ores of
Lovnoozero (1.1-1.5 times). The IMe/S parameter ranges similarly. However,
in the ores of the Monchegorsk pluton and the Allarechka area, it lies within a
more narrow range than the NilS ratios; a wider range in the ratio occurs in the
Lovnoozero area. Based on lump ore (about 400 lumps) analyses performed at
the Institute of Geology, both parameters gradually diminish (1.6 -1. 7 times)
toward the breccia and massive ores at Pechenga (Yakovlev et al. 1980). The
absolute value of both parameters for ores from all the areas fluctuates over a
wide range: from 0.08 to 0.45 for NilS and from 0.10 to 0.66 for IMe/S. The
sulfide phase of the Allarechka ore is the richest in Ni, one of the Pechenga and
Monchegorsk ores is somewhat poorer, and one of the Lovnoozero ores is the
poorest, especially in the ultramafites.
The parameter characterizing the Ni fraction of total metal (Nil IMe) is
notable for its stability. Its value ranges from 0.61 to 0.70 in low-grade dis-
seminated ores of all the areas, and from 0.64 to 0.73 in high-grade impregnated
ores. A similar range is shown in the ores of the Monchegorsk pluton. In all
studied ores its range is 0.60 - 0.82, but the maximum value only characterizes
localized breccia ore from the Lovnoozero deposit. The minimum value of the
parameter (0.52) is established at the Sueinlagash ore showing in the Lovnoozero
area. Thus, the Ni fraction of the total metals is persistent everywhere indepen-
dent of the grade of metamorphism.
Judging from changes in all geochemical characteristics (component contents
and ratios), the Allarechka ore is the most unusual of the impregnated ores. If
consideration is given to all ores (I - IV types), the Allarechka and Monchegorsk
ores show the greatest contrast, the Lovnoozero ores are close to these, and the
Pechenga ores prove to be the least changeable. Thus, the most intensively meta-
morphosed ores (the Allarechka and Lovnoozero areas) and the virtually un-
metamorphosed ores (the Monchegorsk pluton) in mafites-ultramafites differ is
roughly similar manner.

Discussion

Comparisons of mineral and chemical compositions of disseminated and im-

pregnated Ni - Cu ores from the various areas of the Kola Peninsula indicate that
On the Role of Metamorphism in the Formation of Nickel-Copper Sulfide Deposits 149

the degree of metamorphic transformation of the nickel-bearing mafites and

ultramafites is determined by (1) their original composition; (2) their position
within specific tectonic zones; and (3) the structural position of the specific
intrusives. As a rule, the ultramafites are more altered than the mafites within
one zone, and smaller massifs are more altered than larger ones of similar
composition.
The greatest ranges in mineral composition are typical of disseminated and
impregnated ores of the Pechenga area (greenschist facies) and the least differ-
ences are characteristic of the ores from the Lovnoozero area (granulite facies).
The chemical composition of the ores is distinguished by a limited range in most
areas except for great changes in the Allarechka area (amphibolite facies). A
comparison of geochemical characteristics of syngenetic (low-grade dis-
seminated) ores and epigenetic (breccia) ones indicates a similar trend of varia-
tion in all areas. The NilCu and NilCo ratios are greater and the NilS and
EMe/S ratios are lower in epigenetic ores. These data, as well as the constant
Nil EMe parameter, provide evidence on the persistence of major chemical
composition independent of metamorphic grade developed in the original nickel-
bearing rocks.
The ore grade of the mafite-ultramafite massifs does not depend on the
presence or absence of pyrite-pyrrhotite mineralization in the enclosing series.
Ore-bearing and ore-free massifs occur in the enclosing or surrounding rocks
with the pyrite-pyrrhotite mineralization occurring in all areas of the north-
western part of the Kola Peninsula. All geochemical parameters of the pyrite-
pyrrhotite mineralization are 5 -10 times less than these of the Ni - Cu mineral-
ization. The evidence presented above does not support a metamorphic hypo-
thesis for the genesis of Ni - Cu sulfide ores, but points to the key role of
magmatic processes in their formation. Metamorphism results in significant
changes in the composition of primary ores, but is probably insignificant in ore
generation.

Major Conclusions

1. The most intensive changes in mineral compositions of disseminated ores

occur under low-temperature conditions of regional metamorphism, whereas
minimal changes occur under high-temperature conditions. At greenschist
facies, deviations from the original magmatic conditions were at a maximum,
whereas at granulite facies they were minimal.
2. The more the mineral composition of an ore changes, the less its geochemical
characteristics differ from others and vice versa. This fact provides evidence
on the primary nature of the contrast in chemical composition of the various
ore types and on the gradual decreasing of this contrast by the process of
metamorphism.
3. All metamorphic processes involve changes in preexisting ore and are not ore-
generating. Thus, the Ni - Cu deposits of the Kola Peninsula should be
classed as metamorphosed magmatic deposits.
150 Yu. N. Yakovlev and A. K. Yakovleva: On the Role of Metamorphism

References

Arutyunyan LA, Petrenko GV, Mityunin YuK (1978) On sulfurization of olivinites under hydro-
thermal conditions and Ni mobilization. In: Depositional conditions of Ni - eu ores (natural para-
geneses and experiment). Kola Branch USSR Acad Sci, Apatity, pp 73 - 85 (in Russian)
Balabonin NL, Yakovlev YuN, Astafyeva VV, Vorobets AA (1974) On geochemical specialization of
nickel-copper and kies mineralizations in the A1larechka area. In: Aspects of geology and
metallogeny of the Kola Peninsula, vol V, pt 2. Kola Branch USSR Acad Sci, Apatity, pp 185 -194
(in Russian)
Gorbunov 01 (1968) Geology and genesis of Ni - eu sulfide deposits. Nedra, Moscow, 352 pp (in
Russian)
Gorbunov GI, Yakovlev YuN, Astafyev YuA et al. (1973) Atlas of structures and textures of Ni- eu
sulfide ores in the Kola Peninsula. Nauka, Leningrad, 283 pp (in Russian)
Gorbunov 01, Astafyev YuA, Goncharov YuV et al. (1978) Structures of Ni - eu ore fields and
deposits of the Kola Peninsula. Nauka, Leningrad, 160 pp (in Russian)
Gorbunov 01, Vinogradov LA, Makievsky SI, Yakovlev YuN (1982) Basite-ultrabasite magmatism
of the Kola Peninsula and its metallogeny. In: Geology, petrology and correlation of crystalline
complexes in the European part of the USSR. Proc 3rd Region Petrogr Meet, Dnepropetrovsk
1979. Nedra, Leningrad, pp 82-92 (in Russian)
Kozlov EK (1960) Regularities of Ni - eu sulfide deposit emplacement in the Kola Peninsula. In:
Metallogeny of precambrian shields and ancient mobile belts. UkrSSR Acad Sci, Kiev, pp
142 - 149 (in Russian)
Kozlov EK (1973) Natural series of nickel-bearing intrusive rocks and their metallogeny in the Kola
Peninsula. Nauka, Leningrad, 288 pp (in Russian)
Likhachov AI (1974) On Genesis of Ni - eu deposits of the Pechenga and Allarechka Areas. SOy
Geol 7:69 - 84 (in Russian)
Predovsky AA, Zhangurov AA, Ilin YuI, Akhmedov AM (1968) New data on regularities of
emplacement and conditions of formation of Pechenga Ni - eu sulfide deposits. In: Geological
structure and orebearing in the Kola Peninsula. Kola Branch USSR Acad Sci, Apatity, pp 61 - 67
(in Russian)
Smirnov VI (1982) Endogenetic ore formation in geological history. Geol Ore Depos 4:3-20 (in
Russian)
Yakovlev YuN, Yakovleva AK (1974) Mineralogy and geochemistry of metamorphosed Ni - eu ores.
Nauka, Leningrad, 330 pp (in Russian)
Yakovlev YuN, Yakovleva AK, Neradovsky YuN et al. (1980) Principal regularities of changes in
composition of metamorphosed Ni - eu disseminated ores: In: Mineral complexes and minerals of
the Kola Peninsula. Kola Branch USSR Acad Sci, Apatity, pp 55-63 (in Russian)
Yakovlev YuN, Yakovleva AK, Neradovsky YuN et al. (1981) Mineralogy of Ni - eu sulfide deposits
in Kola Peninsula. Nauka, Leningrad, 352 pp (in Russian)
Zak SI, Kochnev-Pervukhov VI, Proskuryakov VV (1972) Ultrabasic rocks of the Allarechka Area,
their metamorphism and mineralization. Kareliya, Petrozavodsk, 129 pp (in Russian)
The Gabbro-W ehrlite Association in the Eastern Part of
the Baltic Shield
E. HANSKl 1

Abstract

Proterozoic mafic and ultramafic intrusive rocks are widespread in the Archaean
Kuhmo greenstone belt, including three minor Ni - Cu deposits. Similar intru-
sions have also been found in the Koli area cutting Jatulian metasediments and
the underlying granitoid basement. They are incorporated in the same magmatic
association "the gabbro-wehrlite association" as the Ni-bearing intrusions of
Pechenga. The cumulate sequences in the layered bodies include olivine, olivine-
clinopyroxene, clinopyroxene, clinopyroxene-magnetite, and plagioclase-clino-
pyroxene-magnetite cumulates. Ti-rich amphibole and mica occur abundantly as
postcumulus phases in ultramafic rocks. The chemical compositions of the min-
erals and rocks differ from those of the Pechenga intrusions, reflecting differ-
ences in the parent magmas and their fractionation stages. The parent magmas in
eastern Finland and Pechenga are similar to picrites of the Suisaarian and
Pechenga types, respectively.

Introduction

Large numbers of both intrusive and extrusive ultramafic rocks have been dis-
covered in the course of the long period over which the Kuhmo Archaean green-
stone belt has been studied. The close spatial relationship between these and the
highly magnesian character of the ultramafic cumulates have led some authors to
take for granted the genetic relationship between the intrusions and the komati-
itic volcanic rocks (Auvray et al. 1982). In 1981, however, a similarity was
detected between the ultramafic and mafic intrusions and those occurring in the
Koli area, where they are known to cut Proterozoic Jatulian metasediments
(Hanski 1982). Age determinations made by the Geological Survey have later
confirmed the Proterozoic age (2.1 - 2.2 Ga) of the intrusions in Kuhmo (Hyp-
ponen 1983).
Further study revealed that the intrusions in Kuhmo and Koli have many
features in common with the host rocks of the Ni - Cu deposits of Pechenga in

1 Department of Geology, University of Oulu, Oulu, Finland.

Present address: Geological Survey of Finland, PL 77,96101 Rovaniemi, Finland

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
152 E. Hanski

the Kola Peninsula, USSR. Russian geologists have used "formational analysis"
to assign the rocks of Pechenga to the gabbro-wehrlite formation (Dagelayskaya
1979). Since the notion of a magmatic formation in the sense defined by the
Russian geologists (Masaytis et al. 1979) is not used in the English terminology,
the term is replaced here by "association", as is done in Suslova et al. (1981).
This work concentrates on the major geological, mineralogical and geo-
chemical features of the gabbro-wehrlite association in Kuhmo and Koli, com-
paring the intrusions with the rocks of Pechenga in the light of data collected
from the literature and whole rock and mineral analyses carried out on some
museum samples.

Regional Geology

The Kuhmo greenstone belt in eastern Finland forms a narrow belt some 200 km
in length running from S to N (Fig. 1). The supracrustal rocks include mainly
tholeiitic, komatiitic and acid metavolcanics, quartzites and mica schists. The
general geology and petrology of the belt and surrounding granitoids are dealt
with in many works by Finnish and French geologists (e.g., Taipale 1983, Blais et
al. 1978).
The rocks of the gabbro-wehrlite association (henceforth OWA) are concen-
trated in the greenstone belt, being virtually absent from the surroundings. They
form an almost uninterrupted chain of ultramafic rocks, over 100 km in length,
in the middle and northern part of the belt. The best-known examples are the
NiUWiniemi serpentinite and the dykes of Petajaniemi. Piquet (1982) noted a
slight difference in metamorphic grade between ultramafic metacumulates and
Archaean metavolcanics, but a more pronounced difference is seen in the
deformation grade, for the metacumulates of the OWA have in most cases not
suffered any pervasive deformation.
The structure and stratigraphy of the Koli area have been described by
Piirainen et al. (1974). Two main stratigraphic units can be distinguished: the
Archaean basement complex and the overlying Proterozoic sedimentary forma-
tions. The latter rocks are divided into the Jatulian quartzites and the overlying
Kalevian graywacke schists.
Both the basement and the Jatulian quartzites are cut by numerous basic
dykes and sills, but nowhere are the Kalevian metasediments found to be cut in
this way. A differentiated sill at Savilahti, near the shore of Lake Pielinen, has
been studied in greater detail. This occurs within the Archaean granitoids near
the overlying unconformity and follows the dip and strike of the J atulian meta-
sediments.

General Features of the Gabbro-Wehrlite Association

The intrusions of the OWA have two principal modes of occurrence. They occur
as layered sills and minor intrusions and dykes affected by flowage differentia-
The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 153

200KM
• ::::::::J

Fig. 1. The occurrences of the gabbro-wehrlite association in the Baltic Shield. 1 Archaean and
Proterozoic granitoids; 2 Archaean greenstone belts; 3 Proterozoic schists; 4 Gabbro-wehrlite
association. Occurrences of the GW A: 1Koli; 2 Kuhmo; 3 Suomussalmi; 4 Puolanka; 5 Kuusamo;
6 Kemi; 7 Kittila; 8 Sodankyla; 9 Savukoski; 10 Pechenga; 11 Imandra-Varzuga; 12 Kuolajarvi;
13 Hautavaara; 14 Onega; 15 Vetrenyi Poyas. Base map simplified mainly after Simonen (1980) and
Bilibina (1980)
154 E. Hanski

tion. The steeply dipping dykes, 20 - 50 m in thickness, are composed almost

totally of ultramafic olivine or olivine-clinopyroxene cumulate.
The exposed parts of the Naataniemi intrusion are composed mainly of
serpentinized metaperidotite, while in a few places the ultramafic cumulates are
found to be overlain by metapyroxenites and metagabbros. The uppermost parts
are not exposed at all. The measured dips of the layering vary in most cases
between 45 0 -70 0 and the estimated thickness of the layered series is about
400m.
In contrast with the Naataniemi intrusions, the last differentiates and the
upper marginal zone of the Savilahti sill at Koli are well-exposed. The estimated
thickness.of the sill is about 300 m and it can be followed along the strike for over
15 km. By combining the available data from both areas, a generalized strati-
graphic section with cumulus minerals can be constructed (Fig. 2).
The layered sequence includes the following cumulates: olivine (-chromite),
"porphyritic" olivine, olivine-clinopyroxene, clinopyroxene, clinopyroxene-mag-
netite and plagioclase-clinopyroxene-magnetite cumulates. The latter rock type
grades upwards into a plagioclase and quartz-rich rock crystallized from the
residual liquid and further into an upper marginal zone. This zone in composed
of coarse-grained gabbro and clinopyroxenite. The cpx-magn cumulate is charac-
terized by a small amount of sulphide.
The contacts between each zone represented in Fig. 2 are relatively sharp,
except for that between the gabbro and the upper marginal zone. Magmatic

cumulus intercumulus ... .. subsolidus

Upper marginal
zone
( <:::: T43048 :9 T T
I

~-J ' I

~~~~
~/#"~
Plag-cpx - magn cum. "
~

....
,"
• ,t .

I \'_1
.... ":',
I .. ' ,
.
''''I~'-
,':I~
"

~
,'-:-'
Cpx-magn cum. 43:43:''''

CPX cum.

... 1.

I
73
Ol-cpx cum. .":4 8 :8
T
1
T I:
1
I I:
• Porphyrit iC· 01 cum, I I'
I 1:

01 cumu late

90 ...
I
~
1
.L
CD !!!
"0 :: .'"
c

..,:;:"'" "
0
.l>
:;:
.,
v""-...e
~ ~ !!! "Ci
E
0
.c
"
c:
&: C
_>(

1"0 ~
c "
.
0
0.
E
0
c:
~
c
0

2'" . "
Co

., .
E ~
;; c
~" ~"
~ Co
:; I
~
U 0 ~~~
U"'_ c:: en
&:
"- U
Co
< "
0 <

Fig. 2. Generalized stratigraphy of the layered bodies in eastern Finland. Section thickness about
400m
The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 155

rhythmic layering is seldom found in any zone. The regular, simple stratigraphy,
coupled with geochemical data, suggests that the layered bodies were formed by
fractional crystallization in a magma chamber formed by a single magma pulse.
In the light of the discovery of a new magmatic event in the Kuhmo green-
stone belt it is interesting to evaluate the possible connection between the known
Ni - Cu deposits and this magmatism. Of the four known deposits, that of
Arola, which has a high average NilCu ratio (28), is probably related to the
Archaean volcanism (Lehtinen 1983). The other three, the Ni-Cu deposits of
Peura-aho and Hietaharju in Suomussalmi, described by Kojonen (1981) and a
so far little studied mineralization at Siivikkovaara can be interpreted on petro-
graphic grounds (see below) as belonging to the GWA.
The Cu/(Cu + Ni) ratios (0.30, 0.33) of the deposits in Suomussalmi
(Kojonen 1981) are similar to the ratios of the syngenetic ores at Pechenga (Zak
et al. 1982), but the PtI(Pt + Pd) ratio at Pechenga (0.55) is much higher than at
Suomussalmi (0.176-0.333). It should be kept in mind, however, that the
published PGE data for Pechenga is so far very limited. Only one analysis of an
ore containing 4.60/0 Ni and 2.45% Cu has been reported by Gorbunov (1968),
who also mentions that the ratio of Pt to Pd increases in a regular manner
between high-grade and low-grade ores.

Petrography of the Ultramafic Rocks

Although the degree of alteration of the metacumulates, particularly in the

Kuhmo area, is high, the magmatic mineralogy can be inferred from the well-
preserved pseudomorphous textures.
The cumulus olivine and intercumulus clinopyroxene and phlogopite in the
lowermost of cumulates are now replaced by serpentine, tremolite and chlorite,
respectively. The quantity of intercumulus minerals in the overlying "por-
phyritic" 01 cumulate increases and sometimes reaches one-half of the volume.
The "porphyritic" appearance of this zone is due to poikilitic clinopyroxene
crystals 2 - 5 mm in size, which give a mottled texture to the eroded surface. The
poikilitic clinopyroxene crystals enclose rounded olivine crystals, and the
clinopyroxene is in turn surrounded by poikilitic brown amphibole and
phlogopite. Chlorite pseudomorphing phlogopite still sometimes inherits the
brownish colour of its precursor. Minor intercumulus minerals are ilmenite and
apatite. Apart from the crystaline form of the clinopyroxene and the absence of
chromite, the mineralogy and texture of the ol-cpx cumulates are the same as in
the zone described above.
In the severely altered host rocks of the Ni - Cu deposits of Peura-aho poiki-
litic textures can still be distinguished where olivine crystals have been enclosed in
clinopyroxene. In addition, there exist relicts of brown intercumulus amphibole
and many samples are seen to have contained a large amount of intercumulus
phlogopite, now replaced by brownish chlorite.
The olivines poikilitically enclosed by clinopyroxene in the ultramafic
cumulates of the Pechenga intrusions often have rounded forms (see Vayrynen
 156 E. Hanski

1938, Fig. 18). Other major intercumulus minerals include brown amphibole and
brown mica (Yakovleva 1981, Yakovleva and Osokin 1981). The texture of the
ultramafic rocks is often said to be "porphyritic" (e.g. Smol'kin 1974a).
The total absence of olivine from the clinopyroxene cumulates and the
rounded forms of the olivine in the underlying cumulates in the intrusions of
Kuhmo and Koli suggest a peritectic relation between olivine and clinopyroxene.

Mineralogy

Olivines. Two generations of olivine are distinguished on textural and chemical

grounds. The first, which is encountered only in the Kuhmo belt, is magnesian
(Fo s6 -%) and relatively high in MnO and very low in CaO (Fig. 3, Table 1). Its
composition is compatible with a metamorphic origin (Piquet 1982). In the
second, magmatic generation the Fo content varies between 90 -730/0. The range
of olivine compositions in Pechenga is more limited, varying in the range FO s3 - 72
(Smol'kin 1978, Smol'kin and Pahomovskiy 1982). It is significant that olivine
ceases to crystallize at the same level of Fe-enrichment as in eastern Finland.
Olivines from Pechenga have a relatively high CaO content (Fig. 3). As the crys-
tallization depth was not essentially different, the difference in CaO content
could be correlated with the lower silica activity of the magma at Pechenga (cf.
Stormer 1973).
Both the Fo and the NiO (0.34 - 0.40%) content of most magnesian olivines
from Kuhmo are high enough for these to have crystallized from a very primitive
magma (Fig. 4). The NiO content decreases below 0.2% with fractionation.

0.8 0.7

[J

• Kuhmo, KO
li 0.6
o K uhm o ( me l am. )
0.6
• Pechenga (E PM A) 0.5
o PeC he nga ( SEP)

0.4
CaD
0.4 MnD

• 0 .3

0.2
• 0 .2

0.1

0
75 80 85 90 95 100 70 75 80 85 90 95 100
Fo Fo

Fig. 3. CaD, MnO, and Fo concentrations in olivines from eastern Finland and Pechenga. The
Pechenga analyses are taken mainly from Smol'kin (1978) and Smol'kin and Pahomovskiy (1982)
Table 1. Representative mineral analyses from eastern Finland and Pechenga
>-l
::r
1 2 3 4 5 6 7 8 9 10 11 12 13 '"0
~
cr'
SiOz 40.99 38.24 42.75 54.29 50.99 50.10 49.97 46.67 47.14 41.56 39.72 36.22 35.85 ..,cr'
0
TiO z 0.01 0.00 0.01 0.40 0.76 1.03 1.38 4.08 4.14 6.72 5.74 7.49 7.20
~
Al z0 3 0.02 0.00 0.01 1.57 2.50 2.94 3.42 8.41 7.95 9.60 13.20 11.70 12.46 '::r"
FeO 9.81 24.01 3.72 4.91 8.63 8.55 10.28 7.88 7.74 9.23 8.26 11.56 12.83 ~
MnO 0.19 0.40 0.52 0.16 0.20 0.18 0.17 0.10 0.09 0.08 0.08 0.40 0.42
MgO 48.18 37.83 52.97 16.41 14.93 15.73 13.53 16.11 16.06 14.09 14.24 20.98 19.23 en
"0>
CaO 0.10 0.06 0.01 21.02 20.64 20.04 19.81 9.80 10.26 11.00 11.47 0.06 0.15 (")

NazO 0.01 0.00 n.d. 0.44 0.50 0.28 0.89 3.25 3.12 3.05 2.36 n.d. 0.07 ~~
0
K20 0.00 0.00 0.01 0.04 0.00 0.03 0.02 0.40 0.44 0.80 0.55 7.41 7.29 ;;
CrZ03 0.02 0.03 0.02 0.48 0.01 0.22 n.d. 0.85 0.72 0.08 0.16 0.76 0.62 S·
NiO 0.37 0.15 0.24 0.04 0.04 0.02 n.d. 0.05 0.05 n.d. 0.06 0.16 n.d. S-
'"ttl
~
Total 99.70 100.72 100.26 99.76 99.20 99.12 99.47 97.60 97.71 96.21 95.84 96.74 96.12 :a...,
;;
'"
Si 1.008 0.996 1.016 1.983 1.914 1.882 1.885 6.709 6.767 6.194 5.893 5.260 5.261 '"C
~
Ti 0.000 0.000 0.000 0.011 0.022 0.029 0.039 0.441 0.447 0.753 0.641 0.818 0.795 ::1.
0
...,
Al 0.001 0.000 0.000 0.068 0.111 0.130 0.152 1.452 1.344 1.686 2.308 2.003 2.155
Fe 0.202 0.523 0.074 0.150 0.271 0.269 0.324 0.948 0.929 1.150 1.025 1.404 1.574 S-
Mn 0.004 0.009 0.010 0.005 0.006 0.006 0.005 0.012 0.010 0.010 0.010 0.049 0.052 '"
I:Cl
Mg 1.766 1.469 1.877 0.894 0.835 0.881 0.761 3.452 3.437 3.130 3.148 4.541 4.206 ~
n·
Ca 0.003 0.002 0.000 0.822 0.830 0.807 0.801 1.510 1.577 1.756 1.824 0.009 0.024 CIl
0.000 n.d. 0.036 0.020 0.867 0.881 0.679 0.020 ::r
n;.
Na 0.000 0.031 0.065 0.906 n.d.
K 0.000 0.000 0.000 0.002 0.000 0.001 0.001 0.073 0.080 0.152 0.104 1.373 1.365 s:
Cr 0.000 0.001 0.000 0.014 0.000 0.007 n.d. 0.096 0.082 0.009 0.018 0.087 0.072
Ni 0.007 0.003 0.005 0.001 0.001 0.001 n.d. 0.006 0.006 n.d. 0.007 0.019 n.d.
Fo 89.7 73.7 96.2
Wo 44.1 42.9 41.2 42.5
En 47.9 43.1 45.0 40.3
Fs 8.0 14.0 13.7 17.2

1 Olivine, 01 cumulate, Kuhmo; 2 Olivine, "porphyritic" 01 cumulate, Koli; 3 Metam. oliv., 01 cumulate, Kuhmo, (Piquet 1982); 4 Clinopyroxene, ol-cpx
cumulate, Kuhmo; 5 Clinopyroxene, cpx-magn cumulate, Koli; 6 Clinopyroxene, 01 pyroxenite, Pechenga, (Zhangurov and Predovskiy 1974); 7 Clino-
VI
pyroxene, plagiopyroxenite, Pechenga, (Smol'kin 1978); 8 Edenitic hornblende, "porphyritic" 01 cumulate, Koli; 9 Edenite, "porphyritic" 01 cumulate, Koli; -..I
10 -11 Kaersutite, 01 cumulate, Pechenga; 12 -13 Phlogopite, 01 cumulate, Pechenga.
-
158 E. Hanski

0.4 Fig. 4. NiO vs Fo for olivines from east-

o ern Finland and Pechenga. Symbols and
o data sources as in Fig. 3

0.3
o
o
0.2

0.1
• o 0

0 .7 0.8 0.9 1.0

Fo

Fig. 5. Comparison of pyroxene

composition trends at Koli (field
K) and Pechenga (P) (analyzed at
the Univ. of Oulu) with others on
a Ca - Mg - Fe diagram. Lines:
1 Alaskan type (Irvine 1974);
2 Black Jack (Murray 1954);
3 Garbh Eilean (Wilkinson 1957);
Mg Fe 4, 5 Skaergaard (Nwe 1976)

Olivine grains from Pechenga separated and analyzed by chemical methods tend
to have a higher NiO content than those analyzed by microprobe. Compositions
of the latter coincide in most cases within the range of olivines from Kuhmo and
Koli, but more microprobe data would be needed to reach any conclusions with
certainty. It should be added that Smol'kin and Pahomovskiy (1982) did not
notice any differences in Ni content between olivines from ore-bearing and
barren intrusions.

Clinopyroxenes. Magmatic and metamorphic varieties can be also distinguished

among the pyroxenes. The latter type consists of diopside with very low A1 20 3,
Ti0 2 , and Nap. The magmatic type is preserved from alteration only in
cumulates underlying gabbroic rocks. It forms a short trend in a ternary plot of
Mg, Ca and total Fe subparallel to the diopside-hedenbergite intersection
(Fig. 5). It has lower Wo, A1 20 3 , and Ti0 2 than the clinopyroxenes from
Pechenga (Table 1). The higher Ti and Al content of the Pechenga pyroxenes
could be ascribed to a lower concentration of silica in the magma (cf. Le Bas
The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 159

1962). Limited Fe-enrichment is also typical of pyroxenes at Pechenga, the most

Fe-rich pyroxene reported has an Fs content of about 230/0 (Smol'kin 1978).

Amphiboles and Micas. Intercumulus amphibole is still well-preserved in many

places in the ultramafic cumulates of the Savilahti sill, consisting of Na-rich
titanian edenite or edenitic hornblende according to the classification of Leake
(1978) (Table 1). The corresponding amphibole in Pechenga is sufficiently rich in
Ti0 2 and poor in Si0 2 to be classified as kaersutite as the Russian geologists have
already called it (e.g. Yakovleva 1981). The Pechenga amphibole also has a
higher KINa ratio, probably reflecting a higher KINa ratio in the parent magma.
Intercumulus mica has been replaced by penninitic chlorite in the ultramafic
rocks of Kuhmo and Koli, but it still has a poikilitic habit. The intercumulus
mica found at Pechenga has been referred to as biotite (Vayrynen 1938,
Yakovleva and Osokin 1981). According to a few microprobe analyses this mica
has an Mg/Fe ratio higher than 2: 1 (Table 1) and is rich in Ti0 2 (3.4-7.5%) and
should properly be called titanian phlogopite.

Geochemistry

Major Elements

Representative whole rock analyses from eastern Finland and Pechenga are pre-
sented in Table 2. The analyses of cumulates from the layered bodies from
eastern Finland define a continuous, strongly curved trend from the MgO corner
to the AIP3 corner in the CMA diagram (Fig. 6). The analyses from Pechenga,
which have mainly been taken from the compilation of Tkachenko (1982), form
a similar but more dispersive pattern.
The trend of the Finnish analyses in the FeO/(FeO + MgO) vs Si0 2 diagram
(Fig. 7) begins at a considerably lower FeO/(FeO + MgO) ratio than in the case of
Pechenga. On the other hand, the most differentiated rocks in Pechenga have a
lower Si0 2 content. The difference in FeO/(FeO + MgO) ratio is consistent with
the higher Fa content of the most magnesian olivines at Pechenga. The
FeO/(FeO + MgO) ratio of the most magnesian rocks in the Savilahti sill,
Petajaniemi dykes and Ni-bearing intrusions in Suomussalmi is higher than the
ratio of layered bodies in Kuhmo, implying fractionation before emplacement.
The difference in silica enrichment may be due to the lack of analyses, but it is
nevertheless compatible with the probable lower silica content of the parental
magma at Pechenga as indicated by the mineralogical data.
Since the AIP3 and Ti0 2 concentrations in olivines are very low, the ratio
between these elements in olivine cumulates reflects this ratio in the intercumulus
liquid and the parental magma. A plot of AIP3 vs Ti0 2 is shown in Fig. 8. The
lowermost olivine cumulates from Kuhmo have a trend of their own which is
probably caused by the removal of Ti0 2 during secondary alteration. The
concentrations of certain other components have also diminished in these rocks,
including CaO and Kp. The "porphyritic" 01 cumulates and ol-cpx cumulates
Table 2. Whole rock analyses of various rock types from eastern Finland and Pechenga ...
a,
0

1 2 3 4 5 6 7 8 9 10 11 12 13

SiO z 38.57 42.75 45.63 50.93 40.03 50.22 36.29 35.85 36.74 47.97 36.47 46.99 49.59
TiO z 0.10 0.76 0.77 1.01 2.80 2.07 0.79 0.66 0.86 2.01 4.21 2.70 3.04
AI20 3 1.95 4.43 4.24 6.53 6.92 14.22 3.05 2.24 2.69 6.30 4.26 10.38 12.84
Fe203TOT 10.86 14.55 12.39 11.49 26.19 15.59 14.65 n.d. n.d. n.d. n.d. n.d. n.d.
Fe203 n.d. n.d. n.d. n.d. n.d. n.d. n.d. 9.54 7.45 2.21 9.95 4.45 1.57
FeO n.d. n.d. n.d. n.d. n.d. n.d. n.d. 8.35 11.51 9.81 16.76 11.35 11.61
MnO 0.09 0.18 0.22 0.29 0.26 0.44 0.20 0.21 0.23 0.15 0.18 0.20 0.18
MgO 36.44 25.52 20.93 12.06 8.40 3.98 29.48 33.21 28.86 11.64 10.62 5.47 3.76
CaO 0.00 5.44 11.17 14.45 12.25 8.02 2.85 2.00 2.60 15.48 13.66 10.53 8.07
Na20 0.18 0.09 0.19 1.58 0.89 3.78 0.14 0.04 0.17 1.10 0.21 2.10 4.45
K20 0.00 0.06 0.03 0.24 0.25 0.97 0.22 0.18 0.29 0.49 0.11 0.76 1.62
P 20 5 0.04 0.08 0.08 0.08 0.05 0.14 n.d. n.d. 0.06 0.19 0.14 n.d. 0.42
Cr203 0.35 0.47 0.36 0.09 0.01 0.01 n.d. 0.48 0.56 0.10 0.01 0.02 0.01
NiO 0.30 0.13 0.10 0.03 0.05 0.00 n.d. 0.43 0.24 0.01 n.d. 0.01 0.00
Loss 11.28 5.28 3.50 1.10 1.14 0.55 n.d. 6.89 7.53 2.36 1.97 4.19 2.74

Total 100.16 100.28 99.61 99.88 99.24 99.99 100.08 99.79 99.82 98.55 99.15 99.90

Eastern Finland: 1 01 cumulate, Kuhmo; 2 "Porphyritic" 01 cumulate, Koli; 3 Ol-cpx cumulate, Kuhmo; 4 Cpx cumulate, Koli; 5 Cpx-magn cumulate, Koli;
6 Plag-cpx-magn cumulate, Koli. Pechenga: 7 Average ultramafite of ore-bearing intrusions (Zak et al. 1982); 80livinite, (anal. 876, Tkachenko 1982);
9 Wehrlite, (911, op. ciL); 10 Pyroxenite, (1108, op. ciL); 11 Magnetite-rich pyroxenite, (903, op. ciL); 12 Gabbro, (1096, op. ciL); 13 Essexite, (1109,
op. ciL).

tI:I
:t:
§
E:
 The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 161

~_xx II< x
~
lIE ••
I- .:Io.. x_ _.. x
xx

CaO CaO
Pechenga Kuhmo+Koli
Fig. 6. Chemical compositions of cumulates from Pechenga and eastern Finland on a CMA diagram.
The Pechenga analyses are mainly from Tkachenko (1982)

1.0 1.0
.. 'I.
• • xx." '.:. ..
.•.•
001 0 .8 )II; . , :
0.8
~ .:~~ .~ x " " 'x
I •• _ .: X

'6 ~

'x ."'.:-
Xx • • •••

Q) 0.6 0.6 x ':r

,.,-~0
'~..
)I
• -'fit
x ~x

. ... <........
Q)
U. 0 .4 Jill II:
x" •
••
0.4
.., x

. '"
x • x )II; )I; x ~
X xx• • 'k~
0.2 • Pechenga 0 .2 .~~ .l'
Kuhmo+Koli
)II; xiJ\. •
o .0 +T"-r--r-.,.....,~-r-1,....,...--r-T-r-T"-r--r-T"-r-,--, o .0 -+-T"-r......,.........,--r-.,.....,~-,--,,....,..-r-r-o-r-r-T""TO
30 40 50 30 40 50 60 Si0270
60 Si0270

Fig. 7. FeOT/(FeO T + MgO) vs Si02 • The Pechenga analyses are mainly from Tkachenko (1982)

2.0
Ti0 2 . ..-... 2.0
Ti0 2

,.
x -; .)11;

1.5
··t:·
..
· "=.::t - 1.5
xX
x
•

)-...
Xx
.... x •

1.0
xI • 1.0 ~ ~
• 'I.
x x

Xx

0.5
~ 0.5

•
-
x_
x
Pechenga Kuhmo+Koli

0.0 0.0
0 5 15
0 5
10 AIO 15 10 AI 20 3
2 3

Fig. 8. Composition of ultramafic cumulates from Pechenga and eastern Finland on a plot of Ti02 vs
Ai20 3 • The data for Pechenga are mainly from Tkachenko (1982)
Table 3. REE analyses of various cumulates from Koli and Pechenga 0\
N
-
1 2 3 4 5 6 7 8 9 10 11 12 13 14

La 5.8 4.8 4.4 4.0 7.6 9.9 19.1 27.0 54.1 5.8 6.5 13.4 10.4 20.7
Ce 13.7 11.2 11.9 11.5 17.9 24.0 44.0 62.0 119.0 12.3 12.1 30.0 25.0 43.0
Nd 8.1 6.9 10.0 10.5 14.0 17.1 28.0 37.0 72.0 6.1 8.7 20.0 19.6 29.0
Sm 2.1 1.8 2.9 3.2 4.0 5.2 7.1 9.0 14.5 1.6 1.8 4.4 6.4 5.9
Eu 0.75 0.62 0.97 0.92 1.30 1.56 2.10 2.90 3.00 0.60 0.65· 1.86 2.20 2.20
Tb 0.39 0.28 0.44 0.48 0.51 0.61 1.05 1.44 2.30 0.26 0.25 0.62 0.75 0.93
Yb 0.84 0.74 1.25 1.16 1.35 1.66 2.70 3.40 6.30 0.42 0.49 1.21 1.28 1.67

Koli: 1 - 2 "Porphyritic" 01 cumulate; 3 Cpx cumulate; 4 Cpx-magn cumulate; 5 - 8 Plag-cpx-magn cumulate; 9 Plag(-cpx-magn) cumulate. Pechenga:
10-11 01 cumulate; 12 "Porphyritic" 01 cumulate; 13 Cpx cumulate; 14 Plag-cpx-magn cumulate.

tIl
:r:
§
E:
 The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 163

100 100
Pechenga

., !
.g ~c
c o
~ 10 J: 10
~ ~
.><
u
o
'"o
u

cr cr

o Plag- cplI: - magn cum.

a Plag- cpx- magn cum_
x Cpx, cpx- magn cum.
+ Cpx cumulate
• Ollv cumulate
• Oliy cumulate

La Ce Pr Nd P.. 5 .. Eu Gd Tb Dy Ho Er Tin Vb Lu La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tin Vb Lu

Fig. 9. Chondrite-normalized REE patterns for various cumulates from Pechenga and Koli

define a steeper trend having an Al 20 3 /Ti0 2 ratio of about 5.5. The analyses
from Pechenga lie on a line which gives a still lower AIP3/Ti02 ratio. The
average ratio calculated from the results of analyses of the well-preserved
cumulates from barren, promising and ore-bearing intrusions, as given by Zak et
al. (1982), is close to 3.6. This is near the value obtained from analyses of chilled
margins or as weighted averages for intrusions (see Table 4).

Rare Earth Elements

The rare earth element concentrations in the various cumulates are seen in Table
3 and are presented in chondrite-normalized form in Fig. 9. The REE patterns
for the olivine cumulates are straight and clearly LREE-enriched. The pyroxene
cumulates have a sloping curve for HREE but an almost flat or rising one for
LREE. The LREE pattern in the gabbroic rocks gradually changes from flat to
sloping with the increase in REE content (with height in the layered series). The
negative Eu anomally in the most differentiated sample from Koli is probably
caused either by an analytical error or by secondary alteration.
In the light of the known crystal/liquid distribution coefficients for REE, the
overall REE concentrations in the cumulates from Koli are consistent with the
LREE-enriched character of the parental magma and an origin for the various
cumulates by fractional crystallization from the same parental magma.

The Parental Magma

A close spatial and genetic relation to basic and ultra basic volcanic rocks is
regarded as a characteristic feature of the GWA (Dagelayskaya 1979). The chem-
164 E. Hanski

Table 4. Analyses of picrites from Onega and Pechenga and of chilled margins and weighted averages
of the Pechenga intrusions

1 2 3 4 5 6 7 8

Si02 44.18 45.59 46.56 46.25 42.08 43.54 42.66 40.92

Ti02 1.54 1.58 1.60 2.15 2.67 1.97 1.83 1.88
A120 3 8.80 9.39 9.52 7.46 9.01 7.38 6.37 6.34
F~03 1.89 2.11 2.46 1.98 4.45 5.29 3.48 6.05
FeO 9.68 9.10 8.83 12.92 11.60 12.23 14.13 10.34
MnO 0.18 0.15 0.15 0.20 0.19 0.18 0.22 0.18
MgO 17.77 14.56 13.32 17.16 13.76 18.91 20.54 19.02
CaO 8.99 9.10 8.90 6.87 9.56 8.51 6.67 6.38
Na20 0.84 1.08 1.43 0.47 0.40 0.62 0.94 0.85
K20 0.39 0.27 0.30 0.12 0.16 0.28 0.44 0.36
P20 S 0.19 n.d. n.d. 0.36 0.27 n.d. n.d. n.d.
Cr203 0.14 n.d. n.d. 0.17 0.11 0.25 0.26 0.20
NiO 0.11 n.d. n.d. 0.09 0.08 n.d. 0.17 0.14
Loss 5.43 7.32 4.95 3.57 6.08 n.d. 1.72 6.93
Total 100.13 100.29 99.02 99.77 100.42 99.16 99.43 99.59

= 10), Onega, (Svetov and Sokolov 1978); 2 Picrite (n = 8), Onega, (Svetov and
1 Picrite (no. of anal.
Golubev 1978); 3 Picrite (n = 17), Onega, (Kulikov 1971); 4 Picrite, Pechenga, (Predovskiy et a1.
1974); 5 Picrite, Pechenga, (Predovskiyet a1. 1974); 6 Chilled margin (n = 10), Pechenga, (Smol'kin
1974b); 7 Weighted average of Pechenga intrusions (n = 88), (Zhangurov and Predovskiy 1974);
8 Weighted average of Pechenga intrusions (n = 88), (Zhangurov and Predovskiy 1974); 8 Weighted
average of Pechenga intrusions (n = 246), (Rabinovich 1978).

ical similarity between the average compositions of the intrusions in the

Pechenga area or the composition of the chilled margins, and the picrites of the
fourth volcanic suite (see Table 4) has led many authors to suggest that the intru-
sions and picrites represent hypabyssal and effusive facies of the same
magmatism (Predovskiy et al. 1974, Smol'kin 1974b). In the Onega trough in
southern Karelia minor intrusions of the OWA are very closely associated with
Suisaarian picrites (Kulikov et al. 1976).
The molar MgO/(MgO + FeO) ratio (Mg' value) of the parental magma can
be calculated using the composition of the most magnesian olivines and the coef-
ficient determining the distribution of Mg and Fe between the olivine and the
liquid (Roeder and Emslie 1970). Using K D = 0.3, Mg' values 0.59 and 0.73 are
obtained for the parental magmas of Pechenga and Kuhmo, respectively, reflect-
ing the more differentiated nature of the magma at Pechenga.
The AlP3/Ti02 ratios and Mg' values of the parental magmas of the OWA
in Kuhmo and Pechenga are compared in Fig. 10 with corresponding data for
picrites from the Pechenga and South Pechenga Complexes, Suisaarian picrites
and komatiites from Kuhmo and Vetrenyi Poyas. The Kuhmo parental magma
comes very close to the Suisaarian picrites, while the picrites from Pechenga are
more primitive than the parental magma of the associated intrusions, but both
have very low AlP3/Ti02 ratios. The picrites from South Pechenga tend to have
higher AIP3/Ti02 ratios.
Many authors have referred to the Pechenga picrites as komatiites (e.g.
Suslova 1982). In spite of the descriptions of spinifex textures (Skufin 1980,
The Oabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 165

25 X Fig. 10. Al20 3 /Ti0 2 vs Mg'. Symbols: 1 parental

Alps x
x magma of the OW A in Kuhmo; 2 parental magma
20
Ti0 2
* XX X x
X
of the OW A in Pechenga; 3 Suisaarian picrite; 4
0
Ox
x 8) ~ i'Olit;P x
Pechenga picrite; 5 picrite from South Pechenga;
6 komatiite from Kuhmo; 7 komatiite from Vetrenyi
o~ xOx Xx 00
Poyas. References for the data sources are available
15 0
from the author
0
10

it <> +.

-
~<> +

...
5
I .' •
• :'.~.;'!; ..
0
0.5 0.6 0.7 0.8 0 .9
Mg'
0-<> · +x o
1 2 3 4 5 67

35 18
MgO Cb
...
Cl
30 16

... '., -:
• • t. . . .
Cl Cl
.....
25 14

..
.: • ... D
'.' <Il>

20 . ~. 12

·.
.: .0."'"
.... :-~ ~ o

.
15 10

...
~::)4~·
10
o
0
.~.
• •
0>
.
8 -..
5;-rT'-rT'-rT'-rT~rT~rr~r. 6;-ro,-ro'-rT'-;T'-ro,-ro, -ro
0 .5 0.6 0 .7 0 .8 0.9 0.5 0 .6 0.7 0.8 0.9
Mg' Mg'

Fig. 11. MgO and total FeO in the Pechenga and Suisaarian picrites compared with other primary
magmas. Symbols: 1 Suisaarian picrite; 2 Pechenga picrite; 3 komatiite; 4 Deccan picrite; 5 Karroo
picrite; 6 Skye picrite; 7 Baffin Bay picrite; 8 picrite from the Reykjanes Peninsula; 9 Hawaiian
tholeiite; 10 MORB; 11 ophiolite basalt; 12 alkali basalt; 13 basanite; 14 olivine melilite. 3 -14 taken
from Basaltic Volcanism Study Project (1981, Tables 1.4.2.1 and 1.4.2.2; for references see footnotes
to these Tables)

Rusanov 1981) they clearly differ chemically from komatiites, and not only in
their higher Ti0 2 content and lower AIP3/Ti02 ratio (Fig. 10). In fact, the
picrites of Pechenga differ from all other primary magma types in having very
low AIP3 and high MgO and total FeO concentrations at the same Mg' value
(Figs. 11 and 12). These features could be ascribed to a very great depth of origin
in the mantle. The primary phase field of garnet expands and that of olivine con-
tracts with increasing pressure, resulting in lower AIP3 and higher MgO and FeO
in the first melt produced (O'Hara 1968).
166 E. Hanski

20 Fig. 12. AI20 3 content of the Pechenga and Suisa-

arian picrites compared with other primary mag-
mas. Symbols and data sources as in Fig. 11
15
o • • ......
10

'b
5
D

O+-'--.-r-r..-r--r-r--.-r-r..-r-r-r-.-r-r~
0 .5 0 .6 0.7 0.8 0.9
Mg'

The (La/Yb)N ratio for the 01 cumulates, which reflects the ratio in the
parental magma, is clearly greater in the rocks of Pechenga (7.5 - 9.5) than at
Kuhmo and Koli (about 4.5). This fact can be interpreted as indicating a lower
degree of melting at Pechenga, provided that parental material was the same for
both areas. Assuming twice the chondritic REE content and garnet lherzolite
mineralogy (01600PX20CPXI0gtlO) of the parental material, and using the melting
proportions, distribution coefficients and equilibrium melting equation given by
Leeman (1977), Terakado (1980) and Shaw (1970), respectively, the melting per-
centages can be calculated to have been about 50/0 at Pechenga and 10% in
eastern Finland.

Areal and Temporal Correlation

In addition to Pechenga, intrusions of the OWA are known to occur in the

Imandra-Varzuga zone (Zhangurov and Smol'kin 1982), in the Kuolajarvi
(Lavrov 1979) and possibly Oaikolskaya structures (Sidorenko 1983), in the
Hautavaara greenstone belt (Popov 1971), in the Onega trough (Kulikov et al.
1976) and in Vetrenyi Poyas (Slyusarev and Kulikov 1973) within the Soviet part
of the Baltic Shield. In Finland rocks petrographically typical of the OW A can be
found at least in Puolanka, Kuusamo (Piispanen 1972), Kemi, Kittila, Sodankyla
and Savukoski in addition to the areas discussed earlier in this report (see Fig. 1).
In the light of the available geochemical data it seems probable that all the above
mentioned occurrences of the OW A, except that at Pechenga and some
intrusions in the Imandra-Varzuga zone, are related to a picritic magma of the
Suisaarian type rather of the Pechenga type. In contrast to Pechenga, only minor
Ni - eu deposits have been discovered in the other occurrences (Kayryak et al.
1976, Kojonen 1981).
The absolute age of the Pechenga intrusions is as yet somewhat uncertain.
The available age determinations for the intrusions and their host rocks, ob-
tained by the Pb isotope, K - Ar and Rb - Sr methods, give ages mostly between
The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 167

1700-1900 Ma (e.g. Gerling et al. 1972). This range is interpreted as reflecting

the time of metamorphism (e.g. Gorokhov et al. 1982).
The intrusions of the GWA and related volcanic rocks of Pechenga and other
areas in the Soviet part of the Baltic Shield have been correlated by many authors
with the Suisaarian volcanic-sedimentary formations (e.g. Sidorenko 1983).
Suisaarian formations, including picrites, have been found to be deposited con-
formably on the Jatulian supracrustal rocks (e.g. Sokolov and Stenar 1980). In
Finland intrusions of the GWA are encountered in the Archaean basement or
within the J atulian metasediments but not within the Kalevian metasediments un-
conformably overlying Jatulian rocks. Age determinations of 2.1 - 2.2 Ga
(Sakko 1971, Hypponen 1983) support a J atulian age for these, since the upper
boundary of the J atulian is interpreted as being about 2000 Ma (Merilainen and
Sokolov 1981). The Suisaarian rocks have not been dated, but their age is
thought to be not far removed from Jatulian (Kratts et al. 1981). Clearly more
work is needed to resolve the time-stratigraphic correlation for the GWA in the
Baltic Shield.

Conclusions

Many conclusions can be drawn from the data presented above, the most impor-
tant being:
1. The area of occurrence of the GWA is extensive, being of the order of
500 X 800km.
2. A reaction pair of olivine and Ca-rich pyroxene seems to be a characteristic
feature of the GWA.
3. The lack of orthopyroxene and the presence of primary hydrous minerals
indicate a low aSi<h and a high water content for the parental magma.
4. The parental magma of the layered bodies in Kuhmo was close to the primary
magma, but that of the Savilathi sill, Petajaniemi dykes and the Ni-bearing
intrusions of Suomussalmi and Pechenga suffered olivine fractionation before
emplacement.
5. In comparison with that in eastern Finland the parental magma in Pechenga
may be deduced from the chemical composition of the minerals and rocks to
have had a lower Si0 2 concentration and AIP3/Ti02 ratio and higher KINa
and (La/Yb)N values and is interpreted as having been produced by a lower
degree of melting at a deeper level in the mantle.
6. The parental magmas in eastern Finland and Pechenga correspond to picrites
of the Suisaarian and Pechenga types, respectively. The latter type differs
from other primary magmas in having a lower AIP3 and a higher MgO and
total FeO content at the same Mgt value.
7. Outside Pechenga only minor Ni - Cu deposits are known to be related to the
GWA.

Acknowledgements. I am deeply grateful to Prof. T.Piirainen for many stimulating discussions and
continuous interest in my work. I would also like to thank Mr. S. Sivonen, Dip\. Eng., for providing
the microprobe analyses and Mr. M. Hicks, M.A., for carefully reviewing the English translation.
168 E. Hanski

References

Auvray B, Blais S, Jahn BM, Piquet D (1982) Komatiites and the komatiitic series of the Finnish
greenstone belts. In: Arndt NT, Nisbet EG (eds) Komatiites. Allen & Unwin, London, pp 131-146
Basaltic Volcanism Study Project (1981) Basaltic volcanism on the terrestrial planets. Pergamon
Press, New York
Bilibina TV (ed) (1980) Map of the geological formations in the eastern part of the Baltic Shield, Scale
1: 150000. In: Sidorenko AV, Bilibina TV (eds) Metallogeny of the eastern part of the Baltic
Shield. Nedra, Leningrad (in Russian)
Blais S, Capdevila R, Jahn BM, Hameurt J (1978) The Archean greenstone belts of Karelia (Eastern
Finland) and their komatiitic and tholeiitic series. In: Windley BF, Naqvi SM (eds) Archean geo-
chemistry. Elsevier North Holland Biomedical Press, Amsterdam New York, pp 87 -107
Dagelayskaya IN (1979) The gabbro-wehrlite formation. In: Masaytis VL (ed) Magmatic formations
of the USSR, vol I. Nedra, Leningrad, pp 257 - 265 (in Russian)
Gerling EK, Lobach-Zhuchenko SB, Gorokhov I, Koltsova TB (1972) Isotopic ages of rocks of the
Precambrian in the Baltic Shield. In: Kratts KO (ed) Geochronological boundaries and geological
evolution of the Baltic Shield. Nauka, Leningrad, pp 76 - 159 (in Russian)
Gorbunov GI (1%8) Geology and origin of the copper-nickel sulphide deposits of Pechenga. Nedra,
Moscow (in Russian)
Gorokhov 1M, Varshavskaya ES, Kutyavin EP, Melnikov NN (1982) Rb-Sr dating of low-grade
metamorphics of the USSR. Precambrian Res 18:145 -156
Hanski E (1982) A comparative study of albite diabases and related ultramafic rocks in the Kuhmo
and Koli areas, eastern Finland. Ark Alueiden Malmiproj Rap 6, Oulun Yliopisto (in Finnish with
English summary)
HyppOnen V (1983) Pre-Quaternary rocks of the Ontojoki, Hiisijllrvi, and Kuhmo map-sheet areas.
Explanation to the maps of Pre-Quaternary rocks, Sheets 4411,4412 and 4413,1: 100000. Geol
Surv Finl (in Finnish with English summary)
Irvine TN (1974) Petrology of the Duke Island Ultramafic Complex, southeastern Alaska. Mem Geol
Soc Am 138
Kayryak AI, Morozov SA, Limonova LP, Popov MG (1976) Copper-nickel and polymetallic pyrite
ores in eastern Karelia. Razved Okhr Nedr 8:2 -7 (in Russian)
Kojonen KK (1981) Geology, geochemistry and mineralogy of two Archean nickel-copper deposits in
Suomussalmi, eastern Finland. Geol Surv Finl Bull 315
Kratts KO, Khil'tova VYa, Vrevskiy AB, Zapolnov AK, Krylov IN, Lobach-Zhuchenko SB,
Manuilova MM, Platunova AP (1981) Stages and types of evolution of the Precambrian crusts in
the Ancient Shields. Nauka, Leningrad (in Russian)
Kulikov VS (1971) The basaltic complex ofVetrenyi Poyas and some causes of variation in the Middle
Proterozoic volcanism in the southeastern part of the Baltic Shield. In: Bel'kov IV, Shurkin KA
(eds) Problems of the magmatism of the Baltic Shield. Nauka, Leningrad, pp 196-202 (in
Russian)
Kulikov VS, Slyusarev VD, Kochnev-Pervukhov VI, Kravchenko AN (1976) The Suisaarian basic-
ultrabasic complex in the Onega synclinorium. In: Kratts KO (ed) Intrusive basic-ultrabasic
complexes of the Precambrian of Karelia. Nauka, Leningrad, pp 98 -109 (in Russian)
Lavrov MM (1979) Hyperbasites and layered peridotite-gabbro-norite intrusions of the Precambrian
in Northern Karelia. Nauka, Leningrad (in Russian)
Leake BE (1978) Nomenclature of amphiboles. Can Miner 16:501 - 520
Le Bas MJ (1962) The role of aluminium in igneous clinopyroxenes with relation to their parentage.
Am J Sci 260:267 - 288
Leeman WP (1977) Mineral constitution of mantle source regions for Hawaiian basalts - rare earth
element evidence for mantle heterogeneity. In: Dick HJB (ed) Magma genesis. Or Dep Geol Miner
Ind Bull 96:169 -183
Lehtinen M (1983) On the geology and Ni deposit of the Arola area, Kuhmo. M Sc Thesis, Univ
Turku (in Finnish)
Masaytis VL, Moskaleva VN, Rumyantseva NA (1979) General problems in the study of magmatic
formations. In: Masaytis VL (ed) Magmatic formations of USSR, vol I. Nedra, Leningrad, pp
5 - 25 (in Russian)
The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield 169

Merilliinen K, Sokolov VA (1981) Some attainments and problems relating to the geology of the
Jatulian formations in Finland and Soviet Karelia. In: Puustinen K (ed) Geological, geochemical
and geophysical investigations in the eastern part of the Baltic Shield. Committee for scientific and
technical co-operation between Finland and the Soviet Union, Helsinki, pp 9 - 18
Murray RJ (1954) The clinopyroxenes of the Garbh Eilean sill, Shiant Isles. Geol Mag 91:17 - 31
Nwe YY (1976) Electron-probe studies of the early pyroxenes and olivines from the Skaergaard
intrusion, east Greenland. Contrib Mineral Petrol 55:105 -126
O'Hara MJ (1968) The bearing of phase equilibria studies in synthetic and natural systems on the
origin and evolution of basic and ultrabasic rocks. Earth Sci Rev 4:69 -133
Piirainen T, Honkamo M, Rossi S (1974) A preliminary report on the geology of the Koli area. Bull
Geol Soc Finl 46:161 -166
Piispanen R (1972) On the spilitic rocks of the Karelidic belt in western Kuusamo, Northeastern
Finland. Acta Univ Ouluensis Ser A 4
Piquet D (1982) Mecanismes de recristallisations metamorphiques dans les ultrabasites: exemple des
roches vertes archeennes de Finlande orientale (Ceintures de Suomussalmi-Kuhmo). These 3. cycle,
Univ Rennes
Popov MG (1971) Proterozoic magmatic complexes of Khjursjulya-Hautavaara (southern Karelia).
In: Bel'kov IV, Shurkin KA (eds) Problems of the magmatism of the Baltic Shield. Nauka,
Leningrad, pp 211 - 217 (in Russian)
Predovskiy AA, Fedotov ZhA, Akhmedov AM (1974) Geochemistry of the Pechenga Complex.
Nauka, Leningrad (in Russian)
Rabinovich YuI (1978) Petrochemistry and nickel content of basic and ultrabasic rocks of some
intrusive complexes in the Pechenga-Allarechensk area. In: Pokrovskiy AD (ed) Additional criteria
for prospecting ore deposits in Karelia and the Kola Peninsula. Min Geol RSFSR, Moscow (in
Russian)
Roeder PL, Emslie RF (1970) Olivine-liquid equilibrium. Contrib Mineral Petrol 29:275 - 289
Rusanov MS (1981) Tholeiite-komatiite formations of the Pechenga Complex. SOY Geol 2:98 -112
(in Russian)
Sakko M (1971) Radiometric zircon ages of the Early-Karelian metadiabases. Geologi 23:117 -118
(in Finnish with English summary)
Shaw DM (1970) Trace element fractionation during anatexis. Geochim Cosmochim Acta
34:237 -243
Sidorenko VV (1983) Major geochemical characteristics of basic-ultrabasic magmatism in the
Karelia-Kola region. SOY Geol 3:99-106 (in Russian)
Simonen A (1980) Prequaternary rocks of Finland, 1: 1000000. Geol Surv Finl
Skufin PK (1980) Characteristics of the volcanism in the Proterozoic Pechenga structure. Pyul.
Mosk. O-va Ispyt. Prirody. Otd GeoI55:120-131 (in Russian)
Slyusarev VD, Kulikov VS (1973) Geochemical evolution of the Proterozoic basic-ultrabasic
magmatism (SE Baltic Shield). Nauka, Leningrad (in Russian)
Smol'kin VF (1974a) Layering in basic and ultrabasic rocks of the ore-bearing Pil'guyarvi intrusion
(Pechenga). In: Bel'kov IV (ed) Petrology, mineralogy and geochemistry. KOl'skiy Filial AN
SSSR, Apatity, pp 24-47 (in Russian)
Smol'kin VF (1974b) Composition of the layered intrusions of Pechenga reconstructed according to
mineral relicts. In: Bel'kov IV (ed) Petrology, mineralogy and geochemistry. Kol'skiy Filial AN
SSSR, Apatity, pp 48 - 54 (in Russian)
Smol'kin VF (1978) Olivine and clinopyroxene of the Pil'guyarvi gabbro-wehrlite intrusion
(Pechenga). In: Tamarinov PM (ed) Minerals and mineral parageneses. Nauka, Leningrad, pp
55 -71 (in Russian)
Smol'kin VF, Pahomovskiy YaA (1982) Nickel content of olivines and chrome spinels from
ultramafic rocks of the Pechenga Ni-bearing complex. In: Minerals and mineral parageneses of
copper-nickel and pyrite ores in the Kola Peninsula. Akademiya Nauk SSSR, Apatity, pp 54 - 63
(in Russian)
Sokolov VA, Stenar MM (1980) Principles of subdivision and correlation in the Precambrian of the
south-eastern part of the Baltic Shield. In: Mitrofanov FP (ed) Principles and criteria of sub-
division of the Precambrian in mobile zones. Nauka, Leningrad, pp 140-173 (in Russian with
English abstract)
Stormer JC (1973) Calcium zoning in olivine and its relationship to silica activity and pressure.
Geochim Cosmochim Acta 37:1815 -1821
170 E. Hanski: The Gabbro-Wehrlite Association in the Eastern Part of the Baltic Shield

Suslova SM (1982) Petrochemical characteristics of the early Precambrian basaltic volcanism in the
Kola Peninsula. SOy Geol 6:85 - 98 (in Russian)
Suslova SN, Staritsyna GN, Dagelayskaya IN (1981) Nickeliferous ultramafic-mafic volcanic-
intrusive complexes of the Karelia-Kola and North Siberian regions. Int Geol Rev 23:1135 -1147
Svetov AP, Golubev AI (1978) The volcanic complex of Suisaari. In: Sokolov VA (ed) Volcanic
complexes of the Proterozoic in Karelia. Nauka, Leningrad, pp 119 - 128 (in Russian)
Svetov AP, Sokolov VA (1978) The volcanic complex of Shidguda. In: Sokolov VA (ed) Volcanic
complexes of the Proterozoic in Karelia. Nauka, Leningrad, pp 111 -119 (in Russian)
Taipale K (1983) The geology and geochemistry of the Archean Kuhmo greenstone-granite terrain in
the Tipasjilrvi area, eastern Finland. Acta Univ Ouluensis Ser A 151
Terakado Y (1980) Fine structures of rare earth element patterns of Tahitian rocks. Geochem J
14:155-166
Tkachenko KN (1982) The Pechenga Complex. In: Gorbunov GI (ed) Chemical analyses of basic-
ultrabasic rocks of the Precambrian complexes in the Kola Peninsula. Kol'skiy Filial AN SSSR,
Apatity, pp 167 - 206 (in Russian)
Vayrynen H (1938) Petrologie des Nickelerzfeldes Kaulatunturi-Kammikivitunturi in Petsamo. Bull
Comm Geol Finl 116
Wilkinson JFG (1957) Clinopyroxenes of a differentiated teschenite sill near Gunnedah, New South
Wales. Geol Mag 94:123-134
Yakovleva AK (1981) Amphiboles. In: Gorbunov GI (ed) Mineralogy of the copper-nickel ores of the
Kola Peninsula. Nauka, Leningrad, pp 96 - 127 (in Russian)
Yakovleva AK, Osokin AS (1981) Micas. In: Gorbunov GI (ed) Mineralogy of the copper-nickel ores
of the Kola Peninsula. Nauka, Leningrad, pp 127 -142 (in Russian)
Zak SI, Makarov VN, Kochnev-Pervukhov VI, Proskuryakov VV, Zaskind ES, Batashev EV,
Kolesnikov GP (1982) Geology, magmatism and ore formation in the Pechenga ore belt. Nedra,
Leningrad (M-vo Geol RSFSR) (in Russian)
Zhangurov AA, Predovskiy AA (1974) Geochemistry of the rocks and the character of the parental
magma of the Ni-bearing intrusives in Pechenga. In: Bel'kov IV (ed) Petrology, mineralogy and
geochemistry. Kol'skiy Filial AN SSSR, Apatity, pp 72 - 83 (in Russian)
Zhangurov AA, Smol'kin VF (1982) Basite-hyperbasites of the middle and late Karelian and
Caledonian stages. In: Gorbunov GI (ed) The Karelian Imandra-Varzuga Zone (Geology, geo-
chemistry and history of evolution). Nauka, Leningrad, pp 145 -165 (in Russian)
Zhangurov AA, Petrov VP, Smol'kin VF, Voloshina ZM, Smirnova TN (1974) Pyroxenes of the dif-
ferentiated intrusives of Pechenga. Mat. Po Miner. Kol'sk. Po-va 10. 52 - 61. Nauka, Leningrad,
pp 52 - 61 (in Russian)
Peridotitic Komatiites and the Origin of Ores
(Southeastern Part of the Baltic Shield)
v. I. KOCHNEV-PERVUKHOV\ E. s. ZASKIND\ and V. V. PROSKURYAKOV 2

Abstract

Komatiite magmatism of the Archean greenstone belts of the southeastern part

of the Baltic shield is considered here. The conformable bedding and inter-
bedding of bodies of komatiites with pillow basalts and horizons of tuffs, with
breccias at their tops, pillows, amygdules, glass and spinifex textures permit the
determination of these rocks as peridotite komatiite lavas. Plutonic analogs of
these rocks are found in ultramafic effusion feeders. Nickel-copper ore mineral-
ization is orthomagmatic and is associated with komatiitic intrusives. The nickel
ore potential of units associated with komatiite magmatism seems .to be cor-
related with the appearance of island-arc basalts: the larger these are, the smaller
the ore potential of the structures. And, in this sense, the nearest analogs of the
belts in the Baltic shield are the Canadian greenstone belts, not the Western
Australian. These may be considered geotectonically as nickeliferous provinces.
The southeastern part of the Baltic shield is a region of wide spreading green-
stone belts; these are filled with Archean series and form a complicated network
of the narrow elongated structures within gneissose granite blocks (Fig. 1). The
extension of the belts ranges from 1 - 2 km to a few tens of kilometers. In a
number of such structures, komatiitic series are found (Fig. 1). The authors
studied the volcanic-plutonic komatiite complex in the Kenozero-Kozhozero
greenstone belt.
Komatiitic bodies have ~arious lengths and thicknesses and conform with
basalt sheets. Deformational and metamorphic changes in such basalts and
komatiites are much the same. The generalized section of the researched bodies
consists of two parts (Fig. 2): the upper, less thick part, is composed of breccias
and vitreous and spinifex rocks; the lower part of serpentinite after ordinary
peridotite and dunite. The ratio of thicknesses of the upper and lower parts fluc-
tuates from 1: 10 to 1: 50.
The glass, the skeletal forms of magmatic minerals, and the spinifex textures
are doubtless evidences of the crystallization of these rocks from a melt under
high temperature-gradient conditions. The composition of the melt is determined
from the composition of altered glass and rocks with spinifex textures (Table 1),
and corresponds to that of peridotite komatiites (Arndt et al. 1979, Brooks and

1 The Central Research Exploration Geology Institute of Basic and Precious Metals, Moscow, USSR
2 The Industrial Geological Trust "Sevzapgeologiya", Leningrad, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et aL)
© Springer-Verlag Berlin Heidelberg 1986
172 V.l. Kochnev-Pervukhov et al.

75km

+
...

Fig. 1. Extending greenstone belts in the southeastern part of the Baltic shield. 1 proterozoic struc-
tures; 2 archaean greenstone belts; 3 granite-gneissic block; 4 belomorides. Greenstone belts
designated by the circles numbers: 1 Kenozero-Kozhozero; 2 Toksha; 3 Koikara; 4 Palaya Lamba;
5 Kostomuksa

" v v v
~l ~4

.
,
~2 0 5
a
~ If ~ 3 1-'-16 I-I
b
~
~ ~

-
~
- - --
11 .
'" it "
Fig. 2. Typical section of peridotitic komatiite flow. 1 Basalts; 2 breccias
~ " "
[ 2.5 m
and ultramafic glass with amygdules; 3 spinifex texture: a random
("birdtrack" after Naldrett and Mason 1968) spinifex, b sheaf spinifex;
p It
.
_ 1_1_1--1
t1
4 apoperidotitic serpentinites with a small block of sheaf spinifex;
5 apodunitic serpentinites; 6 chlorite-Quartz-carbonate shale after acidic
tuffs
Peridotitic Komatiites and the Origin of Ores (Southeastern Part of the Baltic Shield) 173

Table 1. The chemical composition of ultramafic rocks in the Kenozero-Kozhozero greenstone belt"

Nb Si0 2 Ti02 Al 20 3 FeO MnO MgO CaO NiO Cr203 Total

1 44.9 0.34 6.0 11.3 0.22 30.5 5.8 0.11 99.2

2 44.8 0.39 6.5 11.3 0.19 29.6 6.3 0.12 0.33 99.5
3 46.2 0.33 4.1 12.2 0.18 31.7 4.5 0.26 0.34 99.8
4 45.1 0.20 2.8 9.4 0.15 39.6 2.1 0.15 0.35 99.8
5 43.7 0.10 0.8 10.1 0.12 42.1 1.5 0.29 0.23 98.9

" Analysis was done by AA Kuleschova. The analyses were recalculated as water free to 1000/0.
b 1 Altered glass from peridotitic komatiites; 2 rocks with spinifex texture; 3 average composition of
the intrusive bodies (without the olivine cumulate); 4 dunite from peridotitic komatiite flows;
5 dunite in komatiitic intrusives.

Hart 1974), described in Africa (Viljoen and Viljoen 1969), Canada (Pyke et al.
1973), Australia (Ross and Hopkins 1975), and the Russian Platform (Krestin
1978).
The conformable bedding and interbedding of bodies of peridotitic komati-
ites include flows of pillow basalts and horizons of tuffs, the tops of which
contain breccias, pillows, amygdules, and glasses, permit the determination of
the rocks studied as effusives.
The actual discovery of ultramafic effusive rocks among the Archean series
of the greenstone belt is important because it gives additional possibilities for the
discussion of two questions: (1) geologic age and formation type of the ultra-
mafic intrusions of this region, and (2) the general features of the succession
formed, and their connection with, the ore potential of the komatiite series that
are formed in different geotectonic settings. The former has importance only for
the region studied, but the latter is important for any structures containing
komatiites.
1. In the terrain where the Kenozero-Kozhozero greenstone belt is situated,
ultramafic intrusions are widespread. These intrusions are considered as
Proterozoic and are compared with ones of the Vetrenii Poyas structure, which is
the younger of the greenstone belts. Within this intrusive complex, it is possible
to designate a group of bodies, the composition of which (Table 1) is analogous
to the peridotitic komatiites. There intrusives are contained within or near the
greenstone belt and are located in horizons that underlie rock masses that include
komatiites. Usually these are steep dipping isometric or extensive bodies with
considerable predominance of dunite and wehrlite. Peculiarities of the zoning
and the position of these bodies in the belt of rock sequences show symmetrical
zoning relative to the central part of the intrusive body in which the acid differ-
entiates are confined to the contacts. These are localized in horizons that underlie
the komatiite-bearing parts of the column that occur inside the domain of the
komatiitic effusive rocks and transitions of steep-dipping bodies to bedding-
conformable komatiitic flows with spinifex textures. These relations permit then
to be considered as feeders of ultramafic effusions, and to be included in the
Archean komatiite series; they are excluded, however, from the magmatism asso-
ciated with Proterozoic structures. Their communality of origin is confirmed by
174 V. I. Kochnev-Pervukhov et al.

the metalliferous specialization of the rocks: the accessory ore mineralization in

effusive and plutonic facies of the komatiites is qualitatively the same - in such
sulfide material as is present, pentlandite predominates, and the Ni: eu ratio
ranges widely, although essential segregations of the ore minerals occur only in
intrusions, and the ores have features of an orthomagmatic origin. The meta-
morphic process does not lead to accumulation of the sulfide substance, it results
only in a modification of textural-structural peculiarities of the mineralization, in
the decrease of the Ni: eu ratio and in the appearance of a series of impurities
that are nontypical for magmatic segregations. The metamorphism also leads to
the appearance of scattered heazlewoodite mineralization.
2. The rock sequence of the greenstone belt, to which the studied region
belongs consists of three formations: (1) the nondifferentiated basaltic, (2) the
sequentially differentiated volcanic-sedimentary dacite-liparite-andesite-basaltic,
and (3) the komatiite-basaltic. The base of the rock sequence is unknown, and
the age of the granitoids, among with the greenstone belt is located, has not
reliably been determined. In the interpretation of the rock sequence, the posi-
tions of the amygdaloidal zones and of the pillows are confined to basaltic flows,
and those of the spinifex are located in the peridotitic komatiites. On these bases,
the komatiite-basaltic formation is above and the basaltic nondifferentiated
formation is below the sequentially differentiated volcanic-sedimentary forma-
tion. It is not excepted, however, that the nondifferentiated basaltic formation
would have been a facies of the komatiite-basaltic formation with the transition
from one to the other being along the strike. But even if this is true confining of
the komatiite-basaltic formation to the upper part of the column and its position
above the sequentially differentiated formation would not change.
The lower formation consists of pillow and massive tholeiite basalts, among
which the few flows of ferriferous varieties and the subvolcanic bodies of the
quartz-porphyry were found. These basaltic flows are separated by thin (less than
1 m) intercalations of the slates. The rock sequence of the formation is wonder-
fully uniform.
The middle formation consists (from base to top) of basalts AI-basalts,
andesite-basalts, andesites, and dacite-liparites that are interbedded with
horizons of the intermediate and silicic tuffs and rhythmic tuffaceous flysch (the
psammitic parts of the rhythms consist of silicic material, in which there are
plagioclase and fragments of acidic effusive rocks, but grains of quartz are never
found. The changes in the succession take place along the strike: in the northern
part of the belt the succession is essentially volcanogenic and includes inter-
mediate and acidic tuffs; the central part is the tuffogenic flysch with a single
dacite-liparitic flow. The southern part of the belt is composed of ferro-cherty
rocks of the Algoman type (Dimroth 1975) with rare persistent beds of black
shales and acidic tuffs. Flows of andesites occurred only in the northern part of
the belt.
The massive sulfide mineralization of three types (after Smirnov 1968) has
been widely developed:
1. In the northern part of the structure, this mineralization is hydrothermal
copper-zinciferous (eu > Zn) ores in quartz-sericite and quartz-chlorite meta-
somatites, following the silicic effusives and tuffs and the exhalative-sedimentary
Peridotitic Komatiites and the Origin of Ores (Southeastern Part of the Baltic Shield) 175

pyritic mineralization with the copper and zinc in silicic tuffs that are part of the
quartz-sericite metasomatites in the footwall.
2. In the middle part of the belt is a sedimentary pyritic mineralization (in the
pelitic parts of the tuffogenic flysch) with abundant pyritic concretions and
nodules.
3. In the southern part of the belt is a sulfide-oxide mineralization contained
in the cherty rocks.
The zoning of the mineralization of the various types and the association of
pyrites with one and the same silicic volcanites suggest consanguinity and asso-
ciation in time.
The higher formation consists of pillow and massive tholeiitic basalts with a
horizon of the peridotitic komatiites at its base. These basaltic flows inter bed
with basic tuffs and the black shales, whereby the rare pyritic mineralization, has
undergone redeposition and pyrrhotinization. The composition of the basalts is
the same as in the lower formation and differs only in the absence of the fer-
riferous varieties.
It is obvious that the characteristics of the rock sequence and of the ore
'mineralization of the middle formation are analogous of the ones of the island
arcs.
The analysis of the composition of the basalts of the belt by Pearce's method
(Pearce et al. 1977) (Fig. 3), has shown that two petrochemical types can be dis-
tinguished, which are the oceaniclike basalts (in the lower and higher formations)
and the basalts of the island arcs (only in the middle formation).

s FeO'

51-56 %
Si 02

MgO

Fig. 3. MgO - FeO - AI 20 3 for basalts on the Archean greenstone belts: Western Australia, Yilgarn
block (a); Canadian shield - Yellowknife (b), and Abitibi (c); Baltic shield, the Eastern Karelia (d).
In d x represents basalts of the upper and lower parts of the rock sequence; indicates the same from
the middle part. These are designated by numbers: 1 basalts of middle-oceanic ridges and oceanic
floor; 2 basalts of oceanic islands, which are associated with interplate volcanism and the series of
oceanic rises; 3 continental platobasalts and traps; 4 basalts of oceanic island, which are located along
the axis of middle-oceanic ridges; 5 basalts of island-arcs and active continental margins
Table 2. Nickel-bearing provinces ....-..I
Intercontinental systems of epicratonic zones of rifting Zones of rifting in margin continental island-arc systems
'"
Autonomic intercontinental Rift zones connected with Volcanic island-arcs Marginal seas, microcon-
rift and riftlike zones initial stage of passive tinents or regions of early
and regions continental margins forming consolidation

A. Epiarchean

Siberian platform (Mzj) Greenland (Kz) - Skaergaard Northwest Pamirs (Pz3) Kamchatka (Mz2 )
Norilsk Canadian shield coppermine Baltic shield (Pt!) Central Kazakhstan (Pz!)
South Africa (Mz!) - Insizwa district (R 2 ) - Muskox Baltic south Finland, Ziatogorskii complex
Canadian shield (R 2 ) Keweenawan shield, Finland (Pt j ) central Sweden Aldan shield (R3) -
district - Duluth Baltic Kemi-Suhanko North Baikal region
shield (Pt!) - Pechenga Baltic shield (Ptj) - the
Canadian shield (Pt!) - Ungawa, main sulfide belt of Finland
Sudbury, Thompson belt Voronezh Shield (Pt!)
Transvaal Craton (Pt!)
Bushveld

B. Archean

West Australia Transvaal shield - Canadian shield - Abitibi; Yellowknife;

Yilgarn block greenstone belt Barberton Rhodesian shield; West Australia - Pilbara block;
Limpopo belt - Selebi-Pikwe Baltic shield - Granulitic belt of Lapland, :<
East Karelian belts ~
::-::
o
§"
(1)

~
'"C
l:!i
~
:0;"
:0-
~
~
~
Peridotitic Komatiites and the Origin of Ores (Southeastern Part of the Baltic Shield) 177

An analogous study was undertaken previously for the Archean greenstone

belts of the world according to literature data (Zaskind and Ivanjuta 1981) and
this has allowed us to distinguish three types of sequences: without island-arc
basalts (Western Australia, Barberton), with island-arc basalts only (Yellowknife
in Canada), and sequences of mixed rock type (Abitibi in Canada, Rhodesian
belts). If the Barberton belt is not taken into account, the scale of ultramafic
magmatism and its nickel ore potential depend on the sequence type. In fact,
sequences showing an island-arc rock type does not contain nickel-copper ores at
all. Sequences of the mixed succession type contain small deposits and large or
prominent deposits of nickel-copper ores are connected with sequences lacking
island-arc associations. Therefore, the presence in the sequence of island-arc
associations is a negative indication of the lack of nickel-copper ores in them. In
this sense, the nearest analogs to the belts of the region studied are the Canadian
greenstone belts, not those of Western Australia. However, judging by the
example of Western Australia and Barberton, the nickel-copper ore potential of
Archean sequences with komatiites belongs only to one of enumerated types; any
rock sequence with indications of the island-arc development has low ore poten-
.tial, and no sequence without komatiites has a high ore potential.
Literature data on the tectonic environments of geologic complexes (Peive et
al. 1972, Piirainen et al. 1974, Hain 1977, Archibald et al. 1978, Condie 1980,
Baragar and Scoates 1981, Barton and Key 1981, Jackson and Iannelli 1981,
Gaal 1982, Sims et al. 1982, Milanovskii 1983 and other) and their comparison
with the ones in nickeliferous epiarchean provinces, for which the geotectonic
conditions can be determined with more confidence, confirm the above. The
initial point of the paleotectonic analysis, the results of which are in Table 2, is
the Sawkins' model (Sawkins 1972). This model was enhanced by the ideas of a
definite system of the rifting zones (triple junctions discussed by Burke and
Dewey 1973). The main conclusion following from this scheme is that all the
most important provinces throughout the world are confined to the regions of
triple junctions that are intercontinental in the paleotectonic sense and evidently
correspond to zones of epicratonic rifting. These regions can have a spatial con-
nection with margin-continental geostructures, and they can develop out of this
connection. In this sense, the authors consider them as autonomous. Probably,
in Archean time, the belts of the Yilgarn block (Western Australia) corresponded
to structures of this kind. Therefore, it is possible to conclude that the tendencies
for a connection between magmatism and geotectonic position, which are deter-
mined in younger regions, can be traced (though it cannot be so clearly seen) in
the older ones and can be used for the classification and evaluating of the latter.

References

Archibald NS, BeUenay LF, Binns RA, Groves 01, Gunthorpe RS (1978) The evolution of Archean
greenstone terrains, Eastern Goldfields Province, Western Australia. Precambrian Res 6:103 -131
Arndt NT, Francis J, Hynes AJ (1979) The field characteristics and petrology of Archean and
Proterozoic komatiites. Can Miner 17, 2:147 -186
178 V.1. Kochnev-Pervukhov et al.: Peridotitic Komatiites and the Origin of Ores

Baragar WRA, Scoates RFJ (1981) The Circum-Superior belt: a proterozoic plate margin. In: Kroner
A (ed) Precambrian plate tectonics. Elsevier, Amsterdam Oxford New York, pp 297-330
Barton SM, Key RM (1981) The tectonic development of the Limpopo mobile belt and the evolution
of the Archean cratons of southern Africa. In: Kroner A (ed) Precambrian plate tectonics.
Elsevier, Amsterdam Oxford New York, pp 185-211
Brooks CK, Hart SR (1974) On the significance of komatiites. Geology 2:107 -110
Burke KP, Dewey JF (1973) Plume-generated triple junctions: Key indicator in applying plate
tectonics to old rocks. J Geol 81, 4:406 - 432
Condie KC (1980) Origin and early development of the earth's crust. Precambrian Res 11 :183 -197
Dimroth E (1975) Paleo-environ of iron-rich sedimentary rocks. Geol Rundsch 64, 3:751-766
Eriksson KA (1982) Sedimentation patterns in the Barberton mountain land, South Africa, and the
Pilbara block, Australia: Evidence for Archean rifted continental margins. Tectonophysics 81,
3/4:179-193
Gaal G (1982) Proterozoic tectonic evolution and late Svekokarelian plate deformation of the Central
Baltic Shield. Geol Rundsch 71,1:120-129
Hain VE (1977) Problems of the Precambrian Geology. In: Dagelaiskiy VB (ed) Problems of the
geology of Early Pre-Cambrian. Nauka, Leningrad, pp 5 -12 (in Russian)
Jackson GD, Iannelli TR (1981) Rift-related cyclic sedimentation in the neohelikian borden basin,
northern Baffin Island. Geol Surv Can Pap 81-10:269 - 302
Krestin EM (1978) The first find of komatiites in USSR. Rep AN USSR 242-2:412-415 (in Russian)
Milanovskii EE (1983) Rifting in the earth's history. Nedra, Moscow (in Russian)
Naldrett AJ, Mason GD (1968) Contrasting Archean ultramafic igneous bodies in Dundonald and
Clergue townships, Ontario. Can J Earth Sci 5:111-143
Pearce TH, Gorman BE, Birkett TC (1977) The relationship between major element chemistry and
tectonic environment of basic and intermediate volcanic rocks. Earth Planet Sci Lett 36:121-132
Peive AV, Schtreiss NA, Mossakovskii AA (1972) Paleozoic of the Eurasia and some questiones of
evolution of the geosyncline process. SOy Geol 12:7 - 25 (in Russian)
Piirainen T, Hugg R, Isohanni M, Juopperi A (1974) On the geotectonics and ore forming processes
in the basic intrusive belts of Kemi-Suhanko, and Syote-Narankavaara, northern Finland. Bull
Geol Soc Finl 46, 2:93 - 104
Pyke DR, Naldrett AJ, Ekstrand OR (1973) Archean ultramafic flows in Munrotownship, Ontario.
Geol Soc Am Bull 84:955 - 978
Ross IR, Hopkins GMF (1975) Kambalda nickel sulfide deposits. In: Knight CL (ed) Economic
geology of Australia and Papua New Guinea. I. Metals. Aust Inst Min Met, Mon, pp 100-121
Sawkins FJ (1972) Sulfide ore deposits in relation to plate tectonic. J Geol 80, 4:377 - 397
Sims PK, Card KD, Lumbers SB (1982) Evolution of early proterozoic basins of the Great Lakes
region. Pap Geol Surv Can 81-10:379-397
Smirnov VI (1968) Pyrite deposits. In: Smirnov VI (ed) Genesis of the endogenic ores deposits.
Nedra, Moscow, pp 586-647 (in Russian)
Viljoen MJ, Viljoen RP (1969) Evidence for the existence of a mobile extrusive peridotitic magma
from the Komati Formation of the Onvervaht Group. Geol Soc S Afr Spec Pub I 2:87 -113
Zaskind ES, Ivanjuta ZF (1981) The some peculiarities of the petrochemistry ofnickeliferous region's
basalts. In: Stepanov VK (ed) The genesis and localization of the nickel-copper ores. Moscow, p 96
(in Russian)
Geologic Setting of Selected Chromium and Nickel
Deposits of China
P.-F. FAN1

Abstract

Five selected chromium deposits of China are described: Luobusa of Xizang

(Tibet), northwestern and northern Inner Mongolia, western Henan, and
Panzihua of Sichuan. The first four are podiform deposits and the fifth is
stratiform. Six selected nickel deposits of China are described: Jinchuan of
Gansu, Dali and Ailaoshan of Yunnan, weathering products near Ailaoshan, and
sedimentary nickel deposits of Yangzi craton. These nickel deposits can be
classified into four types according to their geologic settings:
1. Rift valley deposits - Jinchuan and Dali
2. Obducted ophiolites - Ailaoshan
3. Weathering products - near Ailaoshan
4. Sedimentary rocks - Lower Cambrian black shales of Yangzi craton

Introduction

The chromium and nickel deposits of China described here were chosen for dis-
cussion because of the availability of published information. They represent
various types of deposits and most of them are considered major deposits. The
chromium deposits are: the Luobusa of Tibet, the northwestern and northern
Inner Mongolia, the western Henan, and the Panzihua of Sichuan. The nickel
deposits are: the Jinchuan of Gansu, the Dali of western Yunnan, and the
Ailaoshan of southern Yunnan (Fig. 1).
A number of chromium occurrences have been reported in China, but most
of them are small. The largest occurrences reported are in the Zhayier
Mountains, northwest of the Junggar basin, 400 km west of Uramchi, the capital
of Xinjiang. The reserves are estimated at 1 million tons of 35"70 Cr203 ore.
Production of chromium in Xinjiang began in 1967 (Brady 1981, Chin 1983a,
Mikami 1983), but no geological reports are available. A number of high-grade
chromium deposits have been found in Tibet, and chromium has also been found
in Panzihua titanium iron ore.

I Hawaii Institute of Geophysics, University of Hawaii, Honolulu, HI 96822, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
180 P.-F. Fan

-,
\"
"'-"'.../1
Heilonoiiong i

Fig. 1. Location of chromium and nickel deposits of China that are discussed here. Symbols: • Cr; X
Nickel; 1 Luobusa; 2 N .W. Nei Monggol (Inner Mongolia); 3 N. Nei Monggol (Inner Mongolia);
4 W. Henan; 5 Panzihua; 6 linchuan; 7 Oali; 8 Ailaoshan

The nickel reserves of China have been estimated at 7 million tons of con-
tained nickel. The largest reserve, estimated at 5 million tons of contained nickel,
is in the Jinchuan deposit of Gansu. Jinchuan, the second largest nickel sulfide
deposit in the world (after Sudbury of Canada), produced 15000 tons of nickel in
1981. Nickel sulfide deposits are also found in Sichuan, Yunnan, Xinjiang,
Hebei, Hunnan, Jiangsu, and Shandong. There are no estimates of the laterite
deposits (Chin 1980, 1983a, b).

Tectonic Setting

China is part of the Eurasian plate and is separated from the Siberia plate by the
Junggar-Solon Suture, from the India plate by the Yarlungzangbu-Indus Suture,
and from the Philippine plate by the Longitudinal Valley of Taiwan (Fig. 2). The
nucleus of China comprises the Sino-Korean, Tarim, and Yangzi cratons.
Mongolia and northern China were consolidated with Siberia during the Precam-
brian and Early Paleozoic Caledonian orogenies. The collision of the Silurian
and the Sino-Korea cratons formed the Solon-Hegen mountain suture (No.3 of
Geologic Setting of Selected Chromium and Nickel Deposits of China 181

Fig. 2; Li 1980). The basement of the Yanzi craton (II of Fig. 2) consolidated
about 700 Ma. The Yanzi craton coalesced with the South China fold belt during
Silurian time and continued moving northwestward. During Permian time these
two cratons were located at 2°N (McElhinny et al. 1981). During Late Triassic
time, the Yangzi-South China cratons collided with the Qinling fold belt, and the
maritime fold belt of southwestern China consolidated with the Yanzi-South
China cratons. The Yarlungzangbu-Indus Cenozoic Suture (No. 13, of Fig. 2) is
the result of the subduction of the Tethys Ocean's crust under the Eurasian
continent and subsequent collision with the Eurasian and Indian continents (Fan
1984).

Chromium Deposits

There are two types of chromium deposits in China: podiform, exemplified by

the Luobusa deposits of Tibet, northwestern and northern Inner Mongolia, and
western Henan deposits; and stratiform, represented by the Panzihua deposits of
Sichuan.

Luobusa. The Luobusa deposits are located in the eastern segment of the
southern Xizang ophiolite belt, south of the Yarlungzangbu rift. The belt, which
trends east-west and plunges southward, is 43 km in length and 3.7 km in width.
The clastic sediments are sandstone and conglomerate with ultramafic boulders.
The belt has a fault contact to the north with Miocene nonmarine clastic
sediments, and a southern contact with upper Triassic flysch deposits consisting
of metamorphosed sandstone, slate, and phyllite. The ophiolite belt was
emplaced during Late Yenshanian to Early Himalayan time (Yen et al. 1979)
during the collision of Eurasian and Indian continents.
The ophiolite belt comprises three layers (from north to south or from
bottom to top). (1) dunite (bottom) - 150 - 800 m in width and containing
veinlets of chromite; (2) dunite and harzburgite - 200 - 1400 m in width; dunite
occurs near the bottom and chromite occurs near the middle of the lower section
with harzburgite; (3) harzburgite and dunite (top) - small chromite bodies only
(Fig. 3). In the ophiolite belts there are 181 chromite deposits and numerous
mineralized regions. Thirteen of the ore bodies are greater than 100 m in length.
Most of them are located in the second belt (dunite-harzburgite) and most of the
veins are clustered together. Chromite and Cr-spinel are the major ore minerals.
Other minerals present are magnetite, millerite, uvarovite, Cr-chlorite, Cr-mica,
bornite, olivine, serpentine, and pyroxene. Os, Ru, Rh, and Ir are also reported
to be associated with the chromite ore bodies (Yen et al. 1979).
The Luobusa deposits are podiform. The magma while intruded into the
deep-seated magma chamber is considered still molten, in a state of crystal mush.
The Luobusa deposits were formed in an unstable tectonic environment and
underwent structural deformation during the process of crystallization. Chi and
Shi (1981) recognized three genetic types ofd chromite in the Luobusa deposits.
(1) Deposits formed by segregation of crystals during a late magmatic stage. They
182 P .-F. Fan
Fig. 2. The plate tectonics of China (Li 1980) o
(I)
o
0'
0<>
m Precambrian Massif ~ Early Mesozoic Fold Belt §1 Melange n·
[f1

m Late Precambrian Massif m Late Mesozoic Fold Belt B Glaucophane Schist i:!
Jg.
B8;3 Median Massif D Cenozoic Fold Belt IE] Subduction Zone o....,
[f1
(I)
§ Early Paleozoic Fold Belt (;2] Ophiolite Belt ~ Deep-Seated Fracture (")

IIlJ] Late Paleozoic Fold Belt

"8-
~ Granite of the Same Tectonic Stage
()
::r
..,
o
3
2·
3
Subduction Zone and Suture Massif and Median Massif ~
::l
p..
1 Altay Early Paleozoic Subduction Zone I. Tarim-Sino-Korean Massif Z
n·
2 Junggar Late Paleozoic Subduction Zone II. Yangzi Massif i'<"
~
3 Solon-Hegen Late Paleozoic Subduction Zone III. Northern margin of Indian Massif o
(I)
4 Nadanhada Early Mesozoic Subduction Zone IV. Jiamusi Median Massif '0
5 Tianshan Late Paleozoic Subduction Zone V. Qaidam Median Massif ~.
VI. Qiangtang Median Massif en
6 Yinshan-Tumen Late Paleozoic Subduction Zone o...,
7 Qilian-Qinling Early Paleozoic Subduction Zone ()
8 Qinghai-Qinling Early Paleozoic Subduction Zone e:
9 Altun Late Paleozoic Subduction Zone (or Transform Fault) Depth Fracture Sl
10 Kunlun Late Paleozoic Subduction Zone
11 Hoh Xil-Jinsha-Ailao Early Mesozoic Subduction Zone (1) Tan-Lu Depth Fracture
12 North Xizang-West Yunnan Late Mesozoic Subduction Zone (2) Deierbugan Depth Fracture
13 Yarlungzangbu-Indus Cenozoic Suture (3) Lishui-Haifeng Depth Fracture
14 Taiwan Cenozoic Suture (or Transform Fault) (4) Garze-Litang Depth Fracture

00
w
-
184 P.-F. Fan

.··...........
.. .. .. ..... .. . .., .....
............. .. ... .. . ............
........
........ ...
·............
. .
..................... .
...--_ . .....
c-p
----_
- -......-.....,.....,--
-.....---

1 Fig. 3. General geologic map of the Luobu a de-

po it, Xizang (Tibet) (After Cui and Qiao t 983).
1 em equals approx. 350 m
82' 86' 96'[

-
D Dunite [ [ ] Quaternary
Peridotite D Lower Tertiary
Lherzolite-Gabbro D Upper Triassic

[2J Fault Ic- pi Carboniferous-Permian

~ Chromile deposit D Himalayan Granite

El Boundary of ore body 1:::_:::1 Yen han ian Gnei sic Granite

could be segregated in situ or by pressure filtering. (2) Primordial cumulative

deposits of a late magmatic stage. The ore-forming Cr-spinel has pisolitic,
nodular, and ellipsoidal shapes. The pisolites, oriented in graded beds, were
formed by gravitational settling and by current differentiation. (3) Deposits
formed by injection during a late magmatic stage. The ore bodies form a sharp
boundary with the country rocks. The three genetic types occur both separately
and in combination. Most of the Luobusa deposits show a zigzag form. Seven
types of zigzag forms were recognized by Hu et al. (1981); they were shear
fissures developed during the time of ore formation.
Podiform chromite deposits are now interpreted as remnants of on-land
oceanic lithosphere. They occur as lenses, pods, slabs, and irregular shapes. The
chromite found in ophiolitic harzburgites is suggested to have originated beneath
Geologic Setting of Selected Chromium and Nickel Deposits of China 185

an oceanic spreading ridge. The dunite wall rocks probably resulted from the
partial melting of harzburgite during ore body emplacement. The basic texture
patterns, seen to be a disruption of the primordial cumulative chromite structure,
resulted from deep-seated mantle subsolidus deformation (Thayer 1969). Late
Mesozoic and Cenozoic examples include deposits in Greece (Moores 1969),
Cyprus (Greenbaum 1977), Turkey (Thayer 1964), Oman (Peters and Kramers
1974), and New Caledonia (Nicolas et al. 1981). Nicolas et al. (1981), in their
study of southern New Caledonia nodular chromite, indicated that nodular
chromite deposits resulted in the concentration of chromite by a process of
elutriation in parts of the magmatic conduits. In the conduits, chromite grains
have an opportunity to agglomerate into pisolitic, pelletlike, ellipsoidal, or
massive shapes and eventually settle.
Northwestern Inner Mongolia. The northwestern Inner Mongolia chromite
deposits, located along the east-west trending rift zone, constitute the eastern
segment of the Hercynian Tienshan fold belt. The country rocks consist of horn-
blende-chlorite schist and mica-quartz schist and are metamorphosed
Ordovician-carboniferous marine volcanics. The ultramafic rocks were emplaced
during the middle Hercynian. The gabbro intrusion occurred before the ultra-
mafic intrusion and the Permian biotite granite intrusion followed the ultramafic
intrusion (Fig. 4). The ultramafic body consists of harzburgite and dunite, with

96' 100' 10d' 10S'E

N 40'

i
N

36'

t:.....l_--L_-L_--'--.::J 32·

0 soo l000~",

[§J III
~

Quaternary Mica-Quartz Schi t

0 Dunite 0.. Granite

[.::-.j Gabbro D Lithological Boundary

C] Serpentine
0 Cr - Fe Mineral Depo it

~ -- Hornblende·Ch lorite Schist Peridotite-Dunite

Fig. 4. General geologic map of northwestern Inner Mongolia (After Wei 1978)
186 P.-F. Fan

chromite occurring in the dunite schlieren, usually in the middle segment of the
ultramafic body or in the upper-middle part of the harzburgite zone. The ores
occur as magmatic segregations in the form of stringers, lenses, veins, pods, and
pillars. Three genetic types of chromite deposits are recognized by Wei (1978):
those formed by (1) in situ magmatic segregation, (2) late magmatic injection,
and (3) multiple phases of late magmatic segregation.

Northern Inner Mongolia. A chromium deposit is reported by Cao (1976) in the

western part of Inner Mongolia (Fig. 1). The chromite-bearing ultramafic body is
located in the Inner Mongolia continental margin and western segment of the
Hercynian fold belt. The country rocks consist of over 10000 m Carboniferous
flysch deposits metamorphosed to arkosic sandstone, slate, phyllite, sericite-
quartz schist, and garnet-mica schist. An east-west trending rift south of the
deposit is related to the formation ofthe ultramafic body. The ultramafic body is
32 km in length and 6 km in width and consists of mostly harzburgite with veins
of dunite. Ores, which occur as veins, pods, lenses, and stringers, form sharp
contacts with the host rocks. The enstatite in harzburgite is characterized by
secondary banded structures, banded crystal faces, and undulating extinction
under microscope. Cao (1976) suggested that this ultramafic complex was
emplaced as a solid intrusion during tectonism.

Western Henan. The chromium deposits in Henan (Fig. 1), described by Wan
(1977) and Zhang (1977), are located at the southern flank of an anticline of the
eastern segment of the east-west trending Qinling range. A rift cuts through the
region and separates two metamorphic facies. Precambrian migmatite occurs
north of the rift and Precambrian marble and quartzite to the south. The
chrome-bearing ultramafic rocks intrude into the migmatite. Chromite and Cr-
spinel are the major ore minerals, although 45 other minerals were identified and
their genesis and paragenesis discussed. Forty-seven chromium ore bodies have
been discovered in this area. The preexisting chromite deposit, which is magmatic
in origin, was altered during migmatization.

Panzihua. The Panzihua deposits are vanadium-bearing titanomagnetite deposits

that are also rich in chromium. Panzihua is located in southwestern Sichuan
(Fig.1) at the western edge of Yangzi craton. The country rocks are Sinian
limestone. A layered ultramafic-mafic intrusion was emplaced during Middle-
Late Paleozoic time (Yen et al. 1979). The intrusive body is 35 km in length and
2 km in width, trending northeast 45° and dipping northwest at 50° - 60°
(Fig.5). Five rhythmic layers are present (from top to bottom). (1) Layered
gabbro - the dark-colored gabbro has a thickness from 500 -1500 m with
stockworks. It is poor in ore content. (2) Upper light-colored layered gabbro -
iron-bearing gabbro with a thickness from 10 -120 m, with stockworks consist-
ing of 5"10 apatite and two ore-bearing zones. (3) Middle dark-colored layered
gabbro - iron-bearing gabbro layer with a thickness of 160 - 600 m and stringers
or ores near the base. (4) Lower dark-colored medium to coarse-grained ore-
bearing gabbro layers. There is a 60 - 570 m thick gabbro layer with a few meters
Geologic Setting of Selected Chromium and Nickel Deposits of China 187

W Quaternary

CJ Syenite

Upper T riassic Seric

ts::n Sinian Limestone

: ..
~ Layered Gabbro

[ill Layered Fine-grained Gabbro

~ Layered Iron-bearing gabbro

Alkali Granite

Ore Body

~ River (An ing River)

Fig. 5. Genera l geologic map of the

Panzihua deposit , Sic huan (A fter
Yen et al. 1979)

of dunite and peridotite, and six ore-bearing zones. (5) Bottom boundary layer -
dark-colored, fine-grained gabbro 10- 300 m thick and poor in ore.
The ore minerals of the Panzihua deposit consist of titanomagnetite,
ilmenite, and Mg - Al spinel. Other minerals are magnetite, perovskite, anatase,
pyrrhotite, nickel-pyrite, and linnaeite (Yen et al. 1979). Vertically, chrome con-
centrates in olivine pyroxenite-dunite layers near the bottom of the rhythmic
layers. Locally, the ores contain from 1.0 to 3.3% Cr203' which is mainly con-
centrated in titanomagnetite (Lu et al. 1980, Luo 1981).

Nickel Deposits

The nickel deposits of China occur in four settings: (1) in ultramafic intrusions in
rift valleys - linchuan deposits of Gansu and Dali deposit of Yunnan; (2) as ob-
ducted ophiolite - Ailaoshan deposit of Yunnan; (3) in deposits of weathering
products - area surrounding the Ailaoshan ultramafics and southern China;
and (4) in sedimentary deposits in Yangzi craton.
188 P.-F. Fan

Jinchuan

Jinchuan, also known as the Baijiajuzi deposit, is by far the largest nickel deposit
in China. The deposit was discovered in 1958; operations began in the open pit in
1964 and in the underground mine in 1969. Reserves are estimated to be 5 million
tons of nickel, 3.5 million tons of copper, and 100000 tons of cobalt. Pt, Os, Rh,
Ir, Te, and Se were also reported to occur in the nickel deposit (Chin 1983a).

@] Quaternary Fig. 6. General geologic ma p and geologic sec-

tion of the linchuan depos it, Gansu (Tang
[AnZ , Pre inian 1982). t em equals approx. 250 m

~ Marble

Schist and Migmatite

Q Q
Granite

Lherzolite

[2J Fault

D Oi seminated Ore Body

~
/..':'
Spongy Ore Body

[2] Blocky Ore Body

0 Contact Ore Body

40'

L------'-----'-.
98' lOr l06'E
Geologic Setting of Selected Chromium and Nickel Deposits of China 189

The linchuan deposits are located at the southeastern uplifted margin of the
Sinokorean craton (Fig. 2), and its ore body is in the eastern part of the northern
Longshan Mountain ultramafic rocks. The ultramafic rocks occur between the
Presinian migmatite and marble of the Baijiajuzi Formation of Changcheng
Group, which is 1.5 billion years old (Shi 1980, Tang 1982). The ultramafic body
is bounded on both the north and south sides by deep faults trending from north-
west 45° -70° to near-northeast (Shi 1980). The ultramafic body is 6500 m in
length, 20 - 500 m in width, strikes in a northwest direction, and dips at 70°. It is
dissected by a fault into four sections. The two central sections (I and II of Fig. 6)
are exposed, whereas the eastern (IV) and western (III) sections are covered by
Quaternary deposits. The ultramafic body consists of lherzolite (64070 olivine and
31 % bronzite and diopside) and sulfide-bearing dunite. Dunite is concentrated at
the middle and upper part of the ultramafic body, and sulfide-bearing dunite is
concentrated near the bottom. Diopside-chlorite schists and tremolite-chlorite
schists, sometimes up to 1 m thick, occur along the contact. The ore minerals
consist of nickel-pyrite, violarite, and chalcopyrite. Other minerals present are
Cr-spinel, Fe-spinel with ilmenite inclusions, carbonite, vallerite, marcasite,
pyrite, and magnetic pyrite. Elements such as Pt, Rh, Ir, Se, Te, Ru, and Os are
also present (Tang 1982).
Four types of ore body are present. (1) The lower magmatic segregated ore
body comprises more than 85% of the total reserve. The elliptical ore bodies
occur near the ultramafic rocks - sulfide-bearing dunite layers. Ni - Cu sulfides,
which occur between olivine crystals, have a spongy, meteoritic iron structure.
(2) The upper magmatic segregated ore body is seen in layers, veinlets, spots, and
spongy form. The ore content of this layer is less than that of the first type.
(3) Intruded ore bodies. Veinlets occur in the lower part of the ultramafic body.
(4) Contact ore bodies. They are usually a few meters to several hundred meters
in length. Thickness ranges from a few meters to tens of meters (Tang 1982). The
tectonic setting of the linchuan deposits, an ultramafic intrusion in a rift valley,
is similar to that of the Skaergaard Complex, Greenland, with its relationship to
the North Atlantic rifting (Naldrett 1981, Chase and Gilmer 1973).

Dali

The Dali deposits are platinum-group mineral deposits with Ni - Cu sulfide as a

by-product. Dali is located at the southwestern edge of the Yangzi craton
(Fig. 1). The country rock consists of biotite-plagioclase gneiss of the Archean
Yuanmou Group, with an age of 4.06 billion years (Tang 1982). The funnel-
shaped intrusive body is 600 m long and 430 m wide and consists of gabbro,
dunite, lherzolite, and pyroxenite. The ore-bearing dunite is usually located near
the base along the contact with the country rocks (Fig. 7) (Yen et al. 1979). Non
information is available on relative amounts of nickel and copper.
190 P .-F. Fan

,,,
i

[§J
,
Quaternary

I
R Gneiss

I fl..
-.....
~ ~ Gabbro
I I
r / 111.1,1,1, 1.
:.:.1,1,1.'1 Pyroxenite
'/
1/
I
A ~
..~"... Peridotite

D Dunite

[2] Fault

,, ~ Ore Body
A
Q

! r

Fig. 7. Geologic map and geologic

section of the Dali deposit, Yunnan
(After Yen et at. 1979). 2.5 em
100'E 104' equals approx. 150 m

Ailaoshan

The Ailaoshan deposits are located in Yunnan province (Fig. 1). The ultramafic
body is 300 - 600 m in length, 50 - 100m in width, and consists of gabbro, py-
roxenite, and lherzolite. Ni - eu sulfides and some platinum-group minerals oc-
cur in the lherzolite (Tang 1982). The nickel-bearing ultramafic body is an ob-
ducted ophiolite that resulted from uplifting of tectonic emplacements at the
southern portion of the North Tibet-West Yunnan Late Mesozoic subduction
zone.
Geologic Setting of Selected Chromium and Nickel Deposits of China 191

Weathering Products

Deposits of weathering products - nickel silicate clay minerals formed south-

west of the Ailaoshan ultramafics - were described by Tang (1982). The follow-
ing eight weathered surfaces average 80-100 m: (1) red residual clay, serpentine
detritus present locally; (2) iron crust, loose, joints are recognizable; (3) iron
crust, serpentine structure recognizable; (4) greenish kaolinized serpentine struc-
ture recognizable; (5) greenish kaolinized serpentine structure recognizable,
loose, granular; (6) fracture serpentine belt, kaolization present locally; (7) sili-
ceous serpentine belt, chalcedony and opal present; (8) serpentine belt, carbona-
tion present, secondary calcite, and dolomite filling fissures of serpentine. Eighty
percent of the weathered crust are ore bodies. Of these eight layers, numbers 2, 3,
and 4 are rich in iron ore; numbers 5, 6, and 7 are rich in magnesium ore; and
numbers 4 and 5 are rich in nickel. The nickel content decreases above and below
layers 4 and 5. This deposit is similar to the nickel laterites of Soroako, Indonesia
(Golightly 1979) and New Caledonia (Troly et al. 1979). They consist of an iron
oxide (goethite and hematite) zone overlying saprolite (altered to unaltered
serpentine) zone.

Sedimentary Deposits

Sedimentary deposits in the Yangzi craton, such as Lower Cambrian black

shales, are good sources for nickel; they are usually associated with pyrite,
phosphate, and coal. The ore-breading shale is usually 1 m thick, thin-bedded,
and of lens to lenticular shape with a few millimeters of nickel sulfides -
millerite, violarite, and nickel-pyrite. The Ni content ranges from 60-215 ppm.
The sedimentary deposits are also rich in Mo, Y, Cu, Pt, Pa, Au, and Ag (Chen
et al. 1982). As this is a preliminary report, no data for deposit grade and
tonnage are mentioned.

Concluding Remarks

Podiform chromite deposits occurring in Luobusa, northwestern and northern

Inner Mongolia, and western Henan are associated with harzburgite and dunite.
They are the result of magmatic differentiation and gravity settling and are also
associated with a subduction zone. The stratiform chromite in the Panzihua
deposit is related to rifting.
The nickel deposits are usually associated with ultramafic bodies, in par-
ticular with lherzolite and dunite. The linchuan and Dali deposits are related to
continental rifting, and the Ailaoshan deposits to a subduction zone. Nickel ores
are also found in weathering products around Ailaoshan and in Lower Cambrian
black shales in the Yangzi craton.
192 P.-F. Fan

Acknowledgments. This project is partly supported by the Resource Systems Institute of the East-
West Center, Hawaii. I am grateful to Charles Johnson and Allen Clark of the Resource Systems
Institute for their support. Sincere thanks to Rita Pujalet of Hawaii Institute of Geophysics for
editorial help. Hawaii Institute of Geophysics Contribution No. 1733.

References

Brady ES (1981) China's strategic minerals and metals. Chin Business Rev 8(5):55 -73
Cao R (1976) On the genesis of the alpine-type ultramafic complex and related chromite deposits in
the western part of the Inner Mongolia orogenic belt. Geochimica 2:113 -125 (in Chinese)
Chase GC, Gilmer TH (1973) Precambrian plate tectonics: The midcontinent gravity high. Earth
Planet Sci Lett 21 :70 - 78
Chen NS, Yang XZ, Liu DH, Xiao SJ, Fan DL, Wang LF (1982) Lower Cambrian black argillaceous
and arenaceous rock series in South China and its associated stratiform deposits. Miner Depos
) :49 - 55 (in Chinese)
Chi S, Shi Q (1981) Genesis of X chromite deposits, Xizang autonomous region. Sci Geol Sin
343 - 359 (in Chinese)
Chin E (1980) The mineral industry of China. In: Mineral yearbook, vol III. US Bur Min, pp
237 -259
Chin E (1983a) China. In: Mineral yearbook, vol III. US Bur Min, pp 225-250
Chin E (1983b) Far East-China. Min Annu Rev 349-357
Cui JW, Qiao ZJ (1983) A study of the tectonic environment of Lobusha chromite deposit. Min
Depos 2(1):48 - 57 (in Chinese)
Fan PF (1984) Geological setting of selected copper deposits of China. Econ Geol 79:9-19
Golightly JP (1979) Geology of Soroako nickeliferous laterite deposits. Int Later Symp, New
Orleans. Soc Min Eng AIME, pp 38 - 56
Greenbaum D (1977) The chromitiferous rocks of the Troodos ophiolite complex, Cyprus. Econ Geol
72:1175-1194
Hu J, Zhang S, Xiang D (1981) The zigzag chromite ore bodies in Luobusa, Xizang (Tibet) and their
stress field features. Earth Sci 2:73 - 84 (in Chinese)
Li C (1980) A preliminary study of plate tectonics of China. Bull Chin Acad Geol Sci Ser 1-2:11 - 22
(in Chinese)
Lu J, Zhang C, Gu G, Gao X, Fang Y, Zhou J (1980) The distribution and state of chromite in the
Hongge layered vanadium-bearing titanomagnetite deposit in Sichuan province. Chin Acad Geol
Sci Bull Ser 111:1- 20 (in Chinese)
Luo Y (1981) The characteristics of Ti-chromite mineralization in Xin Jie layered ultramafic-mafic
intrusion in Panzihua area, China. Geochimica 66 - 73 (in Chinese)
McElhinny NM, Embleton BJJ, Ma XH, Zhong ZK (1981) Fragmentation of Asia in the Permian.
Nature (London) 293:212 - 216
Mikami HM (1983) Chromite. In: Industrial Minerals and rocks (Nonmetallics other than fuels),
vol I. Soc Min Eng AIME, New York, pp 567 - 584
Moores EM (1969) Petrology and structure of the Vourinos ophiolite complex, northern Greece. Spec
Pap Geol Soc Am 118:1-74
Naldrett AJ (1981) Nickel sulfide deposits: Classification, composition, and genesis. Econ Geol 75th
Anniversary Vol 628 - 285
Nicolas DC, Rabinovitch M, Moutte J, Leblanc M, Prinzhofer A (1981) Structural classification of
chromite pods in southern New Caledonia. Econ Geol 76:805 - 831
Peters TJ, Kramers JD (1974) Chromite deposits in the ophiolite complex of northern Oman. Miner
Depos 9:253 - 259
Shi Z (1980) Mineralogy and petrology of Jinchuan ultrabasic rock body with copper-nickel deposits.
Chin Acad Geol Sci Bull X 1-1:78-91 (in Chinese)
Tang Z (1982) A discussion on the main types of nickel deposits in China with an approach to their
relations with the paleoplate tectonics. Miner Depos 1-2:29 - 38 (in Chinese)
Geologic Setting of Selected Chromium and Nickel Deposits of China 193

Thayer TP (1964) Principal features and origin of podiform chromite deposits and some observations
on the Guleman-Soridag District, Turkey. Econ Geol 59:1497 -1524
Thayer TP (1969) Gravity differentiation and magmatic re-emplacement of podiform chromite
deposits. In: Wilson HDB (ed) Magmatic ore deposits. Econ Geol Mon 4:132-146
Troly G, Esterle M, Pelletier B, Reibell W (1979) Nickel deposits in New Caledonia, some factors
influencing their formation. Int Later Symp, New Orleans. Soc Min Eng AI ME 85 -119
Wan X (1977) A geochemical study of chromite spinel alteration in a chromite deposit, Henan
province. Geochimica 4:250 - 261 (in Chinese)
Wei W (1978) On the genesis and genetic types of a chromite deposit in Kansu. Acta Geol Sin
269 - 280 (in Chinese)
Yen JC, Chu SQ, Huo YS (eds) (1979) Ore deposits. Geological Press, Beijing, China, 443 pp (in
Chinese)
Zhang B (1977) A migmatization-altered chromite deposit. Geochimica 147 -155
Part II
Porphyry Deposits
Recent Advances in Porphyry Base Metal Deposit
Research
J. M. GUILBERT!

Abstract

As an introduction to this portion of the Symposium on Copper Deposits, this

paper reviews recent progress and findings in research on deposits associated
with hypabyssal epizonal plutons. The I-type affinity porphyry copper-molyb-
denum series is first established as petrogenically distinct from deposits asso-
ciated with S-type, generally 2-mica granites, the porphyry tin - tin granite-
uranium granite suite. Then five specific areas of advance in the 1970s and 1980s
are described. The first deals with plate tectonic settings and the global geologic-
lithologic distribution of porphyry base metal deposits (PBMDs). Second, the
significance of the recognition of changes in the ratio of magmatic to meteoric
fluids during alteration-mineralization and deposit emplacement is discussed,
along with the concept of "phyllic overprint" and its applications. Techniques
for observing, measuring, and describing the propagation of fluids and fractur-
ing in time and space are given. Third, experimental studies, thermodynamic
modeling, stable and radiogenic isotope analyses, and fluid inclusion chemical
analysis and microthermometric studies and results that have advanced our
understanding of the geochemical and thermal nature of PBMD fluids are cited.
Fourth, geomathematical modeling studies that describe probable fluid source
areas, flow paths, kinetics, and the volume hydrothermally influenced by the em-
placement of PBMD systems are considered, and finally the present status of our
understanding of PBMD fluids in terms of T, P, 1(02), I(S2), salinity, and ion
complexing are briefly stated. It is pointed out that the promise of the next
decade is enormous if field and laboratory studies continue to loop back to
support and extend the whole, and if research continues at its present pace.

Porphyry base metal deposits have been the object of global geologic research
since 1970, especially in the Cordilleran American countries that are effectively
the "type area". The results of that research have been so explicit about pressure-
temperature-composition-tectonic aspects of porphyry base metal deposit
environments of formation that we have collectively made tremendous advances
in understanding both specific ore deposit genesis problems and the more general

I Professor of Economic Geology, University of Arizona, Tucson, AZ 85721, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Recent Advances in Porphyry Base Metal Deposit Research 197

conditions of granite formation and metallogenesis. "S-type" 2 granites are so

little known in North and South America that the word "granite" has come to be
taken by most American geologists as essentially synonymous with "I-type"
systems. We are only just now - in large part through the efforts and scientific
disclosures of scholars from Australia, Europe, Asia, and the USSR - coming
to realize how important the S-type occurrences are economically, geochemical-
ly, and metallogenically and how varied both 1- and S-types can be. This paper
reports the results of research that is heavily directed toward porphyry copper,
porphyry copper-molybdenum, and porphyry molybdenum I-type metallogeny;
it must be extrapolated to porphyry tin and disseminated tin-tungsten-uranium S-
type-affinity deposits in the Americas and elsewhere, and to A-type affinity
Climax-type molybdenum deposits, only with great caution. Many of the
methods of research may serve all these types of disseminated base metal
deposits.
This paper will review recent progress in understanding (1) plate tectonic
settings and PBMD distribution; (2) PBMD mineralization-alteration fluid
histories; (3) PBMD fluid compositions; (4) fluid sources and flow patterns; and
(5) fluid chemistry and physics. It will also be shown how each contributes to
improved geologic models, and finally how field, petrographic, and geochemical-
geophysical studies loop back to improve overall understanding of these systems.

Plate Tectonics and PBMDs

The impact of plate tectonics in the 1970s was felt as much in ore deposit geology
as anywhere else, if not more. Tectonic settings of major ore deposit types have
been defined (Guilbert 1981, Mitchell and Garson 1981, Sawkins 1984), and the
roles of magma types and ore deposition styles in many lithotectonic terranes
have been established. Porphyry copper and porphyry "moly" deposits -
"PCDs" and "PMDs" - are associated with compressive, consumptive margins
(Sillitoe 1972, 1981, Sawkins 1972), commonly with variations in major and
minor elements that appear to be part of an overall metallogenic zoning pattern
first established in plate tectonic terms by Mitchell and Garson (1972). Keith
(1978) and Westra and Keith (1981) have related the geographic distribution of
variations in Cu - Mo - Zn - Pb - Ag - Au in North American porphyry depos-
its to changes in the angle of plate subduction and to distance from the trench to
the arc-orogen hearth of magmatic activity. Recent discoveries of porphyry cop-
per-gold deposits in Argentina east of the porphyry copper-moly deposits of
Chile tend to corroborate those findings. As R. L. Nielsen (personal communica-
tion, 1984) notes, exploration activities and discoveries in the late 1970s reveal
that porphyry copper deposits are much more diverse than was previously

2 For brevity, "S-type" will be used except where noted to stand for transformation-type, leuco-
granite, and ilmenite-series, and "I-type" will be used for syntectic-type, granodiorite, and
magnetite-series granitoids as used by Chinese, European, and Japanese colleagues respectively.
although the terms are not fully equivalent.
198 1.M. Guilbert

thought, especially in terms of associated and genetically related host rocks. We

find deposits associated with calcic-diorite clan rocks (Panguna, El Arco); calc-
alkalic grandiorite clan rocks (mostly in western USA); alkali shoshonitic clan
rocks (Bingham Canyon and OkTedi); and even nepheline-normative alkalic
syenite clan rocks (Stikine, B.C.). Some suggestions exist that each of these
broad groups may be characteristic of a different tectonic setting. The calcic-
diorite rocks seem to be characteristic of island-arc magma systems developed on
oceanic crust. Calc-alkalic and alkali-calcic granodiorite to quartz monzonite
systems are developed in conventional magmatic arcs over subduction zones in
continental margin settings. The syenite host rocks may be related to continental
rifts. Some major advances in understanding can be expected as we get more
hard data on deposits in the southwest Pacific and southeast Asian region. A
topic that has potentially significant economic implications is the variation of
gold content and Mo - Cu - Au ratios in porphyry copper systems in various host
rocks and tectonic settings. This topic also has some genetic implications regard-
ing the sources of metals, in general, and the sources and evolution of porphyry
systems, in particular.
Heidrick and Titley (1982) related the tectonic fabric of southwestern North
America to plate tectonic elements and the PCDs within them. Sillitoe et al.
(1982) have recently shown that the PCDs of Colombia occur in three subparallel
subbelts of different ages (Jurassic, Miocene, and Eocene), each with distinguish-
able Cu-Mo-Au ratios. Westra and Keith (1981) presented a chemical clas-
sification of "Climax-type moly" deposits. They show, as do Mutschler et al.
(1981), that those of the Climax-Urad-Henderson-Mount Emmons type are
almost certainly atectonic or rift related, the extensional to neutral tectonic result
of the annihilation of compressive forces when the North American plate overran
the East Pacific Rise in mid-Cenozoic times. Their absence from the South
American and Canadian scene is thus understandable. Porphyry tin deposits
(Sillitoe et al. 1975) are probably in S-type hypabyssal rocks analogous to the tin
granites; less is known about them and and their distribution in the Americas.
Mutschler et al. (1981) describe "granite molybdenite systems" - the Climax
type - as occurring in rocks that are petrographic-geochemical S-types; they
may actually be closer to A-type granitoids. Further progress in understanding
granite origin and metallogeny can be anticipated from Nd - Sm and Rb - Sr
studies (Farmer and DePaolo 1983) and Lu - Hf systematics (Patchett 1983).
Powerful insights to the lithotectonics of granitoids provided by radioisotopic
dating methods, especially potassium argon techniques (Damon et al. 1981), are
crucial to progress.
Studies are continuing on a broad scale, and we can expect important results
in early years from Keith and others on application of magma chemistry and ore
deposits to plate tectonic settings and metallogeny, from Sillitoe on continued
refinement of PBMD relations to detailed reconstructions of tectonic elements,
from Ren and Guilbert on Fe - Ti - 0 species as determinants in southwestern
North American PBMD systems, from several workers in North America and
especially Europe on tectonic settings analysis through compilation of geo-
chemical-geochronological-geophysical-tectonic aspects of major districts, and
from continued major and minor element whole rock and petrographic-miner-
Recent Advances in Porphyry Base Metal Deposit Research 199

alogic analysis of granitoids, as for example at the CNRS-CRPG near Nancy,

France. Results from many countries and continents need to be compiled and
refined, the potential contribution of the USSR, with several ages of PBMDs,
and of China, with four recognized PCD metallotects, representing outstanding
examples.

Mineralization-Alteration Fluid History

One of the dramatic discoveries of the 1970s was that although the bulbous to
shield-shaped models of alteration-mineralization zoning in PBMDs proposed by
Lowell and Guilbert (1970), Hollister et al. (1975), and others are, in general,
descriptively correct, they represent a summation of many continuous, time-
staged processes rather than a single event. Nielsen (1968) described and Shep-
pard et al. (1971) documented the phenomenon at Santa Rita that Gustafson and
Hunt (1975) were able fully to describe as "phyllic overprint" in the sense of
early, prograde magmatic-water mineralization-alteration followed by later,
lower temperature alteration as part of a massive retrograde influx of meteoric
water at El Salvador, Chile. Sutherland Brown (1976) has further pointed out
that PCDs and PMDs can be classified according to pressure, temperature, and
time variables. We now recognize that however useful the Lowell and Guilbert
model has been in exploration, it is oversimplified in its description of processes.
It appears that early alteration is largely alkalic, or alkali metasomatic, and
weakly hydrolytic, with the early establishment of a central zone of potassic
alteration yielding to an outer zone of propylitic effects, with disseminated and
microveinlet copper-moly mineralization peaking in outer potassic volumes.
Later phyllic and argillic alteration - with substantial redistribution of sulfides
- is then typically "overprinted" onto the mineralized zone, the result of an
influx of meteoric waters convecting through the higher permeabilities of that
mineralized zone. In many porphyry deposits, veinlets flanked by phyllic altera-
tion do indeed cut potassic assemblages on their outer side and propylitic ones on
their inner side; but it is too simplistic to state categorically that all sericite, or
phyllic assemblage material, is late in origin. It is noteworthy that we are aware
of several deposits in which the age separation is especially clear - Silver Bell
and Morenci, Arizonfl, and Butte, Montana, for example - and several at which
the phyllic overprint was weakly developed at best, such as at Ajo and Sierrita,
Arizona; El Arco, Baja California del Norte, Mexico; and Bingham Canyon,
Utah. Further, as later paragraphs will show, solutions clearly evolved in
temperature-pressure-salinity-isotope "space" such that complex fluid histories
are more the rule than the exception.
It was assumed as we entered the 1970s that most fracturing was tectonic in
nature and, except for breccia pipes and obvious internal effects, that fracturing
was an internal strain response to external stress (Mayo 1958, Rehrig and
Heidrick 1972, 1976). Just as alteration-mineralization space-time histories have
proven resolvable, but more complicated than we thought, so also is the propaga-
tion of fractures. A few workers - among them Nielsen and Burnham - sug-
 200 J. M. Guilbert

gested in the early 1970s that the stockwork fracture systems in some porphyry
deposits were intrinsic and related to stock cooling by a process called "hydro-
fracting". Several authors, most notably Knapp and Knight (1977) and Norton
(1982), have since shown that fluids driven outward from cooling stocks can
propagate the fractures along which they then travel by a combination of pro-
cesses including differential thermal expansion of pore fluids, changes in pore
fluid pressure, hydrolysis, and vapor-pressure "wedging". Haynes and Titley
(1980) measured these veinlet sets by documenting the progression of fractures
and fluids in space and time around the porphyry system at Sierrita, Arizona.
They drew squares 1 m on a side on representative outcrops at increasing dis-
tances from the center of the porphyry system, and measured the aggregate
length and crosscutting relationships of the several types of alteration-mineral-
ization-aperture style veinlet sets that occur in each square meter. By studying
P - T conditions indicated by mineralogy and by fluid inclusion filling tempera-
tures and salinities in each of those veinlet types, they were able to describe the
evolution of fracturing, of ore and gangue mineralogy, and of geochemical
patterns in time and space (Fig. 1). A pattern (Fig. 2) of first advancing, then
retreating propylitization, of outward-generated fracturing, and of decreasing
temperature and salinity through time is indicated, with the late development of a
scanty phyllic overprint event. The first detailed study of structural-mineraliza-
tional interdependence and evolution will certainly be followed by other studies
elsewhere, since the fracture density-fluid evolution relations are so vital to
understanding the processes that are basic to "porphyry" mineralization. Preece

.20 - ....... 1

"" \
\
\ I. Total of all altered
\ veins-veinlets - -
\ 2.Quartz-Orthoclase
\ veins-veinlets ...........
J5 \ 3. Ouartz- Sulfide
-----3., \ veins-veinlels '---',
4. Orl hoclase·E pidole
\. \ veins-veinlels .... - .... "

.5\ \
250"-320"" \
'e... .10 \ \
\, \\
, \
\, \\
\ \ Fig. 1. A plot of veinlet mineralization-alteration
\ \
.05 \ \
evolution in space and time around the Sierrita
porphyry copper deposit, Pima County, Arizona.
3400 .360'", \
····· .. ······ .. 2.,..5' \ \ "n" is the aggregate length in meters per square
170"· 23·~:·~· .... ·............ ·\ ...... -->~, meter - or centimeters per 104 cm2 = cm -1) - of
/ \ ........ each veinlet type. 2 is the earliest, 4 the latest, with
emplacement temperatures indicated. Km refers to
.... _ .... _·4.- .... - .. ·., \ .""
.
"""""
" distance in kilometers from the center of the sys-
tem at the present moderately deep erosional ex-
I. 0 2.0 3.0 4.0 posure level. (From Titley 1982)
Km
Recent Advances in Porphyry Base Metal Deposit Research 201

ZON AL COMPOSIT E

'RJTASSIC'~"PHYLLIC " - - ! - - - "PROPYLlTlC" - -- - - _ .... •

Lf.Ore
r She//) --j
n(cm - I), 05 - :-- 1.5 ---··02 ~ 0.1 - - - ~ 001

w :~;:'i:';<'':':'~'::. ''!.~ Quortz aSericite-Pyrite

'.: ~':. ~'.:" ,':- ~,~ ~ . ~...., ...vv..''''''''''''''Yo.'''''''''''''''''''''''''''''Yo.'
:3 ·."
t-
.. ~.......
I ...... I / ... ' ..... no:I.0cm- l _
":;:':':,-:':' :',,'.:.<'':}': ----( Propyl.)
:,'..':.:~;:..~:~':'~~:- Quortz - Mf -Cholcopyrite
~~'''''~J~~'',;'o4o~ " "'. . . . . . . . . . . . . . ., . . . __. . . . . . . . . .,. . . . . . . ,. . ..
~, . . ~ ..... ~ ..... ~,':.. .. ; ...... ~ - 0 2 - Q.lcm- t
:f,~{,~{;~.:/~'~,".:~" ... (Propyl.)
;:,~;.:,,~J::.5'~ ~/~ " . . - Orthoclase - Quortz (py-cpy)
I~ ~ I~ ~~: ~.: ... I " ~ I " - ..~........' :r-
........,.'..........................." .." ...' ..." ' ' '...'''..............' .......

;~:~f:~~;~i:~t~~t n-O.2 ~Icm (Propyl.)

:.:,'\;.:,~/.:~/( ~/~~j '" Biotite- Quartz (mt, Py.cPy)
I ~ ~ ~I!~~~ , .. -' .... ............., ........."",.........'.,. . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .'.........................."""""""" ...,~
::: /

·~,,,,,,~".."i'.."'~"~~'>~': n=0.2 0 .1 Q,Q5cm- 1

)- :.:..::.:,-.;.:..-.;.:,-.:.:...:..:,' ...... {Propyl. )~---
.J
a::: .
I, ~ . . ,. ". . ,. . . ,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
~~..~/::,~~,':i,~;~,':.:,' Selectively Pervasive Biohte Alteration
~~~~)~,.~
~ ~~'(I~ ..~~~-:,~'..:~~~~'. . n=02 -------.0.1 0 ,05
Propylitic Alteration
~
OOlcm~ 1
:~/~~~~~~.....~~'~~(~~,: Sinuous QuortzVeins
.:- .. ~ " . .." . ".' ,
I NTRUSION WAL L ROCK

Fig. 2. Asummary diagram of alteration-mineralization progress in time (upward) and space (lateral)
at Sierrita. See text and caption to Fig. 1 for explanation. (From Titley 1982)

and Beane (1982) and Ahmad and Rose (1980) reported detailed analyses of the
evolution of fluids through the mineralization process at Sierrita, Arizona, and
Santa Rita, New Mexico, respectively. In both deposits, high temperature-high
salinity fluids, possibly the result of demonstrated boiling, were followed by
lower temperature, lower salinity late fluids, a general pattern which seems to be
emerging. For example, potassic alteration at Sierrita occurred in association
with 35 - 40 wtOJo NaCI equivalent fluid at 300° - 370°. Slightly later fluids were
less saline at 10- 15 wt% NaCI and hotter at 400° - 430°, at which temperature
boiling occurred. Mineralization from these latter solutions ranged from 370°C
down to 200°C. At Santa Rita, deposition occurred at 600°C or more down to
200°C, also from two main types of fluid, one of high 30 - 60% NaCI equiva-
lent, the other less saline at 1% - 20% . Boiling and condensation were common
phenomena, especially in the high salinity type. In general, fluids responsible for
hypogene metallization range from 800° down to 200 °C with an apparent zenith
of metal sulfide precipitation importance at 375 ° - 325°C. Complex fluid
histories, including boiling, condensation, and resultant variations in oxidation,
salinity, and volatile content are common. Indicated pressures range from 200 to
2000 bars, or from depths of 0.5 to 5 km, with emphasis on the shallower depths.
The high salinity fluids are distinctive and repetitive enough that their presence in
fluid inclusions localized over mineralized volumes is explorationally significant
in distinguishing potentially productive from barren occurrences.
 202 1.M. Guilbert

Fluid Compositions

A third area of great recent progress is that of the geochemical and thermal
nature of hydrothermal fluids. That progress comes from at least four
subdisciplines, including synthetic (experimental) studies, thermodynamics,
isotope studies, and fluid inclusion analytic techniques. Synthetic studies have
culminated in such papers as the Meyer and Hemley (1967) and Rose and Burt
(1979) papers on the K20-Alz03-SiOz-HzO-HCI system and its alteration-
mineralization implications and a series by Burnham, most recently Burnham
and Ohmoto (1980), on hydrothermal fluid generation in intermediate to felsic
magmas and the compositions of those fluids. Thermodynamic approaches, es-
sentially starting with Helgeson's 1964 treatise on complexing and hydrothermal
ore deposition, have been dramatically advanced through improved basic
thermodynamic data in pertinent systems such as HzO - NaCI- COz through the
use of computers for their speed, enormous data bases, and graphic capabilities.
The use of activity-activity diagrams - in which the compositions of fluids in
equilibrium with minerals and mineral assemblages in terms of ratios of ions in
solution can be examined - have greatly clarified hydrothermal fluid evolution
patterns. It is now convenient, for example, to relate alteration and mineraliza-
tion (Fig. 3) by analyzing K +/H + and Cu +/H + behavior with respect to
Fe +z/(H +)z at appropriate temperatures. Uncertainties are still important at high
temperatures where even basic measured thermodynamic data are scant, but each
year sees more progress. Computer management of mass transfer, thermal
transfer, reaction progress, and solid solution-fluid equilibration is increasingly
valuable.

+I"!-.
(\j

~~
+

bn
C>
o
py
muse
(a) (b)

Key to Minerals:
biot = biotite ep = chalcopyrite orth = orthoclase
bn = bornite mag = magnetite py = pyrite
chI = chlorite (14..\) muse = muscovite

Fig. 3a, b. Two activity-activity diagrams showing silicate and oxide-sulfide phases in equilibrium
during pen mineralization-alteration as functions of ion ratios and time (along the arrows). The ion
ratios describe the fluid in equilibrium with potassic biotite-magnetite, biotite-orthoclase-magnetite,
orthoclase-chalcopyrite, etc., to muscovite-pyrite phyllic assemblages. (From Beane and Titiey 1981)
Recent Advances in Porphyry Base Metal Deposit Research 203

Isotopic studies have had such a far-reaching and profound impact that their
results are difficult even to summarize. Taylor (1979) has reviewed oxygen-
hydrogen isotopic systematics that have verified meteoric water involvement in
porphyry systems described in the next section, and Ohmoto and Rye (1979) have
done much the same for sulfur and carbon isotopes. Sulfur isotopic analysis is
now useful for establishing the geochemical origins of sulfur in sulfides, for geo-
thermometry, and for analysis of geochemical conditions and mechanisms of
deposition. Ohmoto and Rye report two maxima of o34s values in PCDs at - 3
to + 1 and + 8 to + 15, suggesting that the complexed sulfur was derived largely
from igneous sources either directly or by resolution of earlier magmatic sulfides.
Some deposits (Galore Greek, Canada, and Morococha, Peru) may involve some
sedimentary sulfur. PBMD sulfate-sulfide isotopic temperatures typically fall in
the 450° - 650°C interval. Early stable isotopic studies by Garlick and Epstein
(1966) at Butte, Montana, and by Sheppard et al. (1971) at Santa Rita, New
Mexico, both in the USA, first established the importance of meteoric water in-
vasion of PBMD systems. Subsequent oxygen-hydrogen isotopic analyses have
confirmed that importance.
Fluid inclusion analytic techniques have yielded results as profoundly impor-
tant as those of isotopic ones. They have indicated - as stated above - that
fluids to evolve through time, that they are commonly hot and strongly saline
early in mineralization history, and that boiling is common in the environment of
deposition. The inclusion petrographer sees daughter mineral species in the
vacuoles that include halite (NaCl), sylvite (KCI), anhydrite (CaS04), gypsum
(CaS04·2H20), several mixed Ca - Mg - Na - K hydrous sulfate salts, pyrite,
chalcopyrite, hematite, magnetite, and several others that reveal the chemical
complexity of hydrothermal fluids. Carbon dioxide, methane, carbon disulfide,
and several organic and organo-metallic compounds have also been identified.
Study of mineral paragenesis and paragenetic sequence in skarn systems has also
greatly clarified fluid behavior in porphyry systems and their wall rocks (Einaudi
et al. 1981).

Fluid Sources and Flow Paths

Speculation over the last hundred years about the degree to which meteoric fluids
mingle with hypogenic ones have been replaced in the last decade by factual in-
formation and deduction stemming largely from isotopic studies and geochemi-
cal-geophysical-geomathematical modeling. Stable isotopic studies - particular-
ly of hydrogen-oxygen and sulfur - and radiogenic ones - especially strontium-
rubidium - have assured us not only that meteoric waters can enter hydrother-
mal systems, but also that they do, commonly overwhelmingly. Early discussions
of the sort by White (e.g., White 1957, 1968) lead to formulation of models of
vapor-dominated shallow hydrothermal systems (e.g., White et al. 1971, White
1981), thereby providing conceptual support to the isotopic findings of Taylor
and his colleagues and others. These studies documented substantial involvement
in PBMD of water with incontrovertible surficial deuterium-hydrogen and
204 1.M. Guilbert

18/160xygen signature. Important single early studies include Sheppard et al.

(1971), and many isotopic studies of individual deposits have followed (Field and
Gustafson 1976, Eastoe 1982) which require involvement of meteoric waters.
How that water enters has been studied by Whitney (1975), Cathles (1977),
Henley and McNabb (1978), and especially Norton (1978, 1982). Computer
management of geophysical, geochemical, thermal, and kinetic data permit con-
struction of two-dimensional profiles of the consequences of shallow intrusion of
a pluton into the crust. " Average" geologic compositing of PBMD charac-
teristics with assigned permeabilities, solution densities and viscosities, and
absolute and relative temperatures and heat fluxes permits computation of ther-
mal-hydrothermal dynamics, including kinetics. Norton (1978, et seq.) defined
"sourcelines" (the "starting front" of convectively driven fluids), "pathlines"
(the travel paths of aliquots of fluid), and "source regions" (the total volume
from which fluids convectively move). Norton indicates that the source region
might extend as far as 5 km from a 2.7-km-wide 4.5-km-tall stock intruded to
within 2 km of the surface. Pathlines are probably not the simple arcuate inward
and upward bending arrow shapes so commonly sketched, but rather are
complicated by quantitative and qualitative permeability variations, local
thermal perturbations, and variations in density-specific gravity-viscosity rela-

1 I- - f _-sulf urous fumarol es __ ____ depth.

__ 0km

1'\'"
__ --:::::- \I 1-/ r'"

-f :1-.
. .-;: -. .
___ - ---- I -, \ .---
___ ~~ 1 --- grpu~ater
~
boll s
\ .1 ~ ~

polymetallic ~elns

,,",,,~ w,,~
",- -"...groundwater mixes With
~apor 2

condensate

\
t£Section at Pang.Jna 3
'~ , pyrrtlc shell

,,
,
, -'
.I
I
4

I
I
I?

+ +

Fig. 4. Interrelationships between magmatic and meteoric waters, salinity, temperature, boiling, and
alteration-mineralization in porphyry systems as deduced at Panguna by Eastoe (1982)
Recent Advances in Porphyry Base Metal Deposit Research 205

tions in the fluid, variations in turn affected by T - P - X, alteration progress,

and other variables. Figure 4 shows some of the complications of fluid flow pre-
dicted by theoretical and observational studies of the sort described in the preced-
ing paragraphs.

Fluid Chemistry and Physics

A clearer understanding of the physicochemical nature of hydrothermal fluids

and the nature of metal complexing in them has resulted from the studies briefly
described above. Fluid inclusion studies have demonstrated that mineralizing
aqueous fluids in PCD systems are NaCl-saturated and on or near the two-phase
boiling surface in the NaCI- H 20 system (Ahmad and Rose 1980). Eastoe (1982)
concluded that at Panguna "the copper was deposited mainly by salt-rich liquid
expelled directly from the magma. Significant salt-rich liquid may also have con-
densed from a vapor plume" above a boiling zone (Fig. 4). Variations in the
fugacities of volatile components are vital; volatile constituents depress the
critical point of a hydrothermal fluid, while nonvolatile solutes, in general,
increase it. Oxygen fugacity is relatively high (10- 30 to 10- 35 ) near the hematite-
magnetite buffer to sustain the sulfate so commonly recorded in porphyry
systems as anhydrite and also the hematite-magnetite paragenesis, although
magnetite predominates (Fig. 3). Sulfur fugacities of 10- 10 to 10- 12 atm buffer
chalcopyrite-pyrite-tennantite assemblages at appropriate temperatures; higher
I(S2) requires enargite-bornite assemblages. Complex ions of the metals are dom-
inated by mixed metal halides, with lesser involvement of complex ions involving
(HS) - and S02. Eastoe (1982) determined that vapor transport of complexes of
the transition elements is variable. Vapor transport of Fe in the vapor plume of
Henley and McNabb (1978) and Fig. 4 is insignificant; it can be important for
Cu, Zn, and Mo. Space does not permit further review; the scholar is referred to
Eastoe (1982), Beane and Titley (1981), the Barnes volumes (1967 and 1979), and
the many articles already referred to for further detail.
Most importantly, the physics and chemistry of the hydrothermal fluid are
yielding to modern study. The greatest needs are almost certainly in the area of
physical, chemical, and thermodynamic properties of the basic fluids in the
system oxygen-hydrogen, ultimately with consideration of important amounts of
Na, CI, K, and so on to include the metals themselves.

Conclusions

The results of each of the bodies of information above - plate tectonics, miner-
alization-alteration fluid histories, fluid compositions, fluid sources and flow
patterns, and fluid physical geochemistry - are continuing to loop back into
each other. The flow of field, chemical analytic, isotopic, and petrographic data
continues to permit refinement and extension of theoretical analyses. Field
206 J. M. Guilbert

studies are of a whole new generation in the 1980s, including such analyses of
veinlet histories as that by Haynes and Titley (1980). Careful, quantitative altera-
tion mapping (Guilbert and Park 1986), the determination of mineralogic-chemi-
cal equilibrium by phase rule and petrographic analysis, and the measurement of
alteration mineral solid solution characteristics and cryptic zoning in alteration
assemblages are ongoing and will feed back - with the fracture density, fluid in-
clusion, and isotopic studies described - to provide increasingly convincing and
refined mass balance-mass transfer P - T - X relationships. Quantitative altera-
tion mapping permits computerization of alteration data, and for the first time
allows regression analysis to identify academic and explorationally important
variables in that context. The data base growth in tectonic, structural, and litho-
tectonic information improves exploration for these globally important ore-
forming systems. Many of the techniques of analysis that are described here for
I-type systems will apply to S-type ones, although the results will of course differ.
Computerized X-ray diffractometric data, mineral polytype studies, and electron
microprobe studies are all generating mineralogic directional variation data that
is valuable to both science and exploration. Progress in field, quantitative labora-
tory, and digital-numerical analytic studies will provide a productive decade of
the 1980s for both I-type and S-type mineralized-altered PBMD-endowed ter-
ranes.

References

Ahmad SN, Rose AW (1980) Fluid inclusions in porphyry and skarn ore at Santa Rita, New Mexico.
Econ Geol 75:229-250
Beane RE, Titley SR (1981) Porphyry copper deposits, Part II, Hydrothermal alteration and mineral-
ization. Econ Geol 75th Anniversary Vol, pp 235 - 269
Burnham CW, Ohmoto H (1980) Late-stage processes of felsic magmatism. In: Ishihara S,
Takenouchi S (eds) Granitic magmatism and related mineralization. Min Geol Spec Iss 8, Soc Min
Geol, Tokyo, Jpn, pp 1-12
Cathles LM (1977) An analysis ofthe cooling of intrusives by groundwater convection which includes
boiling. Econ Geol 72:804 - 826
Damon PE (1970) Correlation and chronology of ore deposits and volcanic rocks. US Atom Energ
Comm Annu Rep COO-689-130
Damon PE, Shafiqullah M, Clark KF (1981) Age trends of igneous activity in relation to metallo-
genesis in the southern Cordillera. In: Dickinson WR, Payne WD (eds) Ariz Geol Soc Digest
14:137 -154
Eastoe CJ (1982) Physics and chemistry of the hydrothermal system at the Panguna porphyry copper
deposit, Bougainville, Papua New Guinea. Econ Geol 77:127 -153
Einaudi MT, Meinert LD, Newberry RJ (1981) Skarn deposits. Econ Geol75th Anniversary Vol, pp
317-392
Farmer G, DePaolo DJ (1983) Origin of Mesozoic granite in the western United States and implica-
tions for pre-Mesozoic crustal structure: II. Nd and Sr isotopic studies of barren and Cu - Mo
mineralized granite in Precambrian craton. J Geophys Res 88:3379 - 3401
Field CW, Gustafson LB (1976) Sulfur isotopes in the porphyry copper deposit at EI Salvador, Chile.
Econ GeoI71:1533-1548
Garlick DG, Epstein S (1966) The isotopic composition of oxygen and carbon in hydrothermal
minerals at Butte, Montana. Econ Geol 61 :1325 -1335
Guilbert JM (1981) A plate tectonic-lithotectonic classification of ore deposits. In: Dickinson WR,
Payne WD (eds) Ariz Geol Soc Digest 14:1-10. (Available Ariz Bur Geol Univ Ariz, Tucson,
USA)
Recent Advances in Porphyry Base Metal Deposit Research 207

Guilbert JM, Park CF, Jr (1986) The geology of ore deposits. Freeman, New York, 985 pp
Gustafson LB, Hunt JP (1975) The porphyry copper deposit at EI Salvador, Chile. Econ Geol
70:857-912
Haynes FM, Titley SR (1980) The evolution of fracture-related permeability within the Ruby Star
Granodiorite, Sierrita porphyry copper deposit, Pima County, Arizona. Econ GeoI75:673-683
Heidrick TL, Titley SR (1982) Fracture and dike patterns in Laramide plutons and their structural
and tectonic implications. In: Titley SR (ed) Advances in geology of the porphyry copper deposits.
Univ Ariz Press, Tucson, Ariz, pp 73 - 92
Helgeson HC (1964) Complexing and hydrothermal ore deposition. Pergamon Press/MacMillan,
New York, 128 pp
Henley RW, McNabb A (1978) Magmatic vapor plumes and ground water interaction in porphyry
copper emplacement. Econ Geol 73:1-20
Hollister VF, Anzalone SA, Richter DH (1975) Porphyry copper deposits of southern Alaska and
contiguous Yukon Territory: CIM Bull 68, 756:104-112
Keith SB (1978) Paleosubduction geometries inferred from Cretaceous and Tertiary magmatic
patterns in southwestern North America. Geology 6:516 - 521
Knapp RB, Knight JE (1977) Differential thermal expansion of pore fluids: Fracture propagation and
microearthquake production in hot pluton environments. J Geophys Res 82:2515 - 2522
Lowell JD, Guilbert JM (1970) Lateral and vertical alteration-mineralization zoning in porphyry ore
deposits. Econ Geol 65:373 - 408
Mayo EB (1958) Lineament tectonics and some ore districts of the southwest Min Eng 10:1169-1175
Meyer C, Hemley 11 (1967) Wall rock alteration. In: Barnes HL (ed) Geochemistry of hydrothermal
ore deposits, Chap 6. Reinhart, New York, pp 167 - 235
Mitchell AHG, Garson MS (1972) Relationship of porphyry copper and circum-Pacific tin deposits to
palaeo-Benioff zones. Trans Inst Min Meta1l81:BI0-25
Mitchell AHG, Garson MS (1981) Mineral deposits and global tectonic settings. Academic Press
(Geol Ser), London New York, 405 pp
Mutschler FE, Wright EG, Luddington S, Abbott JT (1981) Granite molybdenite systems. Econ Geol
76:874-897
Nielson RL (1%8) Hypogene texture and mineral zoning in a copper-bearing granodiorite stock,
Santa Rita, New Mexico. Econ Geol 63:37 - 50
Norton D (1978) Sourcelines, sourceregions, and pathlines for fluids in hydrothermal systems related
to cooling plutons. Econ Geol 73:21-28
Norton DL (1982) Fluid and heat transport phenomena typical of copper-bearing plutons environ-
ments: Southeastern Arizona. In: Titley SR (ed) Advances in geology of the porphyry copper
deposits. Univ Ariz Press, Tucson, Ariz, pp 59-72
Norton DL (1984) Theory of hydrothermal systems. Annu Rev Earth Planet Sci 12:155 -177
Ohmoto H, Rye RO (1979) Isotopes of sulfur and carbon, Chap 10. In: Barnes HL (ed) Geochemistry
of hydrothermal ore deposits, 2nd edn. John Wiley & Sons, New York, pp 509-567
Patchett P J (1983) Importance of the Lu - Hf isotopic system in studies of planetary chronology and
chemical evolution. Geochim Cosmochim Acta 47:81-91
Preece RK III, Beane RE (1982) Contrasting evolutions of hydrothermal alteration in quartz mon-
zonite and quartz diorite wall rocks at the Sierrita porphyry copper deposit, Arizona. Econ Geol
77:1621-1641
Rehrig WA, Heidrick TL (1972) Regional fracturing in Laramide stocks of Arizona and its relation-
ship to porphyry copper mineralization. Econ Geol 67:198 - 213
Rehrig WA, Heidrick TL (1976) Regional stress during the Laramide and late Tertiary intrusive
periods, Basin and Range Province, Arizona. Ariz Geol Soc Digest 10:205 - 228
Rose AW (1970) Zonal relations of wallrock alteration and sulfide distribution at porphyry copper
deposits. Econ GeoI63:920-936
Rose AW, Burt DM (1979) Hydrothermal alteration. Chap 5. In: Barnes HL (ed) Geochemistry of
hydrothermal ore deposits, 2nd edn. John Wiley & Sons, New York, pp 173-235
Sawkins FJ (1972) Sulfide ore deposits in relation to plate tectonics: J Geol 80:377 - 397
Sawkins FJ (1984) Metal deposits in relation to plate tectonics. Springer, Berlin Heidelberg New York
Tokyo, 325 pp
Sheppard SMF, Nielsen RL, Taylor HP, Jr (1971) Hydrogen and oxygen isotope ratios in minerals
from porphyry copper deposits. Econ Geol 66:515 - 542
208 J. M. Ouilbert: Recent Advances in Porphyry Base Metal Deposit Research

Sillitoe RH (1972) Relation of metal provinces in western America to subduction of oceanic litho-
sphere. Oeol Soc Am Bull 83:813 - 818
Sillitoe RH (1981) Ore deposits in Cordilleran and island-arc settings. In: Dickinson WR, Payne WD
(eds) Ariz Oeol Soc Digest 14:49-69 (Available Ariz Bur Oeol, Univ Ariz, Tucson, Ariz, USA)
Sillitoe RH, Halls C, Orant IN (1975) Porphyry tin deposits in Bolivia. Econ Oeol 70:913 - 927
Sillitoe RH, Jaramillo L, Damon PE, Shafiqullah M, Escovar R (1982) Setting, characteristics, and
age of the Andean Porphyry Copper Belt in Colombia. Econ Oeol 77:1837 -1850
Sutherland Brown A (1976) Morphology and classification. In: Sutherland Brown A (ed) Porphyry
copper deposits of the Canadian Cordillera. CIM Spec Vol 15, pp 51- 55
Taylor HP, Jr (1979) Oxygen and hydrogen isotope relationships in hydrothermal mineral deposits,
Chap. 6. In: Barnes HL (ed) Oeochemistry of hydrothermal ore deposits, 2nd ed. John Wiley &
Sons, New York, 798 pp
Titley SR (1982) The style and progress of mineralization and alteration in porphyry copper systems.
In: Titley SR (ed) Advances in geology of the porphyry copper deposits. Univ Ariz Press, Tucson,
Ariz, USA, pp 93 -116
Westra 0, Keith SB (1981) Classification and genesis of stockwork molybdenum deposits. Econ Oeol
76:844-873
White DE (1957) Magmatic, connate, and metamorphic waters. Oeol Soc Am Bull 68:1659-1682
White DE (1968) Environments of deposition of base-metal ore deposits: Econ Oeol 63:301- 335
White DE (1981) Active geothermal systems and hydrothermal ore deposits. Econ Oeol 75th
Anniversary Vol, pp 392 - 423
White DE, Muffler LJP, Truesdell AH (1971) Vapor-dominated hydrothermal systems compared
with hot-water systems. Econ Oeol 66:95 - 97
Whitney JA (1975) Vapor generation in a quartz monzonite magma: A synthetic model with applica-
tion to porphyry copper deposits. Econ Oeol 70:346 - 358
The Geochemical Behaviour of Copper and Molybdenum
in Ore-Forming Processes
A. V. KUDRIN, L. N. VARYASH, Yu. N. PASHKOV, and V. I. REKHARSKy 1

Abstract

The geochemical behaviour of molybdenum and copper during the formation of

copper-molybdenum ore deposits is considered. Molybdenum and copper miner-
alization is closely associated with the hydrothermal-metasomatic associations.
The probable values of the physicochemical parameters of copper- and molyb-
denum-bearing hydrothermal solutions are estimated on the basis of calculations
and experimental data. The problem of the types of copper and molybdenum
species in ore fluids is considered. Hydroxyl alkali chloride or alkali molybdate
complexes are suggested as the molybdenum-bearing species responsible for ore
transport. Copper chloride complexes are suggested as copper-bearing transport
forms. Cooling of the hydrothermal solutions appears to be one of the main ore
deposition factors.
Geological and geochemical investigations indicate that the scale of copper
and molybdenum mineralization in copper-molybdenum deposits is directly asso-
ciated with the hydrothermal-metasomatic rock associations (Rekharsky 1980,
Shipulin et al. 1975).
The feldspar-quartz hydrothermal-metasomatic association is considered as
the main productive ore formation in respect to molybdenum mineralization. It
is characterized by feldspar and quartz-feldspar metasomatites with potassium
feldspar-biotite-quartz assemblages in their inner zones. The metasomatic rocks
are accompanied by quartz and quartz-feldspar veins with molybdenite as the
main ore mineral. Molybdenum mineralization in feldspar-quartz metasomatic
rocks is present usually in the form of disseminations in the inner zones of the
metasomatites.
Copper mineralization is associated mainly with the quartz-sericite metaso-
matic association and partly with propylites, berezites (quartz-sericite-carbonate
metasomatites), etc. The quartz-sericite association is represented by quartz-
sericite metasomatites as well as by veins and veinlets of a similar mineral com-
position. The main copper ore mineral, chalcopyrite, is localized in veins and
marginal zones in the quartz-sericite metasomatites. Chalcopyrite is pre-
dominantly associated with pyrite, whereas molybdenite, bornite, tetrahedrite
and other ore minerals are found only in subordinate quantities.

1 The Institute of Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry, USSR
Academy of Sciences, Moscow, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
210 A. V. Kudrin et al.

According to fluid inclusion investigations the temperature of mineralization

of the ore-bearing hydrothermal-metasomatic rocks is estimated in the range of
500° -:200°C. The molybdenum ores were formed mainly from 450° to 300°C
and copper ores from 320° to 200°C (Pavlova 1978, Popov 1977).
Alkaline metal chlorides and carbon dioxide are considered as the dominant
components of the ore-forming hydrothermal solutions on the basis of the fluid-
inclusion chemical composition. The latter was determined in samples of quartz
coprecipitated with molybdenite and chalcopyrite. In several cases high fluorine
concentrations were determined. The published data on fluid-inclusion chemical
composition and its dependence on temperature and the character of the altered
rocks are sometimes controversial and few in number. The highest temperature
inclusions as usual are characterized by high salinity, up to 40 - 50070 eq. wt% of
NaCI in several cases. The fluid inclusions from early stages of mineralization
contain not only "brines", but a low-density water-carbon dioxide fluid with
carbon dioxide concentration up to 8 - 10 mol %. The temperature decrease as
usual resulted in a decrease of total salinity notwithstanding the reverse relations
(Eastoe 1982, Popov 1978, Rekharsky et al. 1980).
Sulphide concentration and redox state of the ore-forming solutions are of
great importance for the interpretation of copper and molybdenum geochemical
behaviour. The probable range of these parameters may be evaluated using the
geochemical interpretation of the mineral assemblages of the ore deposits in
question (Distler et al. 1982). Hydrogen sulphide concentrations during ore
deposition were estimated as 10 - 2 to 10- 3m, gradually decreasing with a
decrease in temperature. The oxygen fugacity is evaluated as corresponding to
that of the Fe304 - Fe203 and Ni - NiO buffers.
The data on the physicochemical parameters of the ore-forming solutions
based on experimental data are used in examining the role of different molyb-
denum and copper compounds in ore transport and the causes of ore deposition.
Among the probable forms of molybdenum migration, complex compounds
of molybdenum with sulphur, fluoride and chloride complexes as well as molyb-
dation and products of its hydrolization and some other compounds were
proposed by a number of investigators (Korenbaum 1970, Rekharsky 1973). Yet
according to current experimental data on the stability of these compounds at
elevated temperatures and the results of thermodynamic calculations, only the
hexavalent molybdenum hydroxyl complexes are considered as important for
molybdenum transport (Kolonin and Kosals 1979, Kolonin and Laptev 1975,
Turgarinov et al. 1973). It should be emphasized that the conclusions presented
are based on extrapolation of the experimental values obtained in the low-tem-
perature range up to 200° - 250°C and published elsewhere.
To test these conclusions as well as to evaluate the possible role of trivalent
and tetravalent molybdenum in providing the reduced character of hydrothermal
solutions we carried out an experimental determination of the Mo0 2 solubility in
aqueous solutions. The experiments were undertaken for different Eh and pH
values at the temperatures of molybdenite formation. The hexavalent molyb-
denum hydroxyl complexes, HMo04(sol) and H2MoO~(SOI)' were found to be the
main molybdenum species over a broad interval of redox conditions up to 10- 5
102 for Ni - NiO buffer and acidity from 0.1 m HCI up to 0.001 m NaOH over-
 The Geochemical Behaviour of Copper and Molybdenum in Ore-Forming Processes 211

Fig. 1. The solubility of molybdenite in

hydrothermal solutions.
-2
Im(Na,K)Cl = 1 mol kg- 1 H 20;
IaH2S(sol) = 0.003 mol kg- 1 H 20;
-3 /H2S = 0.03 (t = 450°C);
/02 values correspond to Ni - NiO buffer.
1 450°C; 2 350°C; 3 300°C

=
2:
o -5
:::<:
o
E' -6

-7
--1

-8
---2
-,-,-3

-9

2 7 8

lapping the maximum probable variations in the hydrothermal solutions. The

existence of chlorine-bearing (?) low-valence molybdenum cation forms is
probable only in extremely acid solutions with HCI ~ 1 m.
The solubility of molybdenite was calculated using the experimental solubility
data on Mo0 2 • The latter was found to form hydroxyl complexes in the tempera-
ture interval of molybdenite formation (300° - 450°C) (Fig. 1). It was assumed
that the total activity of sulphur (S -2) is equal to the average value within the
estimated range (0.003 m; /H2S = 0.03 at 450°C) and the redox parameters are
equivalent to the equilibrium values of /02 for the Ni - NiO buffer.
According to a number of workers, the minimum concentration of the metal-
bearing component in ore-forming solutions producing economic deposits is
estimated as not lower than 10- 4 m (Crerar and Barnes 1976). These conditions
are attained only in fairly alkaline environments at 300°C in ore solutions equi-
librated with respect to the molybdenite (Fig. 1). The solubility of molybdenite is
moderately increased with increase in temperature up to 450°C, but "ore-
forming" molybdenum concentrations in the form of the HMo04(sol) complex
are predicted only in highly alkaline solutions. The molybdenum mineralization
at porphyry copper deposits is often associated with feldspar-quartz-muscovite
mineral assemblages. Thus we assume that the ore-forming solutions at the stage
of ore deposition were near-neutral or weakly acidic. Evidently the molybdenum
hydroxyl complexes could not support molybdenum transport even in the high-
temperature region. The dominant role in molybdenum transport is attributed to
some other species.
We assumed a substantial role was played by the alkaline metal chlorides
during the formation of molybdenum-bearing complexes, as the chlorides are im-
portant ingredients of the hydrothermal solutions. Our experiments on the
solubility of Mo0 2 in alkaline chloride solutions at 300° -450°C and under a
pressure of 500 bar indicated that the ore-component content in the ore solutions
212 A. V. Kudrin et al.

increases with an increase of temperature and salinity. The dependence of

solubility of Mo0 2 on the solution chemistry is an argument for the existence
of two types of Mo + 6 compounds, such as HMoO 4 • (K, Na) CI(-;ol) or
(K, Na) HMoO~(sol).
The formation of the first type of molybdenum-bearing complex is presum-
ably governed by a substantial decrease in the NaCI and KCI dissociation con-
stants with an increase in temperature in the range 300° -450°C, resulting in an
increase in the activities of the neutral molecules NaCI~sol) and KCI~sol). This
prediction is in agreement with the marked increase in Mo0 2(cryst) solubility
within the experimentally investigated temperature interval. On the other hand,
the formation of (K, Na) HMoO~(sol) could be considered as the result of a general
tendency for the formation of neutral complex compounds in the high-tempera-
ture region. For an unambiguous solution of the problem, additional experi-
ments are needed. Yet the available data can be used for approximate estimates
of MoS 2 solubility resulting in the formation of the assumed species (Fig. 1). The
MoS 2 solubility in 1 m (Na, K)CI solutions at 300°C is similar to its solubility in
nonchloride solutions, whereas at 450°C the difference is about 1.5 orders of
magnitude. Thus high-temperature neutral and even weakly acidic hydrothermal
solutions are to be considered as productive in respect of molybdenum ore
deposition.
Copper, like molybdenum, is known to exist in different valence states in
hydrothermal solutions. The available data indicate that the predominent trans-
port of copper is performed in the monovalent form, whereas molybdenum is
transported in the highest valence state Mo(VI). According to data on the
chemical compositions of ore-forming solutions, we predict chloride and
sulphide complexes as the most probable copper-transporting species.
The formation of complex compounds of Cu(I) with chlorine was inves-
tigated by a number of authors (Ahrland and Kawsthorne 1970, Crerar and
Barnes 1976, Nikolaeva et al. 1974, Varajash and Rekharsky 1981). According to
these data, copper-bearing chloride complexes are observed to be dominant even
at low chlorine concentrations (_10- 3 m). The formation of CuCI~sol)'
CuCli(sol), CuOHCI(-;ol) and other species is a function of acidity and chlorine
concentration in solution. Yet estimates of the stability of these species,
especially in the high-temperature region, are controversial. Thus, according to
Nikolaeva et al. (1974), the dominant copper-bearing species in dilute
(10- 2 -10- 3 m) chloride solutions at temperatures up to 150°C is represented by
CUCli(sol)' while Crerar and Barnes (1976) reported that CuCI~sol) is formed even
in highly concentrated solutions (up to 10.4 m NaCl) in the 200° - 350°C tem-
perature range. Evidently the disagreement could lead to erroneous conclusions
concerning the ore-forming properties of the complexes.
A number of experiments on metallic copper solubility at 300° - 350°C in
chlorine-bearing solutions under specified redox conditions were carried out. It
was found that CuCli(sol) complex compounds are the dominant species in acidic
and near-neutral solutions with 0.003 - 1.0 m NaCI concentration. In more dilute
solutions the monovalent ion, CU(~ol)' is the dominant copper-bearing species.
Thus, our experimental data indicate that CuCI~sol) is unstable not only at 25°C
(Ahrland and Kawsthorne 1970), but also in high-temperature solutions.
The Geochemical Behaviour of Copper and Molybdenum in Ore-Forming Processes 213

These experimental data, as well as thermodynamic data on CUCI2(sol) and

CU(~ol) for the low temperature region (Ahrland and Kawsthorne 1970, Naumov
et al. 1971, Nikolaeva et al. 1974), were used in the solubility calculations.
Chalcopyrite was taken to be in equilibrium with magnetite (350°C) and pyrite
(300° and 200°C), and the solubility data for a 1.0 m NaCI solution with
CUCI2(sol) and CU(~ol) as the products are plotted on Fig. 2. On the same figure the
chalcopyrite solubility with the Cu(HS)2(sol) complex as the final product is
shown according to Crerar and Barnes' (1976) data. The hydrogen sulphide
activity was taken as 0.003 m, similar to that in the molybdenum solubility
calculations. From Fig. 2 it is evident that the copper concentration in the
solution abruptly increases with increase in temperature from 200° up to 350°C.
At 200°C the ore-productive copper concentrations are predicted in solutions
with pH < 3, while at 350°C the region is extended up to near-neutral conditions.
The dominant copper-bearing forms are represented by the chloride compounds,
yet the hydrogen sulphide complexes playa subordinate role even at high H 2S
concentrations. It is evident that the cation CU(~OI) cannot be considered as one of
the transport forms of copper in the hydrothermal solutions.
Thus, the copper-bearing chloride complexes and molybdenum-bearing
hydroxyl alkaline chloride complexes (or alkaline molybdate complexes) are the
most probable species for the hydrothermal transport of copper and molyb-
denum.
In this section we discuss the factors involved in copper and molybdenum
deposition in the form of chalcopyrite and molybdenite, respectively. On Fig. 3
the evolution of a hypothetical hydrothermal solution containing ore-productive
copper and molybdenum concentrations within the 500° - 200°C temperature
interval in terms of calculated and experimental data is shown. The total alkali

--1
-1 ---2

Fig. 2. The solubility of chalcopyrite in

equilibrium with magnetite (350°C) and
pyrite (300° and 200°C) in hydrothermal
solutions. 0Cl- = 1 mol kg- t H 20;
.EoH2S (sol) = 0.003 mol kg- t H 20.
Cu(HS)i 1 350°C; 2 300°C; 3 200°C; solid lines cor-
-92~~3~~4~~5~~6~~7~~8~--­
respond to /02 Ni - NiO; thin lines = /02
Fe304 - Fe203
pHt
 214 A. V. Kudrin et al.

Fig. 3. The solubility of molybdenite (1)

and chalcopyrite (2) in hydrothermal
-1
350 0 solutions. Im(Na,K)Cl = 1 mol kg- 1
300 0 450 0 H 20; IaH2S(sol) = 0.003 mol kg- 1 H 2O;
-2 / /02 values correspond to the Ni - NiO
/ I
/
/
/350 0 buffer
/
~ -3 / I
0 / I 1
/
0 / 1300 0
~ -4 / 1
/ /
/
- /
C. -5
0
L
~ -6
S!'
-7
---
-8 ---1
--2
-9
2 3 4 5 6 7 8
pHI

chloride concentration was taken as 1.0 m, .EaH2S(sol) = 0.003 m, redox conditions

correspond to the Ni - NiO buffer, copper and molybdenum concentration start-
ing values were taken as 10 - 3 m and 10 - 4 m, respectively. Evidently the acidity of
the solution at 500°C is close to the neutral region, as the opposite case would
correspond to ore over saturation with respect to molybdenum in the acidic
region or to copper in the alkaline region at extremely high temperatures
exceeding the known interval of ore deposition.
A temperature decrease from 500° to 350°C would result in molybdenite
deposition as its solubility abruptly decreases. At the same time the hydrothermal
solution remains undersaturated in relation to chalcopyrite. An increase in
acidity as a result of the temperature decrease is considered as an additional
factor favouring molybdenite crystallization and preventing copper deposition.
Chalcopyrite could be deposited in equilibrium with molybdenite during progres-
sive cooling of the hydrothermal solution (350° to 300°C) as the chalcopyrite
solubility in this temperature range is lower in relation to the initial copper
concentration even within the moderately acid region. At 300°C, practically all
molybdenum would be deposited in the solid form, whereas copper deposition is
prolonged to 200°C. The neutralization of the solution is considered as an
additional factor favouring a similar evolutionary path.
The stability of molybdenum- and copper-bearing compounds in hydro-
thermal solutions is governed not only by temperature and acidity variations, but
also by changes in redox conditions and variations in S - 2 and alkali metal
chlorides contents in solution. Unfortunately the lack of reliable data on
numerical values of these parameters and their dependence on the stages of
evolution of the hydrothermal process prevents the estimation of the specific role
of each parameter in the ore-forming process. We suggest that the dominant in-
dependent factor in the deposition of the molybdenum sulphides and copper
The Geochemical Behaviour of Copper and Molybdenum in Ore-Forming Processes 215

sulphides is the temperature decrease of the hydrothermal solutions. Our model

of the successive formation of molybdenite and chalcopyrite in terms of gradual
cooling of the hypothetical hydrothermal solution is considered to some extent to
agree with this conclusion. The model is consistent with the oberserved prevalent
association of molybdenum ore mineralization with high-temperature feldspar-
quartz metasomatites, whereas the copper ores are observe in assemblage with
low-temperature quartz-sericite metasomatites.

References

Ahrland S, Kawsthorne J (1970) The stability of metal halide complexes in aqueous solution. VII.
The chloride complexes of copper (I). Acta Chem Scand 24:157 -172
Crerar DA, Barnes HL (1976) Ore solution chemistry. 5. Solubilities of chalcopyrite and chalcocite
assemblages in hydrothermal solution at 200° to 350°C. Econ Geol 71 :772 -794
Distler VV, Rekharsky VI, Pashkov YuN et al. (1982) The geochemistry of ore forming processes.
Nauka, Moscow, 270 pp (in Russian)
Eastoe CJ (1982) Physics and chemistry of the hydrothermal system at the Panguna Porphyry copper
deposits, Bougainville, Papua New Guinea. Econ Geol 77:127 -153
Kolonin GR, Kosals YaA (1979) Physico-chemical parameters of molybdenum-tungsten ore
formation in ore deposits of quartz-greisen rock association. In: Main parameters of the natural
processes of the endogeneous ore formation, vol I. Nauka, Novosibirsk, pp 236 - 251 (in Russian)
Kolonin GR, Laptev YuV (1975) The spectrophotometric investigation of the stability of sulfide
molybdenum complexes versus temperature. In: Experimental mineralogical investigations
(1974 -1975). Inst Geol Geophys, Novosibirsk, pp 38 - 43 (in Russian)
Korenbaum SA (1970) Physico-chemical conditions of tungsten- and molybdenum-bearing mineral
crystallization in hydrothermal solutions. Nauka, Moscow, 268 pp (in Russian)
Naumov GB, Ryzhenko BN, Khodakovsky IL (1971) Handbook of thermodynamic constants for
geologists. Atomizdat, Moscow, 240 pp (in Russian)
Nikolaeva NM, Ehrenburg AM, Skorokhod LS (1974) The influence of temperature on the equilib-
rium constant values of the reactions of substitution in the halide complexes of copper (I). Izv Sib
Otd Akad Nauk SSSR Ser Chim 3:44 - 48 (in Russian)
Pavlova IG (1978) The copper-porphyry ore deposits. Nedra, Leningrad, 274 pp (in Russian)
Popov VS (1977) The geology and genesis of the copper- and molybdenum-porphyry ore deposits.
Nauka, Moscow, 201 pp (in Russian)
Rekharsky VI (1973) The geochemistry of molybdenum in the endogeneous processes. Nauka,
Moscow, 268 pp (in Russian)
Rekharsky VI (1980) Geochemistry of copper-molybdenum deposits of the Lesser Caucasus. In:
European copper deposits. Dep Econ Geol Fac Min Geol, Belgrade, Yug, pp 104 -107
Rekharsky VI, Pashkov YuN, Ignatjeva IB, Nosik LP (1980) The formation of ore-bearing hydro-
thermal-metasomatic rock formations in molybdenum ore deposits. In: Thermobarogeochemistry
and ore genesis. Vladivstok, pp 83 - 86 (in Russian)
Shipulin FK, Rekharsky VI, Rozbianskaya AA et al. (1975) The intrusions, hydrothermal-metaso-
matic rocks and copper-molybdenum ore mineralization. Nauka, Moscow, 231 pp (in Russian)
Tugarinov AI, Khodakovsky IL, Zhidikova AP (1973) Physicochemical conditions of molybdenite
formation in hydrothermal uranium-molybdenum ore deposits. Geokhimiya 7:975 - 984 (in
Russian)
Varjash LN, Rekharsky VI (1981) On monovalent copper behaviour in hydrothermal chloride solu-
tions. Geokhimiya 7:1003 -1008 (in Russian)
Batholith-Volcano Coupling in the Metallogeny
of Porphyry Copper Deposits
P. E. DAMON 1

Abstract

In the course of our study of the porphyry copper deposits of Mexico, the author
became aware of the vast extent of the composite batholiths of Sonora and
Sinaloa. We observed that the mineralized porphyritic plutons intrude the pene-
contemporaneous batholith and are associated with dominantly andesitic
volcanism. An explanation of this phenomenon is provided by the phenomenon
of dome-in-dome structures observed by Ramberg (1981) in centrifuge model
experiments where less dense material is overlain by material of greater density. I
have applied this concept to the case of a cooling batholith. The interior of a
batholith is inherently unstable gravitationally with respect to the more rapidly
cooling upper part which is in contact with the country rock. As a consequence,
the upper reaches cool faster, begin crystallizing sooner, and eventually become
denser than the lower reaches. Thus, the interior of the batholith becomes
buoyant and intrudes the upper part. The early batholith forms a broad, gentle
dome intruded by later domes. The process is repeated with buoyant upwelling of
the interior of the intrusive domes forming cupolas, some of which breach the
surface and form cones or stratovolcanos.
As a result of the severe temperature regimen and dome-in-dome structures,
large-scale hydrothermal fluid transfer occurs dominated by volcanic orifices
constituting hydrothermal "pumps" or "artesian vents". Blind porphyries, those
that do not breach the surface, have hydrothermal systems that are subsidiary to
the master volcanic artesian vents. Such master hydrothermal systems have iso-
topically dated lifetimes equal to the isotopically dated long-lived hydrothermal
activity of stratovolcanos, i.e., up to at least 2.5 m. y. The fluids exported to the
volcanic pumps carry ore components extracted from the batholith and country
rock. The ore components are distributed and precipitated in the manifold forms
of occurrence observed within porphyry systems.

1 Laboratory of Isotope Geochemistry, Department of Geosciences, University of Arizona, Tucson,

AZ 85721, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 217

Introduction

During the past decade, the author has had the opportunity to work with Ing.
Guillermo P. Salas, former Director of the Consejo de Recursos Minerales, and
other colleagues (see acknowledgments) on a project entitled "Dating the ore
deposits and defining the metallogenetic provinces of the Republic of Mexico" .
This project had as the initial impulse the construction and modification of the
metallogenetic map of Mexico (Salas 1975) and developed the following con-
cepts: (1) Most of the metallic ore deposits of the Southern Cordillera of the
U.S.A. and Mexico are related to volcanism; (2) volcanos and related plutons are
not randomly spaced and usually occur in volcanic arcs; (3) volcanic arcs of dif-
ferent ages can be distinguished and their paleogeography reconstructed through
field work and isotopic dating; (4) volcanic arcs migrate with time; (5) as they
migrate away from the associated trench, the silicic magma composition becomes
more alkalic (Dickinson and Hatherton 1967, Dickinson 1975, Keith 1978, Clark
et al. 1982, for Mexico see Damon et al. 1981b, Fig. 8); (6) the compositions of
ore deposits change as magma compositions change (ibid.); (7) this results in
metallogenetic provinces that are long, relatively narrow belts, more or less
parallel to the paleo-subduction zone (Fig. 1, see caption for explanation); (8)
erosion is accelerated by volcanic arc tumescence and by continental uplift
proceeding inward from the convergent margin (Damon 1979, 1983); (9) as a
result, secondary enrichment takes place, and if erosion is not interrupted, the
ore deposit will be removed (Livingston et al. 1968, Damon et al. 1983a); (10) ore
preservation depends upon tectonic events, such as faulting and sedimentation,
or the ore being buried under a volcanic blanket, such as within the Sierra Madre
Occidental and southwestern U.S.A. (Livingston et al. 1968, Damon et al.
1983a).
In recent years, there has been much interest in the Mesozoic-Cenozoic
Cordilleran volcanic arcs that are related to the convergent margin of western
North America associated with the present cycle of seafloor spreading. It is now
generally recognized that these arcs have migrated over long distances. By earlier
Jurassic time a volcanic arc appears to have been established throughout the
length of the Southern Cordillera from northern Nevada through Chiapas, Mexico
(Damon et al. 1981b). By early Cretaceous time, the volcanic arc had regressed
westward to the ancestral Pacific margin from which it transgressed slowly east-
ward again in early Cretaceous time (Lindgren 1915, Henry 1975, Coney and
Reynolds 1977, Anderson and Silver 1977, Clark et al. 1979, Damon et al. 1981b,
Chen and Moore 1982). During the Laramide orogeny, the volcanic arc moved
rapidly inland to the east, toward the Gulf of Mexico, and then following the
Laramide orogeny, it rapidly regressed to the Pacific margin (Coney and
Reynolds 1977, Clark et al. 1979, Damon et al. 1981b).
In this paper I will restrict my attention to the Mexico states of Sonora and
Sinaloa (Fig. 3). Dated igneous rocks from Sonora and Sinaloa are plotted on
Fig. 2 relative to the envelope of dates for igneous rocks of all of northern
Mexico. It is generally accepted that all of these igneous rocks, both plutonic and
volcanic, with the exception of the two-mica granites and pegmatites and post-
subduction extrusive rocks, are related to the migrating Cordilleran volcanic arcs
218 P.E. Damon

M1NERAL ASSEMBLAGE

c..-f.ktH F--;'3 Hol-Sb

Cu-Mot:::-:::::::::j k ({(:lu·
W ~ mc..-Bo

Au -AQtf.~;\;~' f.

PACIFIC
PLATE

, .............
• T. . .
MJI'lII\o DI.rr,<:t

o
KIn

Fig. 1. The metallogenetic provinces of Mexico (modified from Clark et at. 1979). The patterned
areas enclose provinces in which significant numbers of important deposits are known to occur (see
also Salas 1975). Note succession of north-north west-trending metallogenetic provinces resulting
from different stands of the Cordilleran Volcanic Arc which becomes increasingly alkalic with
distance from the trench (Damon et at. 1981 b). The Cu - Mo province is the focus of interest in this
paper. Note the concentration of deposits in that province as shown in the inset of Fig. 3

of North America. The two-mica granites and pegmatites are related to the ther-
mal event accompanying the volcanic arc, but have no known volcanic equiv-
alents. The postsubduction basalts and andesites are related to a taphrogenic
event, the Basin and Range Disturbance, that occurred throughout the Basin and
Range Province and southward in western Mexico to Cabo Corrientes at the
northern end of the Middle American Trench. Figure 3 shows that the
Cordilleran Jurassic volcanic arc that terminated in late Jurassic time (145 Ma),
was followed by a long magma gap in Sonora and Sinaloa lasting 40 m. y. to the
west and 90 m. y. to the east. The arc then rapidly regressed westward through
Sonora and Sinaloa during Oligocene and early Miocene time. This was followed
by dominantly basaltic and andesitic associated with taphrogenesis.
During the eastward advance of the volcanic arc in the Cretaceous period,
porphyry copper deposits were emplaced, forming the metallogenetic province
 Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 219

20
• 0

40

a. 60
ai
cti
~ 80
\oJ
>-
o Plulon . unm.lntrollud
• Pluton · mrntl"Olteo
• felsic ",o!conlc' l nctI bGiolt1c vo6conitl
D Bo.~u

"" Two mlco QrdnUft and peqm oUlu

o
15
o
17
o
190 o

100 200 300 400 500 600 700 800 900 1000
KILOMETERS FROM TRENCH

Fig. 2. Migration of the Cordilleran volcanic arcs within northern Mexico with specific reference to
Sonora and Sinaloa (modified from Clark et al. 1979). The outer envelope includes over 95070 of all of
the dated rocks of northern Mexico. The dates plotted are for Sonora and Sinaloa only. Note the
magma gap of varying duration between Jurassic and Cretaceous-Cenozoic magmatism. In late
Jurassic-early Cretaceous time, the volcanic arc had returned to the Pacific margin, but then migrated
slowly eastward until the beginning of rapid convergence during the Laramide orogeny, at 80 Ma.
During the Laramide orogeny, 80 to 40 Ma, the Cordilleran Volcanic Arc migrated rapidly eastward.
Following extensive changes in plate motions at the end of the Laramide (40 Ma), the Cordilleran
Volcanic Arc migrated rapidly back to the Pacific margin. Taphrogeny followed with extrusion of
dominantly basaltic and andesitic magma. Data are from Damon and Mauger (1966), Henry (1975),
Gastil et al. (1976, 1978), Anderson and Silver (1977, 1979), Gastil and Krummenacher (1977),
Anderson and Roldan (1979), Clark et al. (1979), Anderson et al. (1980), Damon et al. (1983a, b)

shown in Fig. 1. Twenty-nine deposits form a long, NNW-trending linear belt

(inset, Fig. 3). This was followed by emplacement of tungsten deposits associated
with two-mica peraluminous granites. The Au - Ag metallogenetic province
shown in Fig. 1 was formed during the regression of the volcanic arc to the
Pacific coast.
In this paper, as previously mentioned, I will use the porphyry copper metal-
logenetic province of Sonora and Sinaloa as an example, in order to demonstrate
the importance of batholith-stratovolcano coupling in the formation of porphyry
copper deposits.
220 P.E. Damon

J .., ..,
I I LEGEND
STATES Of AM~RICA
I D Lot. Tertio,-, of'ld Ouolttnoty

I
SICiIne",ary Roth

D Middl, T.rlie" Sedimentary

······,
Roell.

GIJ
•
........
. .
'. '
Pcm - EoctM Vokanic Roc ...

Sonoton-Slnaloon 8alhol11h ond

Anodattd VOk:onlc Rotb

D
L.aromidl Pot"phyry Coppw
Deposit.

Pr,· Lor-amid, C,.tac:.oul

P... ...,.,y ~ Do9Otlt
I LoU/R.....oIr

oz:t~ ClO 00 100 HO

••y••""
.t\I ___- - ._ _ _ _- ....._ (Il" .

PACIFIC

OCEAN

"

Fig. 3. Simplified geologic map of Sonora and Sinaloa emphasizing the Laramide batholiths. Note
that the actual extent of the batholiths must be much greater because voluminous Mid-Tertiary
volcanic extrusions, dominantly ignimbritic, and sedimentation in basins have buried much of the
batholith . Data are from the Carta Geologica de la Republica Mexicana compiled by Lopez Ramos
(1976)
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 221

Field and Age Relationships

The late Cretaceous-early Cenozoic (Laramide) batholiths of Sonora and Sinaloa

cover much of the two states. The batholiths and their associated volcanic rocks
crop out over an area of 13600km2 in Sinaloa and 12800km2 in Sonora, com-
prising a total area of 26400 km 2 (Fig. 3). This area may be compared with the
area of the Sierra Nevada batholith, which covers some 60000 km 2• The
Sonoran-Sinaloan batholiths are covered by Mid-Tertiary volcanic rocks of the
Sierra Madre Occidental (King 1939) and of the Basin and Range Province
(Shafiqullah et al. 1980), as well as the sedimentary rocks of the Oligocene-
Miocene Baucarit Formation (King 1939) and late Tertiary-Quaternary
sediments (Fig. 3). Thus their actual extent must rival some of the greatest
batholiths of the world. They intrude rocks of varying age ranging from older
Precambrian (Livingston and Damon 1968) through marine limestone and clastic
rocks of early Cretaceous age (King 1939). The batholiths are composite
intrusions consisting primarily of calc-alkaline granodiorite and granite intruded
by penecontemporaneous late-stage porphyries, many of which contain econom-
ic copper deposits (Fig. 3, inset).
The bulk of the premineralization batholith cooled at a rate which seems, at
first, to be surprisingly fast, as demonstrated by the concordance of many horn-
blende and biotite K - Ar dates for samples of unmineralized portions of the
Sonora-Sinaloa batholiths (Table 1). K - Ar mineral dates may be related to the
time at which the temperature is too low to activate significant diffusion (the
"blocking temperature").
Despite the widely different approximate blocking temperatures for horn-
blende (500 0q and biotite (300 o q, 670/0 of the hornblende-biotite pairs showed
no significant difference in age, demonstrating very rapid cooling such as shown
for the fast cooling curve in Fig. 4. There is a significant difference in age
between the pairs constituting the remaining 33% of the samples. This indicates a
slow cooling rate as shown also in Fig. 4. The rate of cooling of a batholith is
very much dependent upon its depth of emplacement. For example, Larsen
(1945) has calculated that the top of a batholith with a cover of 5 km of sedi-

Table 1. Statistical analysis of 46 biotite-hornblende pairs from unmineralized portions of the

Sonora-Sinaloa batholiths

Difference between hornblende Number of samples 070

and biotite age (Ma) in class interval

-0, i.e., number falling between ± 2 0- 31 67.4

assuming normal distribution
>0 to ~5 Ma 9 19.6
>5 Ma to ~10 Ma 5 10.9
>10Ma 1 2.1
-
1: = 46 100.0

Note: Data for Sonora are from Damon et al. (1983a), Gastil and Krummenacher (1977), Gastil et al.
(1977). Data from Sinaloa are from Henry (1975) and unpublished data from this laboratory.
222 P.E. Damon

Fig. 4. Difference in age for different min-

erals for fast and slow cooling regimens.
Blocking temperatures are a function of
cooling rate; consequently, the blocking tem-
800 peratures for fast and slow cooling rates are
o ZIRCON different. The values used here are generally
EI HORNBLENDE accepted "rule of thumb" values based upon
t:. MUSCOVITE several decades of field and laboratory ob-
600 servations
o BIOTITE

400

200 ~--SLOW COOLING

r---~----- RAPID COOLING

O~-~-.---.---~r-----r----
10 20 30 40 TIME (rna)

mentary rock would cool from its emplacement temperature of 820 to 300°C in
about 2 Ma. With a cover of 1 km, it would take only about 0.1 Ma to cool.
Seven Ma would be required for cooling to 300°C at a depth of 3 km within the
batholith having a cover of 5 km. With reference to Larsen's graphs, the dif-
ferences in hornblende-biotite ages (Table 1) are consistent with the upper part of
the batholiths having intruded to within 5 km of the surface, and erosion having
cut deep enough to expose, for the most part, the upper reaches of the batholith;
but in other parts erosion has extended to a depth of as much as 5 km below the
upper surface, exposing the slowly cooled depths of the batholith. This would
require erosion rates of only about 5 to 20 cm per 1000 yrs. since emplacement of
the batholiths. Such rates would be quite low for a mountainous area (Gregory
and Walling 1973). The Rocky Mountains of the U.S.A. have been estimated to
be eroding at a rate of 50 cm per 1000 yrs. (ibid.). However, the parallel ranges
and Basin and Range Province on the west flank of the Sierra Madre Occidental
in which the batholiths crop out are not as rugged as the Rocky Mountains and,
so, the rate of 5 to 20 cm per 1000 yrs. is reasonable.
During field reconnaissance associated with our study of the porphyry copper
deposits of Mexico, we observed that the most common wall rocks of the
porphyries are penecontemporaneous early-stage batholith rocks and associated
Laramide volcanic rocks (Damon et al. 1983a). Table 2 shows that 69070 of the
wall rocks for Mexican porphyries are comprised of penecontemporaneous
batholithic rocks and associated volcanic rocks.
Epicontinental limestones that are barren in copper comprise 17%, and all
other sedimentary rock types comprise only 14% of the wall rocks. Partly as a
consequence of this relationship, Damon et al. (1983a) concluded that copper
was being transported from the calc-alkalic batholithic wall rocks to the
hypabyssal roots of the overlying volcano, "where the metals are distributed and
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 223

Table 2. Wall rocks intruded by Mexican porphyry copper deposits

Type of wall rock Frequency

Number 010

Penecontemporaneous composite batholith 15 37

Laramide volcanics 13 32
Epicontinental limestones 7 17
Shales, sandstones, and metasediments 5 12
Pre-Laramide submarine basalts 1 2
-
E = 41 100

precipitated in the manifold forms of occurrence observed within the porphyry

systems" (Damon et al. 1983a, p. 1069).
Table 3 includes K - Ar dates from six locations (see inset, Fig. 3) for both
mineralized and unmineralized facies of the batholiths. If the anomalous datum
for Mina Washington is excluded (possibly reset by post-volcanic arc two-mica
granite), the average difference is 3.1 ± 0.8 (0-) Ma with the mineralized facies
always younger. Thus, about 3 m. y. after emplacement of the early stages of the
batholiths, the late-stage porphyries had been intruded and were in the process of
accumulating copper mineralization.
Following our early reconnaissance work (Damon et al. 1981a) and K - Ar
dating of the batholiths and their associated porphyries and volcanic rocks, Jean
Jacques Cocheme, of the Northwestern Regional Station of the Institute of
Geology, Autonomous National University of Mexico, mapped an area in
Sonora on the edge of the Sierra Madre Occidental (Damon et al. 1983b). A
slightly modified version of Cocheme's map is shown in Fig. 5. The map shows
that in areas where erosion has not cut too deep, hypabyssal cupolas are still
surrounded by Laramide volcanic rocks (NW area). On the other hand, where
erosion has exposed the batholith, the hydrothermally altered porphyries clearly
intrude the batholith (central and south areas on the map).
Summarizing, I have shown that the vast batholiths of Sonora and Sinaloa
rose close enough to the surface for a rapid cooling of their upper surface to have
taken place. Deeper areas cooled at a measurably slower rate. Plutons and
cupolas intruded the early-stage batholiths several million years after their
emplacement. Lastly, the porphyry copper deposits appear to be associated with
stratovolcanos (Sillitoe 1973), giving rise to the associated, dominantlyandesitic
volcanic rocks that are associated with the porphyries (see Fig. 5).

Dome-in-Dome Structures

The relationship of hypabyssal porphyries intruding the penecontemporaneous

batholith was incontrovertible, but at first puzzled me. Sillitoe's model (1973, his
Fig. 1) has small «2 km diam) hypabyssal porphyries passing into unminer-
Table 3. Age difference between unmineralized batholith and mineralized porphyry copper tv
~
Map No., name, state (Fig. 3) Coordinates Mineral and rock dated K-Ar date Difference
reference (Ma) (Ma)

(11) Mina Washington, Sonora 29°54'29"N Biotite, premineralized 56.4± 1.2

Damon et al. (1983a) 110 0 05'26"W granodiorite 10.7±1.6
29°53'47"N Sericite, mineralized 45.7±1.0
110 0 03'59"W breccia pipe
(18) Suaqui La Verde, Sonora 28°25'12"N Hornblende, premineralized 58.8± 1.3
Damon et al. (1983a) 109°49'01 "W granodiorite 2.1 ± 1.7
28°24'41"N Sericite, mineralized quartz 56.7 ± 1.1
109°48' 11 "W diorite (dissemination)
(22) Cerro Colorado, Chihuahua 26°17'10"N Biotite, premineralized 48.0± 1.2
Damon et al. (1983a) 107°20'36"W granodiorite 1.7 ± 1.6
26°15 '08"N Sericite, mineralized 46.3± 1.0
107°19' 46" W breccia
(24) Tameapa, Sinaloa 25°39' N Hornblende, premineralized 56.9± 1.2
Damon et al. (1983a) 107°23 ' W granodiorite 2.7 ± 1.6
25°40' N Secondary biorite, quartz monzonite 54.1 ± 1.1
107°21' W mineralized stockward
(28) Minas Las Higueras, Sinaloa 25°00'39"N Biotite, unmineralized 54.9± 1.2
Damon et al. (1983a) 107°13'42"W granodiorite 5.9± 1.6
25°00'39"N Sericite from dissemination 49.0± 1.0
107°13' 45 "W in mineralized granodiorite
(31) Mal pica, Sinaloa 23°14' N Biotite, premineralized 57.3±0.6
Henry (1975) 106°08' W granodiorite 3.3±1.7
23°16' N Biotite, breccia filling in 53.8±0.6
106°07' W granodiorite porphyry
Average difference 4.4± 1.4i7 "t:l

Average difference (excluding 3.1 ±0.7i7 ~

anomalous location 11) 0$l)
9
0
::s
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 225

D Recent alluvium - - Principal faults

[ ] Saucarlt formation - - Dikes

O
'.'
Mid Tertiary volcanic rocks
(Calcic lallte 10 rhyOlite)
Mine

o hyabyssal facies (cupolas wit"

hydrothermal alteration) o
-=-
I 2 3Km.

E:TI Laramide granodiorite Scol.

fill] ~~a~:~cl~~1esltes {luffs, flows

=
r=;:::::q Sedimentary rocks (limestones,
sandstones and breccias)

Fig. S. Geologic map of the region of Santa Rosa de Yecora, Sonora (modified from map by Jean
Jaques Cocheme in Damon et al. 1983b). Note that in the southern and central areas of the map, the
batholith is intruded by hypabyssal cupolas with hydrothermal alteration. In the northeast, the
Laramide volcanic cover has not yet been removed . Erosion has removed all but a small remanent of
the Paleozoic-Mesozoic pre-batholith sedimentary cover. This remanent crops out in the west-central
area of the map

alized phaneritic granodiorite at a depth of from 7 to 8 km. Branch (1976)

suggests that small, high-level stocks (1- 2 km diam), within and below much
larger stratovolcanos, are fed by a main volcanic conduit directly from the
mantle. In the Basin and Range Province of the southwestern United States,
where taphrogeny has more severely disrupted the basement than in the area of
226 P.E. Damon

this study on the flanks of the Sierra Madre Occidental in Sonora and Sinaloa,
Titley (1982) and Norton (1982) model the porphyry copper deposits as high-level
diapirs intruding basement host rocks that are at "slightly higher than normal"
temperatures (Norton 1982, p. 61). Norton suggests that a 30°C km -1 gradient
seems reasonable. However, where batholithic intrusions precede the hypabyssal,
porphyritic plutons, as in Sonora and Sinaloa, much higher temperature gradi-
ents must prevail.
The author's perplexity was relieved after a discussion with Dr. Hans
Ramberg of the University of Uppsala. After I described the field relationships to
Dr. Ramberg, he assured me that his scale models predict this sort of
relationship, and referred me to his discussion of dome-in-dome structures in
Chap. 11 of his book (Ramberg 1981). I then proceeded to model these structures
using his equations and a simple numercial computer model.
The model is shown in Fig. 6. For initial conditions, I chose 8 km of country
rock (unit 1) underlain by a granodioritic magma 8 km thick (unit 2 + 3), that in
turn is underlain by a more basic magma (unit 4) of greater density than the over-
lying magma ({!(4) > {!(2+3», which, consequently, is gravitationally stable. The
combined layer (i = 2 + 3) has a density less than the country rock ({!(2+3) < (!(1)
and hence is gravitationally unstable. The viscosity ratio critically determines the
wavelength, A, of the doming of the batholith's upper surface. I chose a viscosity
ratio, /1(1//1(2+3)' of 104, yielding a A of 470 km. A larger viscosity ratio would
produce an even larger A. Consequently, the upper surface has the form of a
broad, gentle dome. However, as the upper part of the batholith cools more
rapidly than the lower part, it will eventually have a higher density than the lower
part and the lower part will become gravitationally unstable. To model this, I
separated layer (i = 2 + 3) into two layers, i = 2 and i = 3, with {!(1) > {!(2). I chose
a viscosity ratio of /12//13 = 100, which yielded A = 50 km. Upon discussing this
with my colleague, Professor Spencer Titley, he pointed out that Pitcher (1972)
had mapped similar spacings for centered complexes of arcuate intrusions that
form an essential component of the coastal batholith of Peru. Ramberg (1981)
gives several other examples with similar spacings.
Layer 3, as it rises into the gently domed upper batholith and country rock, is
provided with space by the thinning of the overlying batholith and country rock.
The overlying material moves away from the rising domes and then down into
the space provided between the domes as layer 3 is "sucked" up into the rising
domes. The downward-moving material is compressed and deformed as the
dome rises. Ramberg (1981) has modeled a similar situation with two competent
modeling-clay sheets (his Fig. 11-13, p.256). After erosion (Fig.6f), the
resultant geologic terrain would look very much like the map of the Santa Rosa
area (Fig. 5).
Summarizing, the lower part of a rising and cooling batholith is inherently un-
stable with respect to the upper part, because eventually its density will be less than
the cooled upper reaches of the batholith. As a consequence, the lower part will rise
forming dome-in-dome structures. Ramberg's equations (1981) predict spacings of
the rising domes that are in accord with spacings of mapped plutonic complexes in-
truding into batholiths. This explains the field and age relationships between early-
stage batholiths and late-stage porphyries that we have observed in Mexico.
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 227

NO VERTICAL EXAGGERATION KM 1"'.-_-....

o 10 20 30 40 50
1:100000

Fig. 6a - f. Dome-in-dome structures: a Batholith forms. Layer 1 represents the covering country
rock. Layers 2 + 3 represents the internally stable granodioritic batholith underlain by diorite (layer 4)
which is denser than the granodiorite, hence gravitationally stable. bGentle doming is accompanied
by cooling and partial crystallization of the upper part of the granodiorite batholith. The upper part
gradually becomes denser than the lower part and gravitationally unstable. Layers 2 + 3 becomes
layer 2 and layer 3, with densities {h. and {!J, where {h. > l!3 ' c Buoyant lower granodiorite (layer 3)
forms rising domes intruding the country rock, and granodiorite layer 2 sinks to occupy the space left
by layer 3. d One dome breaches the surface to form a juvenile stratovolcano. The sinking part of the
batholith and cover rocks continue to be deformed. e A second stratovolcano is formed . The
previously formed volcano matures and deformation continues. Because the upper part of the domes
will also cool and crystallize faster, the lower part will become buoyant and form cupolas, which are
not shown here. Surface tumescence over the rising dome is also not shown. f Erosion occurs, leaving
a small remnant of country rock and exposing the domes that intrude the batholith. The resulting
outcrop pattern resembles the geologic outcrops within the southern part of the region of Santa Rosa
de Yecora, Sonora (Fig. 5)

Batholith-Volcano Coupling via Dome-in-Dome Structures

Figure 7 b is an expanded view of the section around the first stratovolcano to

erupt (see Fig. 6d). It is shown as it appears in Fig. 6e. The hypabyssal dome is
idealized. Actual domes would usually have a cupola or cupolas because, as in
the batholith itself, the exterior of the dome would cool against country rock and
batholith at lower temperatures. Consequently, the interior of the dome would
become buoyant with respect to the exterior with formation of cupolas. This phe-
228 P.E. Damon

o 5 10 20 30 40 50
NO VERTICAL EXAGGERATION (K M) r'I",."""",,·_~--'l~_J--___"";'
1' 400, 000

Fig. 7. a Illustration of an extensive hydrothermal system set up by the dome-in-dome structure.

Arrow show the direction of fluid flow with distance between arrows representing the distance
traveled by a fluid packet during 2 x 104 yrs. transit time (total transit time is 1.8 x lOs yrs.). Perme-
abilities and transit times were derived from Norton (1979, 1982). The permeability of the dome
beneath the stratovolcano was assumed to be very high, after Norton, because of extensive fracturing
(10 millidarcys, 1 md = 10- 11 cm 2 ). The permeability decreases to 2 md in the central part of the
dome, then increases to 4 md in less crystallized parts at depth. The sunken batholith was given a
permeability of 0.5 md at its contact with the country rock and 2 md at its contact with the diorite
below. The permeability of the country rock is high, 4 md, due to deformation with fracturing. The
large hydrothermal system with long transport distances is the result of high geothermal gradients and
the long lifetime of the stratovolcano (up to 2.5 m. y.). The stratovolcano is an orifice to the surface
that acts as a hydrothermal "pump". "Blind" domes, those that do not breach the surface, have
hydrothermal systems that are subordinate to the master stratovolcano. Each stratovolcano vents
enormous quantities of water and acid gas to the surface. The water transports metals to the
hypabyssal roots of the volcano where they are distributed and precipitated in the manifold forms of
occurrence observed in porphyry ore deposits. (Letters A to K designate different fluid pathways
during the 1.8 x lOs yrs. transit time). b Region of large stratovolcano in Fig. 6e (cupolas and
tumescence not modeled)

nomenon can be observed in the mapped area of Fig. 5 where the hydrothermally
altered hypabyssal intrusions occur as clusters of small area. These clusters may
have fed eruptive centers of a large stratovolcano or clusters of small cones. I
intend to numerically model this phenomenon in a future paper.
Norton has modeled fluid and heat transport phenomena around cooling
magmas with specific reference to the porphyry copper environment (Norton and
Knapp 1977, Norton 1978, 1979, 1982). His work has been based on a model in
which the pluton is a high-level diapir intruding country rock with a moderately
high geothermal gradient (30°C km -1). My model differs in that the dome
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 229

intrudes batholith and country rock that are under a more severe temperature
regimen ( - 90 ° C km - 1, see also Larsen 1945). The dome is also larger, although
cupolas, not modeled, would probably be the size of his small plutons. As a
result of their small size and less severe temperature gradient, the plutons
modeled by Norton cool rapidly, and "by 2 x 10 5 years elapsed time the thermal
anomaly consists only of a slight (3 km) upward perturbation in the 200°C
isotherm, which is centered on the pluton-granite contact, and a broad (3 km)
upward perturbation in the 100°C isotherm" (Norton 1982, p. 65).
Stratovolcanos may have a long, eruptive history. For example, the eruptive
history of the silicic eruptive center constituting the San Francisco Peaks strato-
volcano in northern Arizona and its satellitic cones began at 2.7 Ma and lasted to
0.2 Ma, indicating a lifetime of 2.5 m. y. (Damon et al. 1974). Page and
McDougall (1972) have demonstrated that pluton emplacement and subsequent
Au - Cu mineralization in the New Guinea Highlands occurred over a time
interval of from 1 to 2 m. y. The model presented here is consistent with such
long duration of an eruptive volcano-hydrothermal system. Figure 7a, which is
based on parameters from Norton's papers, shows that given the dome-in-dome
structure and consequently more severe thermal regime, hydrothermal transport
will continue much longer than predicted for high-level diapiric cupolas, and will
result in long-distance fluid transport. Neighboring "blind" domes that do not
have eruptive centers above them will contribute to the hydrothermal transport
system which will be governed by eruptive centers that vent to atmospheric pres-
sure. Fluids at depth will be under a high lithostatic-hydrostatic head, and so the
eruptive centers will act as fluid pumps or "artesian vents".
The model in Fig.7a assumes, after Norton's analysis, high permeability
within the hypabyssal roots of the stratovolcano, a somewhat higher perme-
ability than assumed by Norton for the country rock resulting from deformation
and consequent fracturing, and increased permeability within crystallizing parts
of the batholith and dome. Figure 7a shows the transport after 1.8 x 105 yrs. By
that time, fluid packets derived from neighboring domes will have reached the
central vent system. Interstitial fluids from a large part of the batholith will also
have arrived at the vent system. As pointed out by Norton, convection systems
will have been set up in the surrounding country rock. The convection cells will
probably be guided by fissuring at the axes of large-scale folds producing higher
permeabilities.

Consequence for Ore Deposits of Batholith-Volcano Coupling via Dome-in-

Dome Structures

Emmons (1940) emphasized the importance of batholiths in ore formation. Most

of his experience was within the Canadian Shield, where batholiths of vast extent
dominate the landscape and the imagination. He was less familiar with Neogene
volcanism and, as a consequence, volcanism does not appear as an important
controlling factor in his models. On the other hand, Sillitoe (1973), with much
experience within the Andes, where Neogene stratovolcanos dominate the land-
230 P.E. Damon

scape, produced a brilliant model of the tops and bottoms of porphyry copper
deposits in which stratovolcanos played a dominant role. His model used Andean
volcanos and porphyry copper deposits to illustrate cross-sections of strato-
volcanos and their hypabyssal roots at different levels of dissection. However,
Sillitoe's porphyry copper bottoms disappear into unspecified depths. The dome-
in-dome structures, as previously pointed out, provide effective batholith-
volcano coupling. Thus, the dome-in-dome structures serve to link Emmon's
batholiths with Sillitoe's stratovolcanos.
The stratovolcanic "pump" is capable of dominating a very large hydro-
thermal system, including interstitial batholith fluids as well as large volumes of
water from a more direct meteoric source. Table 4 presents the estimates of Zies
(1938) for the amounts of acid gas and water emitted by fumaroles in the Valley
of Ten Thousand Smokes, on the flanks of the Katmai andesitic cone, during the
year 1919. These amounts may be compared with Norton's (1979) estimates of
the fluid mass flux through the central parts of a porphyry copper deposit. He
estimates that about 1000 kg cm - 2 of fluid mass would pass through the central
parts of a porphyry copper deposit in 2 X 105 yrs. Thus, a porphyry of 3 X 10 10
cm2 area (3 km 2 ) would have had 3 X 10 10 metric tons of fluid mass pass through
it. This amount is equivalent to the flux of water to the surface in 63 yrs. at the
1919 rate of the Valley of the Ten Thousand Smokes (Table 4). When one
considers the long duration of the hydrothermal activity of stratovolcanos, their
dominant role in fluid transport becomes clearly evident.

Table 4. Amount of water and acid gases

emitted by fumaroles in the Valley of Ten
Thousand Smokes in 1919

Gas Metric tons per year

480.0 X 106
1.3 X 106
0.2 X 106
0.3 X 106

Source: Zies 1938.

Summarizing, dome-in-dome structures provide effective batholith-strato-

volcano coupling. The long lives of stratovolcanos (as much as 2.5 Ma) allow for
long-distance transport of fluids containing ore components derived from a large
volume of batholith and country rock. Each volcano forms a surface orifice at
which there is pressure release and export of enormous quantities of fluid to the
surface. Thus, hydrothermal solutions transport ore components from the
batholith as well as from the country rock to the hypabyssal roots of the volcano
where the metals are distributed and precipitated in the manifold forms of occur-
rence observed within the porphyry systems.
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 231

Summary and Conclusions

As a result of our study of the Sonoran-Sinaloan batholiths and their hypabyssal

plutonic intrusions, I have arrived at the following conclusions:
1. The vast batholiths of Sonora and Sinaloa rose close enough to the Earth's
surface for a rapid cooling of their upper surfaces to have taken place. This is
shown by 670/0 of hornblende-biotite K - Ar dates being concordant. The
remaining 33% are discordant, having been derived from deeper reaches of
the batholiths which cooled at a measurably slower rate. The distribution of
concordant and discordant K - Ar dates suggests that erosion occurred on the
western flanks of the Sierra Madre Occidental at the modest rate of from 5 to
20 cm per 1000 yrs.
2. Several million years after emplacement, the early-stage batholiths were
intruded by hypabyssal domes and porphyritic cupolas, some of which vented
to the surface as stratovolcanos of varying size and complexity.
3. The lower part of a rising and cooling batholith is inherently, gravitationally
unstable with respect to the upper part because eventually its density will be
less than the cooled upper reaches of the crystallizing batholith. As a
consequence, the lower part will become buoyant and rise, forming dome-in-
dome structures. Similarly, the interior of a rising dome will become buoyant
and rise, forming porphyritic cupolas, some of which will be at the roots of
stratovolcano vents. This explains the field and age relations between early-
stage batholiths and late-stage porphyries that we have observed in Mexico.
4. Modeling, using Ramberg's equations (1981) and assuming reasonable
viscosity ratios, predicts spacings of the rising domes that are in accord with
mapped spacings between intrusive complexes within batholiths.
5. Dome-in-dome structures from large-scale hydrothermal systems that last for
periods of time consistent with lifetimes of stratovolcanos, as much as 2.5
m. y. As a consequence, there will be long-distance transport of fluids, guided
by eruptive centers, and intimate hydrothermal coupling, via the dome-in-
dome structures, between the batholithic system and the stratovolcano with its
hypabyssal porphyritic plutonic roots.
6. Each stratovolcano forms a surface orifice at which there is pressure release
and export of enormous quantities of fluid to the surface. Thus, hydrothermal
solutions transport ore components from the batholith as well as from the
country rock to the hyabyssal roots of the volcano where the metals are dis-
tributed and precipitated in the manifold forms of occurrence observed within
the porphyry systems.
Lastly, it should be possible to evaluate the model in Figs. 6 and 7 through
geophysical and field geologic studies. The model predicts extension and con-
sequent thinning of the cover above the domes with transport to the margins
where deformed cover would fill in the space vacated by the rising domes. This
magmatotectonic deformation would be superimposed on the back-arc deforma-
tion that precedes the advancing volcanic arc. Under favorable conditions in
areas where the dome has intruded a relatively thick and layered sedimentary
sequence, geologic, gravity, and seismic studies could be used to define the
232 P.E. Damon

batholith-cover interface and structural analysis could ascertain whether the

deformational characteristics are appropriate. The geophysical studies could also
be used to detect blind domes that had not breached the surface to form strato-
volcanos. Also, it would be interesting to know what, if any, ore deposits are
related to the blind domes whose hydrothermal convection cells would be sub-
ordinate to the domes coupled to stratovolcanos.

Acknowledgments. This work would not have been possible without the consistent support and en-
couragement of Ing. Guillermo P. Salas, former Director of the Consejo de Recursos Minerales
(C.R.M.). The author is particularly grateful to Prof. Kenneth F. Clark of the Department of
Geological Sciences, University of Texas at El Paso, for many stimulating conversations and for his
companionship in the field aspects of this work; to Dr. Muhammad Shafiqullah for his constant and
able help both in the field and in the laboratory; and to Prof. John M. Guilbert for magnanimously
encouraging a geochronologist to enter into the "sacred sanctuary" of economic geology and for
giving much helpful advice. The author also benefited from stimulating conversations with his col-
leagues: Profs. Christopher Eastoe, Jibamitra Ganguly, Denis L. Norton, Spencer R. Titley, and
Joaquin Ruiz.
Many Maxican geologists have contributed logistical support and encouraged our effort. In
particular, I would like to thank Ings. Jaime Islas-Lopez, Fernando de la Fuente-Lavalle, Antonio
Bustamante-Yanez, and Miguel Carrasco Centeno of the C.R.M.; Drs. Ariel Echavarri-P., Director
of the Direccion de Mineria, Geologia y Energeticos of the State of Sonora; Dr. Guillermo A. Salas,
formerly head of the Department of Geology at the University of Sonora; and Ing. E. Nestor Silva-
Mejia of the Comision Federal de Electricidad.
I am grateful to my collaborators in the field aspects of this work: Ing. Jaime Roldan-Quintana
and Dr. Jean Jaques Cocheme, for stimulating conversation leading to further insights and for many
pleasant days in the field together. Dr. Cocheme's map of the Santa Rosa de Yecora area was
particularly helpful in my understanding of the relationship between the hypabyssal cupolas and the
batholith. Ings. Islas, Roldan, and Silva accompanied the author on several field trips. Ings. Juan
Carlos Garcia and Claudio Bartolini-Navarro assisted the author during field work in the Sierra
Madre Occidental.
Thomas C. Kirschner, Robert Butcher, and Claudio Bartolini drafted the final figures. The
manuscript was reviewed and edited by Austin Long, Joaquin Ruiz, Daniel Lynch, and Yemane
Asmerom. Alice Jeanne Woodward typed the manuscript and contributed many editorial comments
and corrections.

References

Anderson TH, Roldan JQ (eds) Field trip log. Geology of northern Sonora, guidebook, field trip
#27. Annu Meet Geol Soc Am, pp 74-91
Anderson TH, Silver LT (1977) U - Pb isotopic ages of granitic plutons near Cananea, Sonora,
Mexico. Econ Geol 72:827 - 836
Anderson TH, Silver LT (1979) The role of the Mojave-Sonora megashear in the tectonic evolution of
northern Sonora. In: Anderson TH, Roldan JQ (eds) Geology of northern Sonora, Guidebook,
field trip # 27. Annu Meet Geol Soc Am, pp 59 - 68
Anderson TH, Silver LT, Salas GA (1980) Distribution and U - Pb isotopic ages of some linea ted
plutons, northwestern Mexico. In: Crittenden MD Jr, Coney PJ, Davis GH (eds) Cordilleran meta-
morphic core complexes. Geol Soc Am Mem 153:269-283
Branch CU (1976) Development of porphyry copper and stratiform volcanogenic ore bodies during
the life cycle of andesitic stratovolcanos. In: Johnson RW (ed) Volcanism in Australia. Elsevier,
New York, pp 337 - 342
Chen JH, Moore JG (1982) Uranium-lead isotopic ages from the Sierra Nevada batholith, California.
J Geophys Res 87:4761-4784
Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits 233

Clark KF, Damon PE, Schutter SR, Shafiqullah M (1979) Magmatismo en el Norte de Mexico en
relacion a los yacimientos metaliferos, Associacion de Ingenieros de Minas, Metalurgista y
Geologos de Mexico. Mem Tec XIII:57
Clark KF, Foster CT, Damon PE (1982) Cenozoic mineral deposits and subduction-related magmatic
arcs in Mexico. Geol Soc Am Bull 93:533 - 544
Coney PJ, Reynolds SJ (1977) Cordilleran Benioff Zones. Nature (London) 270:403-406
Damon PE (1979) Continental uplift at convergent boundaries. Tectonophysics 61:307-319
Damon PE (1983) Continental uplift, compensation and shunting during trench-spreading center
collision. Tectonophysics 99:T1 - T8
Damon PE, Mauger RL (1966) Epeirogeny-orogeny viewed from the Basin and Range Province.
Trans Soc Min Eng 235:99-112
Damon PE, Shafiqullah M, Leventhal JS (1974) K - Ar chronology for the San Francisco volcanic
field and the rate of erosion of the Little Colorado River. In: Karlstrom TNV, Swann GA,
Eastwood RL (eds) Geology of northern Arizona, pt 1. Regional studies guidebook. Geol Soc Am
Rocky Mountain Sect Meet, pp 221- 235
Damon PE, Clark KF, Shafiqullah M, Roldan JQ, Islas LJ (1981a) Geology and mineral deposits of
southern Sonora and the Sonoran Sierra Madre Occidental (excursion 11). In: Ortlieb L, Roldan
JQ (eds) Geology of northwestern Mexico and southern Arizona, field guides and papers.
Guidebook, 1981. Cordilleran Sect Geol Soc Am, pp 369 - 428
Damon PE, Shafiqullah M, Clark KF (1981b) Age trends of igneous activity in relation to metal-
logenesis in the southern Cordillera. In: Dickinson WR, Payne WD (eds) Relations of tectonics to
ore deposits in the Southern Cordillera. Tucson, Ariz, Geol Soc Digest 4:137 -154
Damon PE, Shafiqullah M, Clark KF (1983a) Geochronology of the porphyry copper deposits and
related mineralization of Mexico. Can J Earth Sci 20:1052-1071
Damon PE, Shafiqullah M, Roldan JQ, Cocheme JJ (1983b) El batolito Laramide (90-40 Ma) de
Sonora. Pap 15th Nat Convent Asoc Ing Min Metal Geol Mex, Guadalajara, Jalisco, October 1983
Dickinson WR (1975) Potash-depth (K - h) in continental margin and intra-oceanic magmatic acrs.
Geology 3:43 - 56
Dickinson WR, Hatherton T (1967) Andesitic volcanism and seismicity around the Pacific. Science
157:801- 803
Emmons WH (1940) Principles of economic geology. McGraw-Hill, New York
Gastil RG, Krummenacher D (1977) Reconnaissance geology of coastal Sonora between Puerto
Lobos and Bahia Kino. Geol Soc Am Bull 88:189-198
Gastil RG, Krummenacher D, Bushee J, Barthelmy D (1976) La zona batolitica del sur de California
y el occidente de Mexico. Bol Soc Geol Mex 37:84 - 90
Gastil RG, Morgan JG, Krummenacher D (1977) Mesozoic history of peninsular California and
related areas east of the Gulf of California. In: Mesozoic paleogeography of western United States.
Pacific Coast Paleogr Symp
Gastil RG, Krummenacher D, Minch J (1978) The record of Cenozoic volcanism around the Gulf of
California. Geol Soc Am Bull 90:839 - 857
Gregory KJ, Walling DE (1973) Drainage basing form and process: A geomorphological approach.
Wiley & Sons, New York
Henry CD (1975) Geology and geochronology of the granitic batholithic complex, Sinaloa, Mexico.
Ph D Diss, Univ Texas, Austin
Keith SB (1978) Paleo-subduction geometrics inferred from Cretaceous and Tertiary magmatic
patterns in southwestern North America. Geology 6:516-521
King RE (1939) Geological reconnaissance in northern Sierra Madre Occidental of Mexico. Geol Soc
Am Bull 50: 1625 - 1722
Larsen ES (1945) Time required for the crystallization of the great batholith of southern and lower
California. Am J Sci 243A:399-416
Lindgren W (1915) The igneous geology of the Cordilleras and its problems. In: Problems of
American geology. Yale Univ Press, New Haven, Conn, pp 234-286
Livingston DE, Damon PE (1968) The ages of stratified Precambrian rock sequences in central
Arizona and northern Sonora. Can J Earth Sci 5:763 -772
Livingston DE, Mauger RL, Damon PE (1968) Geochronology of the emplacement, enrichment and
preservation of Arizona porphyry copper deposits. Econ Geol 63:30 - 36
234 P. E. Damon: Batholith-Volcano Coupling in the Metallogeny of Porphyry Copper Deposits

Lopez Ramos E (1976) Carta geologica de la Republica Mexicana. Com Carta Geol Mex, 4th edn,
. scale i : 2,000,000
Norton DL (1978) Sourcelines, source regions, and pathlines for fluids in hydrothermal systems
related to cooling plutons. Econ Geol 73:21- 28
Norton DL (1979) Transport phenomena in hydrothermal systems: The redistribution of chemical
components around cooling magmas. Bull Miner 102:471-486
Norton DL (1982) Fluid and heat transport phenomena typical of copper-bearing pluton environ-
ments. In: Titley SR (ed) Advances in geology of the porphyry copper deposits: Southwestern
North America. Univ Ariz Press, Tuscon, Ariz, USA, pp 59-72
Norton DL, Knapp RB (1977) Transport phenomena in hydrothermal systems: Cooling plutons. Am
J Sci 277:937 - 981
Page RW, McDougall I (1972) Ages of mineralization of gold and porphyry copper deposits in the
New Guinea Highlands. Econ GeoI67:1034-1048
Pitcher WS (1972) The coastal batholith of Peru: Some structural aspects. 24th Int Geol Congr Sect 2,
pp 156-163
Ramberg H (1981) Gravity, deformation and the Earth's crust, 2nd edn. Academic Press, London
New York
Salas GP (1975) Metallogenetic chart of Mexico. Geol Soc Am Map Chart Ser, Map MC13, scale
1 : 2,000,000
Shafiqullah M, Damon PE, Lynch DJ, Reynolds SJ, Rehrig WA, Raymond RH (1980) K-Ar
chronology and geologic history of southwestern Arizona and adjacent areas. Ariz Geol Soc Digest
12:201-260
Sillitoe RH (1973) The tops and bottoms of porphyry copper deposits. Econ Geol 68:799 - 815
Titley SR (1982) The style and progress of mineralization and alteration in porphyry copper systems.
In: Titley SR (ed) Advances in geology of the porphyry copper deposits. Southwestern North
America. Univ Ariz Press, Tucson, Ariz, USA, pp 93-116
Zies EG (1938) The concentration of the less familiar elements through igneous and related activity.
Am J Sci 35A:385 - 404
Space-Time Distribution, Crustal Setting and Cu/Mo
Ratios of Central Andean Porphyry Copper Deposits:
Metallogenic Implications
R. H. SILLITOE 1

Abstract

At least 80 porphyry copper centres, 18 of them major deposits, are known from
the central Andes of Chile, Argentina and southern Peru. They are assignable to
four longitudinal sub-belts of Meso-Cenozoic age, which become progressively
younger from Late Cretaceous in the west, through Paleocene and Late Eocene-
Early Oligocene, to Middle-Late Miocene in the east. A fifth, Late Carboni-
ferous-Early Permian sub-belt is overprinted by the Middle-Late Miocene sub-
belt.
The porphyry copper sub-belts are underlain by sialic crust ranging from
about 35 to 60 km in present thickness. Outcropping pre-Mesozoic basement
rocks range from 2,OOO-m.y. granulites to Paleozoic magmatic and sedimentary
units.
H~pogene Cu/Mo ratios of major deposits range from 22 to 88. No sys-
tematic variation in Cu/Mo ratios is discernable within or between the three
Cenozoic sub-belts. Neither are ratios dependent on the lithology or age of
known pre-Mesozoic basement rocks, nor on crustal thickness.
The lack of correlation between Cu/Mo ratios and spatial or upper-crustal
parameters is consistent with derivation of the principal metals in the porphyry
copper deposits from subcrustal levels. The fact that the regionally extensive
series of temporally and spatially discrete sub-belts parallels the axis of the Peru-
Chile trench, and migrated systematically away from it, strongly supports in-
volvement of fundamental subduction-related processes in the liberation of
metals incorporated into central Andean porphyry copper deposits.

Introduction

During the last decade or so, the source of metals (Cu, Mo, Au, Ag) and other
components (S, K) in porphyry copper deposits has remained a subject of debate.
Opinions have been polarized between sub crustal and crustal sources.
Burnham (1981), Oyarz(m and Frutos (1980), Sillitoe (1972, 1981) and Westra
and Keith (1981) support a subcrustal source, with subducted oceanic crust, both

1 8 West Hill Park, Highgate Village, London N6 6ND, England

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et a1.)
© Springer-Verlag Berlin Heidelberg 1986
236 R, H, Sillitoe

........................
.
. . . ., . , . . . , . .
. . . . .. " .. ... ... ... ..... .
~

100

150
km

I::;: :1SIAliC CRUST ~ OCEANIC CRUST

mmn SUBCRUSTAl'UNDERPLATE ' ~ LITHOSPHERIC MAN TLE

rn ASTHENOSPHERIC MANTlE
Fig. 1. Possible metal sources for porphyry copper deposits, Subcrustal sources include: subducted
oceanic crust (1a), the overlying mantle wedge (Ib), and an "underplate" of magmatic material
(derived from sources 1a and/ or 1b) at the base of the sialic crust (Ie), Crustal sources include: lower-
crustal sites of partial melting (2a), sites of selective contamination (2b) of subcrustal or lower-crustal
magma, and upper-crustal sites of hydrothermal scavenging (2e)

layer-2 basalts and layer-1 pelagic sediments, considered to playa preeminent

role (Fig. 1). In this model, the continental crust does not contribute a significant
proportion of porphyry copper components, although it does provide a site for
fractional crystallization of magmas to take place.
In contrast, Griffiths and Godwin (1983), Lowell (1980) and Titley and Beane
(1981) favor extraction of metals and other elements from the continental crust,
either at the site of partial melting, or by magma contamination resulting from
assimilation, or as a consequence of scavenging by circulating hydrothermal
fluids (Fig. 1). A number of materials, including shales (Jensen 1971), volcanic
rocks (Laznicka 1976) and preexisting mineralized bodies (Lowell 1974), have all
been proposed as specific crustal sources of metals in porphyry copper deposits.
Moreover, molybdenum, in particular, has been cited as a metal derived as a con-
taminant from thick sialic crust (e.g., Hollister 1975).
If metals are extracted from sialic crust, then it seems to be a reasonable
premise that Cu/Mo ratios of porphyry copper deposits should vary with
changes in crustal constitution or thickness. This supposition is inherent in
Hollister's (1975) explanation for high Cu/Mo ratios (deficiency of molyb-
denum) in porphyry copper deposits emplaced into the thinner, more mafic crust
of island arcs. Conversely, if Cu/Mo ratios fail to correlate with crustal
parameters, then a subcrustal metal source appears the more likely alternative.
The central Andes, comprising northern and central Chile, southern Peru and
northwestern Argentina, provide an ideal location to test the validity of the two
alternative metal sources. At least 80 porphyry copper deposits and prospects
(including groups of tourmaline breccia pipes) in the central Andes (Fig. 2; Table
1) possess an orderly distribution in both space and time, as well as being
emplaced into continental crust of different types and thicknesses. In this report,
the spatial and temporal distribution of central Andean porphyry copper miner-
alization and the characteristics of the region's crust are reviewed, preparatory to
a consideration of Cu/Mo ratios and their implications.
MetaIlogenic Implications 237

o KM 400

BOLIVIA

20 0

f
,- r-
\'\ I.i
\.... ._ /
' '\
"V
i
ANTOfAGASTA + I

24.., . 26 ,;')
25 """,
/ . 56
+ SALTA
CHILE ; · 57
\
\
,,27 .28 !
290.. I
31 .. '-1301,\ + SANMIGUfL
32" , ,- ' DE TUCUMAN
"33 I
"'34 I
;
35 /
$36! · 64
065

Major deposit.
in production
(no, 30 worked ou') 0

.
\
\
079 Important occurrence
Major deposit •
0
54 i Minor occurrence ..
(
i
~'
•80 Breccia pipes only ..
i Fig. 2. Distribution of porphyry
I copper-type mineralization in the
711' 6. 0
central Andes. Numbers keyed to
names in Table 1
238 R. H. Sillitoe

Table 1. Porphyry copper-type mineralization in Chile, Argentina and southernmost Peru (keyed to
Fig. 1)

Peru
1. Cerro Verde-Santa Rosa 41. EI Indio
2. Quellaveco 42. Andacollo
3. Cuajone 43. EI Altar
4. Toquepala 44. El Sauce
Chile 45. La Lana
5. Lluta (Campanani) 46. La Loica
6. Ticnamar 47. Llamuco
7. Chilpe (Q. Camarones) 48. Los Pelambres
8. Queen Elizabeth 49. Rio Blanco
9. Mocha 50. Disputada
10. Cerro Colorado 51. El Teniente
11. Yabricoya (La Planada) 52. Rosario de Rengo
12. Copaquire 53. Estero San Jose
13. Collahuasi 54. Galletue
14. Quebrada Blanca Argentina
15. El Abra 55. Pancho Arias
16. Conchi 56. Taca Taca
17. Puntillas 57. Inca Viejo
18. Pampa Norte 58. Bajo del Durazno
19. Chuquicamata 59. Bajo de Pampitas
20. Sierra Gorda 60. Bajo de La Alumbrera
21. Bella Esperanza 61. Bajo de San Lucas
22. Centinela 62. Mi Vida (Cerro Rico)
23. Lomas Bayas 63. Filo Colorado
24. Chimberazo 64. Cordillera de La Brea
25. La Escondida 65. Los Bayos (Famatina)
26. Imilac 66. Arroyo Chita
27. Carrizalillo de Las Bombas 67. San Francisco de Los Andes
28. El Jardin 68. Alcaparrosa
29. El Salvador 69. El Pach6n
30. Potrerillos 70. Yunque
31. San Pedro de Cachiyuyo 71. Cerro Mercedario
32. Cachiyuyo de Llampos 72. La Honda
33. Los Azules 73. La Colorada, Yalguaraz
34. Cabeza de Vaca-Zapallar 74. San Jorge, Yalguaraz
35. El Bolsico 75. Paramillos Norte
36. Los Morteros (Porvenir) 76. Paramillos Sur
37. Domeyko 77. Rio de Las Vacas
38. Chacritas 78. Santa Clara
39. Pablo y Pablo 79. Campana Mahuida
40. Los Loros 80. La Voluntad

Space-Time Distribution

In a previous compilation (Sillitoe 1981), porphyry copper-type mineralization in

Chile, Argentina and southernmost Peru was shown to be assignable to five sub-
belts (Fig. 3). Each sub-belt was emplaced during a discrete time interval: Late
Carboniferous-Early Permian (281 - 267 m.y.), Late Cretaceous (90 -73 m.y.),
Paleocene (67 - 54 m.y.), Late Eocene-Early Oligocene (41- 29 m.y.) and
Metallogenic Implications 239

/
I
15· (
\

~
PERU 0 ~M

•........
-..,
I

BOLIVIA
J::j
v:
zi
",,:
~!
"./
""i
.... :
-:
:x::
,:
vi
25· ::::>:
<>::
",,:
"-i
.... ;
Of
-
><::.
V):

,,:

Fig. 3. Porphyry copper sub-

belts in the central Andes. All
porphyry copper centres and
breccia pipes from Fig. 2 are
represented by dots. Numbers
refer to K - AT ages determined
on hydrothermal or, where un-
derlined, magmatic minerals.
Numbers in parenthesis are al-
ternative ages. K - Ar dates
given to nearest m.y.; where
Porphyry Copper
I deposit • more than one apparently re-
CHILE\ or occurrence liable date is available, an aver-
age is given. Data sources are
~j Middle-Late detailed in Sillitoe (1981), ex-
~~ Miocene cept for new determinations for
Quebrada Blanca (Hunt et al.
~
Late Eocene - 1::3
Early Oligocene S 1983), La Escondida (Minera
~~~"
7~'':,~
'a' ~:.:;~
Paleocene
Late Cretaceous ~
L:::l Utah de Chile, Inc., unpub-
lished data 1983), Domeyko
.~~.~, (Shell Chile, unpublished data
iy
r!}\]
r...• • ..•

lot':.~a;:::~;~~~~~
;
1983), El Indio (Araneda 1982),
i Disputada (Warnaars 1983),
: 75·
.I and Estero San Jose and Gal-
letue (Alfaro 1980)
240 R. H. Sillitoe

Middle-Late Miocene (16- 5 m.y.) (Fig. 3). Recently, seven more porphyry
copper centres, at Quebrada Blanca, La Escondida, Domeyko, EI Indio, Dis-
putada, Estero San Jose and Galletue, have been dated radiometrically by the
K - Ar method. (See Fig. 3 for data sources). Quebrada Blanca and La
Escondida are major deposits and have been confirmed as part of the Late
Eocene-Early Oligocene sub-belt. The major Disputada deposit, along with low-
grade porphyry-type mineralization related genetically to the recently discovered
EI Indio Au - Ag - Cu vein deposits (Araneda 1982), lie in the Middle-Late
Miocene sub-belt. Domeyko, Estero San Jose and Galletue are minor
occurrences and make up part of the Late Cretaceous sub-belt. Addition of these
new data to the previous compilation requires no significant modification in the
positions of sub-belts (Fig. 3), although the two separate portions of the Late
Eocene-Early Oligocene sub-belt may be united now with reasonable confidence
on the basis of the La Escondida determination, and the southern portion of the
Late Cretaceous sub-belt may be extended considerably. The Paleocene K - Ar
ages of porphyry copper deposits in southern Peru (Sillitoe 1981, Fig. 3) have
recently been confirmed by a Rb - Sr whole-rock isochron age (Beckinsale et al.
1985) and by additional K-Ar dating (Zimmermann and Kihien Collado 1983).
Figure 3 documents Paleozoic porphyry copper emplacement, mainly in
Argentina, during the Late Carboniferous-Early Permian and, on the basis of a
single dated occurrence (Taca Taca, 323 m.y.), the Early Carboniferous. A well-
defined Late Carboniferous-Early Permian sub-belt is depicted on the basis of
two dated occurrences and four additional occurrences (Cordillera de La Brea,
San Francisco de Los Andes, La Colorada at Yalguaraz, Santa Clara; Figs. 2
and 3) inferred to be of similar age. The two other radiometrically dated
occurrences of Late Carboniferous-Early Permian age, including Imilac which
yielded two imprecise Rb - Sr ages (Halpern 1978) not included in earlier
compilations, plot as isolated points north and south of the main sub-belt
(Fig. 3).
At the beginning of the Mesozoic, the locus of magmatism stepped abruptly
westwards to the vicinity of the present Pacific coastline. Meso-Cenozoic
porphyry copper emplacement occurred in at least three narrow sub-belts, which
migrated progressively eastwards with time, and appear not to have overlapped
one another (Fig. 3). In the Early Miocene, however, the locus of magmatism
expanded farther eastwards to form a wide band of dispersed volcanism and
stock emplacement, to which the Middle-Late Miocene porphyry copper sub-belt
is confined (Fig. 3). The Middle-Late Miocene sub-belt therefore overlaps much
of the Late Carboniferous-Early Permian sub-belt.
The well-documented Meso-Cenozoic sub-belts contain all the major
porphyry copper deposits in the central Andes. Preeminent is the Late Eocene-
Early Oligocene sub-belt, which hosts Quebrada Blanca, Chuquicamata, La
Escondida, EI Salvador and Potrerillos. Major deposits farther north, at Cerro
Verde (Santa Rosa), Quellaveco, Cuajone and Toquepala in Peru, and Cerro
Colorado in Chile make up the Paleocene sub-belt, which southwards, from
latitudes 26 to 32°S, contains several important groups of tourmaline breccia
pipes (Figs. 2 and 3). In the southern part of the central Andes, all major
deposits, at Los Pelambres, EI Pach6n, Rio Blanco, Disputada and EI Teniente,
Metallogenic Implications 241

as well as the EI Indio Au - Ag - Cu deposit, constitute the western edge of the

Middle-Late Miocene sub-belt. The concentration of major deposits of younger
age in the southern part of the central Andes is attributed in large measure to
rapid unroofing by glacial erosion south of latitude 300 S (Sillitoe 1981).
The porphyry copper sub-belts, including the Late Carboniferous-Early
Permian one, possess a marked parallelism with the axis of the Peru-Chile trench
(Fig. 3), a relationship which strongly implies the involvement of subduction-
related processes in porphyry copper genesis. It would appear, however, that the
Late Cretaceous sub-belt and the present trench axis diverge appreciably in the
southern part of the region under study.

Crustal Setting

The Meso-Cenozoic central Andes were formed on a continental, sialic base-

ment, which is known to include both Paleozoic magmatic and sedimentary
rocks generated at earlier subducting margins (e.g., Herve et al. 1981), and at
least two Precambrian complexes (Fig. 4). The Paleozoic basement includes the
Early Carboniferous and Late Carboniferous-Early Permian porphyry copper
mineralization described above.
Although Paleozoic rocks are widespread, especially in western Argentina,
the Precambrian is relatively restricted in outcrop. The oldest known Precam-
brian complex, which includes abundant granulite-facies metamorphic rocks, has
been dated by both Rb-Sr and U -Pb methods at nearly 2,000 m.y. (Cobbing et
al. 1977, Dalmayrac et al. 1977, Shackleton et al. 1979). Its outcrop is restricted
to the Arequipa massif of southwestern Peru, which may possibly extend south-
wards to include the Belen Schist Formation in northernmost Chile (Fig. 4), for
which Pacci et al. (1980) obtained a 1,OOO-m.y. Rb-Sr isochron. The eastward
and southward subsurface extensions of the Arequipa massif are poorly
constrained, although it should be pointed out that a late Precambrian Rb - Sr
age was obtained for granitic rock encountered during deep drilling at San
Andres in northwestern Bolivia (Fig. 4, Lehmann 1978). Although most inves-
tigators have concluded that the mid-Proterozoic rocks of the Arequipa massif
are autochthonous and constitute the western leading edge of the Brazilian Shield
(Cobbing et al. 1977, Shackleton et al. 1979), a slight possibility exists that they
are allochthonous in origin and constitute an exotic suspect terrane in the sense
of Coney et al. (1980). However, accretion could not have occurred any later
than the Silurian since Devonian strata overlap the outcropping edge of the
Arequipa massif.
Precambrian rocks are also known to underlie parts of the Sierras Pampeanas
massif in western Argentina. Rocks range in metamorphic grade from green-
schist to amphibolite facies, and normally yield late Precambrian radiometric
ages, although remnant nuclei of granulite-facies rocks as old as 1,800 m.y. have
been reported (Caminos et al. 1982).
Crustal thicknesses in the central Andes (Fig. 4) range from > 70 km beneath
the Altiplano of southern Peru and western Bolivia to 30 km along the Chilean
242 R. H. Sillitoe

I
65·

300

30·

,,'i Crustal thickness. in kms /

i
i (dashed ••
where extrapolated)
l
.i
)
Fig. 4. Outcropping Precambrian
I Porphyry Copper massifs (from Shackleton et al.
i deposit. •69 1979, Ramos and Pesce 1982),
i with Cu/ Mo ratio
CHI L E \ crustal thicknesses (mainly from
I
Arequipa Massif James 1971a but influenced south
i ., of lat. 27°S by Cummings and
(", Sierras Pampeanas =.. Shiller 1971), and Cu/Mo ratios
I Mossif '.... _ .... ") of principal porphyry copper de-
i
i posits (from Table 2) in the cen-
i 70. 65· tral Andes. B = Belen Schist For-
i
mation; S = San Andres granite
Metallogenic Implications 243

littoral (James 1971a, Cummings and Shiller 1971). Much of the increase of
crustal thickness to the extreme of 50 to 70 km in northern Chile-southern Peru-
western Bolivia is believed to have been a consequence of voluminous Meso-
Cenozoic magmatism (James 1971b).
In the central Andes of northern Chile-southern Peru-western Bolivia,
Miocene to Recent volcanic rocks have been shown to possess intial 87Sr/8t>sr
ratios >0.705. These relatively high initial ratios, combined with other isotopic
and trace-element criteria, are interpreted to signify that mantle-derived magmas
underwent small degrees of crustal contamination (Thorpe et al. 1981, James
1982). In southern Peru, at least part of the contaminant has been shown to be
2000-m.y.-old granulites in the upper crust (Tilton and Barreiro 1980, James
1982). The significantly lower «0.705) initial ratios for andesitic volcanic rocks
from central and southern Chile, where the continental crust is < 40 km thick,
are taken to indicate less crustal involvement in magma genesis, and to imply that
contamination farther north is due to passage of magmas through an unusually
thick (> 50 km) crust (Harmon et al. 1984).

Cu/Mo Ratios

It is difficult to obtain precise hypogene Cu/Mo ratios for much of the porphyry
copper mineralization in the central Andes (Fig. 2) because drilling has been in-
sufficient to obtain representative Cu and Mo grades. Satisfactory Cu/Mo ratios
are available for only 18 major deposits, induding all the operating mines (Table
2). The grade information is based on assay of 3-m lengths of split core obtained
from extensive programmes (generally > 30000 m) of diamond drilling con-
ducted by the mining companies concerned to determine ore reserves.
Although the data in Table 2 are taken to be reasonably reliable, say within
± 25070 of the true Cu/Mo ratios, two main problems are inherent in the
provision of average hypogene Cu and Mo values for entire porphyry copper
deposits.
First, porphyry copper deposits are commonly characterized by an inhomo-
geneous, sometimes zoned, distribution of metals. This is particularly the case
for molybdenum, which may occur as higher-grade concentrations: in a halo to a
copper-enriched core, as at Bajo de La Alumbrera (Sillitoe 1979) and Potrerillos
(H. Flores, personal communication, 1983); in hydrothermal breccia bodies, as
at Quebrada Blanca (Hunt et al. 1983), EI PachOn (A. N. Lencinas, personal
communication, 1983) and Disputada (Warnaars 1983); in halos to breccias at EI
Teniente (A. Enrione, personal communication, 1983) and Los Pelambres
(Atkinson et al. 1984); and associated with a phase of overprinted sericitic
alteration adjoining the West Fissure at Chuquicamata (Ambrus 1978).
However, although inhomogeneities in Mo distribution may be large, they are
effectively averaged out when metal values become available for entire deposits.
It may be seen from Table 2 that average Mo values range from 0.01 to 0.025
percent, except for Quellaveco and Bajo de La Alumbrera with averages of 0.008
and 0.007%, respectively.
244 R. H. Sillitoe

Table 2. Hypogene copper and molybdenum contents of selected central Andean porphyry copper
deposits 3

Deposit 070 Cu 070 Mo Cu/Mo

Santa Rosa 0.65 0.014 46

Quellaveco 0.7 0.008 88
Cuajone 0.7 0.0135 52
Toquepala 0.6 0.0135 44
Cerro Colorado 0.4 0.015 27
Quebrada Blanca 0.5 0.02 25
El Abra 0.8 0.015 53
Chuquicamata 0.9 0.02 45
La Escondida 0.5 0.015 33
El Salvador 0.55 0.025 22
Potrerillos 0.75 0.025 30
Andacollo 0.6 0.015 40
Los Pelambres 0.6 0.015 40
El Pach6n 0.4 0.016 25
Rio Blanco 1.0 0.017 59
Disputada 0.7 0.016 44
El Teniente 0.86 0.022 39
Bajo de La Alumbrera 0.49 0.007 70

3 All data from unpublished company sources, except for those from Andacollo (Llaumett et al.
1975) and Quebrada Blanca (Hunt et al. 1983).

Second, and more problematic, is the effect of supergene copper enrichment,

which modified most central Andean porphyry copper deposits during the
Cenozoic. In deposits such as Cuajone, Toquepala, Cerro Colorado, Chuqui-
camata, La Escondida, Quebrada Blanca, El Salvador and Potrerillos, charac-
terized by high-grade supergene enrichment, average hypogene grades, com-
monly less well constrained by drilling, were estimated for depths below the
enrichment blankets. Where supergene enrichment is more localized, as at Los
Pelambres, El Pach6n, Disputada and El Teniente, an attempt was made to sub-
tract the supergene contribution from the overall average copper grades. Copper
values are therefore of variable accuracy, with deposits such as El Abra, Rio
Blanco and Bajo de La Alumbrera, which underwent essentially no supergene
enrichment, providing the most reliable figures. Hypogene copper contents vary
from a low of 0.4"70 to a high of 1.0% (Table 2).
From Table 2 and Fig. 4 it may be appreciated that all Cu/Mo ratios fall in
the range from 22 to 88. Based on Table 2, the average Cu/Mo ratio for the 18
central Andean porphyry copper deposits is 43.4. Bajo de La Alumbrera, the
deposit with the second highest Cu/Mo ratio (70), is the only central Andean
porphyry copper deposit with a high gold content. Its average gold grade of
0.62 ppm (Sillitoe 1979) is more than 2.5 times that of Potrerillos and Andacollo,
and at least 5 times that of all other deposits listed in Table 2.
Metallogenic Implications 245

Metallogenic Implications

Cu/Mo ratios are not available for any of the Paleozoic or Late Cretaceous por-
phyry-type occurrences. Nevertheless, it may be reasonably concluded from Figs.
3 and 4 that the Cu/Mo ratios of central Andean porphyry copper deposits are
unrelated to the age of emplacement (Fig. 5), their position relative to the Pacific
littoral (and the Peru-Chile trench), and their latitudinal position within
individual sub-belts. Since systematic variations in crustal constitution are to be
expected both along and across Andean-type margins, and during their evolu-
tion, Fig. 4 implies that the Cu/Mo ratios are probably independent of the
central Andean continental crust. The possibility that Fig. 4 reflects a random
spatial variation of one or more critical crustal parameters is reduced by dif-
ferences in Cu/Mo ratios of as great as 44 between closely spaced deposits in
southern Peru.
Furthermore, reference to Figs. 4 and 6 reveals that Cu/Mo ratios appear to
bear no relation to crustal thickness or, as far as can be determined, to the distri-
bution of outcropping Precambrian crust. Deposits in southern Peru and
northern Chile with Cu/Mo ratios from 22 to 88 (average 42.2 for 11 deposits)
are underlain by crust from 50 to 60 km thick, whereas those in central Chile and

90
•
90
0

80 80

70
• 70
•
60
• 60
•
0
+=
E! 50
• • e
.2
50
o·
~ • • •• • 0
•
•
0 0

:'40
• • • ~40 •
u u
~

• • •
30
• 30
•
• • • • •
20 • 20
30
•
0 10 20 30 50 60 70 40 50 60 70

Age. m.y. Approx. crustal thickness

Fig. 5 Fig. 6 (km)

Fig. 5. Plot of Cu/Mo ratio (from Table 2) vs K - Ar age (from Fig. 3) for selected porphyry copper
deposits in the central Andes. Lack of any correlation is apparent. Age of Cerro Colorado assumed
same as nearby Mocha deposit

Fig. 6. Plot of Cu/Mo ratio (from Table 2) vs approximate crustal thickness (estimated from Fig. 4)
for selected porphyry copper deposits in the central Andes. No correlation is apparent. No Cu/Mo
ratios available for deposits underlain by crust from 40 to 50 km thick. Deposits emplaced in
Arequipa massif denoted by open circles
246 R. H. Sillitoe

westermost Argentina with Cu/Mo ratios from 25 to 59 (avergae 41 for 6

deposits) are underlain by crust <40 km thick. It is particularly noteworthy that
deposits confined to the 2000-m.y.-old Arequipa massif, where the crustal thick-
ness is now (but perhaps not during Paleocene porphyry copper emplacement)
some 50 km thick, are not significantly more molybdenum-rich than those
generated farther south, where known 2000-m.y.-old rocks are lacking. Further-
more, Bajo de La Alumbrera, underlain by the Precambrian Sierras Pampeanas
massif (Fig. 4), is one of the most molybdenum-poor deposits in the central
Andes. The emplacement of porphyry copper deposits (with similar Cu/Mo
ratios) into granulite-facies upper crust in southern Peru and into essentially
unmetamorphosed volcanic and sedimentary sequences elsewhere in the central
Andes fails to support the general applicability of Plant et al. 's (1983) hypothesis
that porphyry copper generation is inhibited by granulite-facies (anhydrous)
metamorphic host rocks.
It should also be stressed that Cu/Mo ratios for Cenozoic deposits emplaced
in northern Chile and southern Peru through thickened crust, as denoted by
Miocene to Recent volcanic rocks with initial 87Srl 8~r > 0.705, are not systema-
tically higher than those for deposits in central Chile, where volcanic rocks
possess initial Sr-isotopic ratios <0.705.
It is concluded that there is no discernable correlation between the Cu/Mo
ratios of central Andean porphyry copper deposits and the thickness, age,
lithology, or other observable or measurable characteristics of the continental
crust into which they were emplaced. This conclusion militates against a crustal
source for Cu and Mo, and possibly other elements, in central Andean porphyry
copper deposits. The data are consistent with derivation of Cu and Mo from a
subcrustal source region. However, since crustal parameters, apart from
thickness, pertain only to upper crustal levels, the lower continental crust cannot
be discounted as a metal source. Observed variations in Cu/Mo ratios (Table 2)
are perhaps attributable to detailed differences in the conditions of magma
generation and evolution, given broadly similar starting products.

Discussion

Karig and Kay (1981) recognized a correlation between 87Srl 8~r and 206pb/ 204pb
ratios for pelagic sediments and island-arc volcanic rocks. They concluded,
contrary to many previous workers, that pelagic sediments overlying the oceanic
plate are selectively subducted, and contribute to arc magmatism. Although the
bulk of calc-alkaline magma is considered to be generated by partial melting of
the mantle wedge above the subduction zone (Fig. 1), a substantial sediment flux
is required to account for the contents of K, Rb, Sr, Pb and REE in arc magmas.
Combined isotopic and trace-element data for arc magmas from many parts of
the world also strongly support a significant contribution from the downgoing
slab (e.g., Hawkesworth 1982).
If it is accepted that the metal contents of porphyry copper deposits in calc-
alkaline magmatic arcs, such as the central Andes, are largely independent of
Metallogenic Implications 247

crustal parameters, then an adequate subcrustal source could be provided by sub-

ducted pelagic sediments (cf., Sillitoe 1981). Representative layer-1 sediments on
the Nazca plate, west of the central Andes, contain up to 1000 ppm Cu and 300
ppm Mo (Field et al. 1983). Although subducted oceanic basalts and the perido-
titic mantle wedge, both characterized by much lower Cu and Mo contents than
pelagic sediments, cannot be ruled out as sources for these two metals, they do
not appear to supply much of the trace Pb found in porphyry copper deposits
(Sillitoe and Hart 1984). The lead-isotopic signatures of porphyry copper miner-
alization from the northern Andes of Colombia (Sillitoe and Hart 1984) fall in
broadly the same isotopic fields as those determined for Disputada (Tilton 1979),
EI Salvador and Toquepala (Tilton et al. 1981) in the central Andes, and are not
incompatible with a pelagic-sediment source. However, since pelagic-sediment
and continental-crustal leads are difficult to distinguish, Pb-isotopic ratios do
not provide a discriminant between crustal and subcrustal Pb sources (cf.,
McNutt et al. 1979), except under special circumstances (Sillitoe and Hart 1984).
It was noted above that small amounts of 87Sr, 207Pb, 180 and other isotopes
may have been supplied as contaminants to subduction-related magmas that
ascended through continental crust > 50 km thick in northern Chile, southern
Peru and adjacent areas (Thorpe et al. 1981, James 1982, Harmon et al. 1984).
However, this analysis has revealed no compelling reason to suspect a significant
contribution of Cu or Mo from this exceptional thickness of crustal rocks,
although it should be pointed out that the continental crust of the Altiplano
region may well have been substantially thinner during Paleocene porphyry
copper emplacement in southern Peru. However, as emphasized by Cobbing and
Pitcher (1983), crustal thickening was probably essentially complete prior to
Eocene magmatism, so that the Late Eocene-Early Oligocene porphyry copper
deposits in northern Chile were emplaced through the thickened crust.
Available 87Sr/81lsr initial ratios for most central Andean porphyry copper
deposits (Table 3), including Toquepala, EI Salvador and Disputada, all approxi-
mate 0.704, thereby confirming that Sr, although known to be a possible crustal
contaminant, is entirely of sub crustal origin. It is therefore extremely unlikely
that eu and Mo in these deposits possess anything other than a subcrustal source.
Andesitic volcanic rocks cogenetic with porphyry copper emplacement at
Bajo de La Alumbrera, northwestern Argentina (Figs. 2 and 3), have yielded

Table 3. Available 87Sr/86Sr initial ratios for central Andean porphyry copper deposits

Deposit 87Sr/86Sr Reference

(mean value)

Toquepala 0.7044 Beckinsale et at. (1985)

EI Salvador 0.7041 Gustafson and Hunt (1975)
0.7033 Halpern (1978)
Disputada 0.7041 Halpern (1979)
Bajo de La Alumbrera 0.7068 a McNutt et at. (1975)

a Determined on comagmatic volcanic rocks only.

248 R. H. Sillitoe

87Srl 8~r initial ratios of 0.706 and 0.7075 (McNutt et al. 1975, Table 3), which
may be taken (as they were by Tilton et al. 1981) to suggest an addition of radio-
genic Sr from subjacent sialic crust, although this is <40 km thick (Fig. 4).
Contrary to expectation, however, the Bajo de La Alumbrera deposit possesses
the highest Au content of any porphyry deposit in the central Andes (see above),
and the second highest Cu/Mo ratio in Table 2, a combination of features that
renders it the least likely candidate to have gained a significant metal contribu-
tion from sialic crust. Bajo de La Alumbrera therefore demonstrates that a
porphyry copper metal budget can lack the generally accepted signs (high
Mo/low Au) of an upper crustal source even where sialic crust appears to have
supplied an 87Sr contaminant. A similar conclusion was reached by Damon et al.
(1983) for porphyry Cu - Mo deposits throughout Mexico, where those emplaced
through Precambrian sialic basement gave a mean 87SrI 8~r initial ratio of
0.7075, significantly higher than a mean value of 0.7048 for those farther south
in accreted terranes lacking sialic basement.

Acknowledgments. The assistance of Sergio Barassi, Jose Cabello, Mario Chabert, Hector Flores,
Andres Lencinas, Joy Merz, Dorian Nicol and Enrique Tidy in the collection of Cu and Mo grades
for central Andean porphyry copper deposits in gratefully acknowledged. Comments on the
manuscript were provided by Nigel Grant and Richard Thorpe.

References

Alfaro HG (1980) Prospecci6n de p6rfidos cupriferos en la zona central-sur de Chile. In: Mineria de
cobres porfidicos, vol II. Anales del Congreso Cincuentenario. Ins Ing Min Chile, Santiago, pp
295 - 313
Ambrus J (1978) Chuquicamata deposit. In: Sutulov A (ed) International molybdenum encyclopaedia
1778 -1978, vol I. Intermet, Santiago, pp 87 - 93
Araneda GR (1982) El Indio, yacimiento de oro, plata y cobre. Coquimbo, Chile. Minerales
(Santiago) 37 (160):5-13
Atkinson WW Jr, Souviron A, Vehrs TI (1984) Geological results of recent exploration of the
porphyry copper deposit at Los Pelambres, Chile. Geol Soc Am (Abstracts with Programs) 16:434
Beckinsale RD, Sanchez-Fernandez AW, Brook M, Cobbing EJ, Taylor WP, Moore ND (1985)
Rb - Sr whole-rock isochron and K - Ar age determinations for the Coastal Batholith of Peru. In:
Pitcher WS, Atherton MP, Cobbing EJ, Beckinsale RD (eds) Magmatism at a plate edge: The
Coastal Batholith of Peru. Blackie, London, pp 177 - 202
Burnham CW (1981) Convergence and mineralization - Is there a relation? Geol Soc Am Mem
154:761-768
Caminos R, Cingolani CA, Herve F, Linares E (1982) Geochronology of the pre-Andean meta-
morphism and magmatism in the Andean Cordillera between latitudes 30° and 36°S. Earth Sci Rev
18:333 - 352
Cobbing EJ, Pitcher WS (1983) Andean plutonism in Peru and its relationship to volcanism and
metallogenesis at a segmented plate edge. Geol Soc Am Mem 159:277 - 291
Cobbing EJ, Ozzard JM, Snelling NJ (1977) Reconnaissance geochronology of the crystalline
basement rocks of the Coastal Cordillera of southern Peru. Geol Soc Am Bull 88:241 - 246
Coney PJ, Jones DL, Monger JWH (1980) Cordilleran suspect terranes. Nature (London)
288:329 - 333
Cummings D, Shiller GI (1971) Isopach map of the earth's crust. Earth Sci Rev 7:97 -125
Dalmayrac D, Lancelot JR, Leyreloup A (1977) Two-billion-year granulites in the late Precambrian
metamorphic basement along the southern Peruvian coast. Science 198:49-51
Metallogenic Implications 249

Damon PE, Shafiqullah M, Clark KF (1983) Geochronology of the porphyry copper deposits and
related mineralization of Mexico. Can J Earth Sci 20:1052-1071
Field CW, Rye RO, Dymond JR, Whelan JF, Senechal G (1983) Metalliferous sediments of the East
Pacific. In: Shanks WC III (ed) Cameron volume on unconventional mineral deposits. Am Inst
Min Metall Petrol Eng, New York, pp 133-156
Griffiths JR, Godwin CI (1983) Metallogeny and tectonics of porphyry copper-molybdenum deposits
in British Columbia. Can J Earth Sci 20:1000-1018
Gustafson LB, Hunt JP (1975) The porphyry copper deposit at El Salvador, Chile. Econ Geol
70:857 -912
Halpern M (1978) Geological significance of Rb - Sr isotopic data of northern Chile crystalline rocks
of the Andean orogen between latitudes 23° and 27°South. Geol Soc Am Bull 89:522 - 532
Halpern M (1979) Strontium isotope composition of rocks from the Disputada copper mine, Chile.
Econ GeoI74:129-130
Harmon RS, Barreiro BA, Moorbath S, Hoefs J, Francis PW, Thorpe RS, Deruelle B, McHugh J,
Viglino JA (1984) Regional 0-, Sr-, and Pb-isotope relationships in late Cenozoic calc-alkaline
lavas of the Andean Cordillera. J Geol Soc (London) 141:803 - 822
Hawkesworth CJ (1982) Isotope characteristics of magmas erupted along destructive plate margins.
In: Thorpe RS (ed) Andesites: orogenic andesites and related rocks. John Wiley & Sons, New
York, pp 549 - 571
Herve F, Davidson J, Godoy E, Mpodozis C, Covaccevich V (1981) The late Paleozoic in Chile:
Stratigraphy, structure and possible tectonic framework. An Acad Bras Cienc 53:361 - 373
Hollister VF (1975) An appraisal of the nature and source of porphyry copper deposits. Min Sci Eng
7:225 -233
Hunt JP, Bratt JA, Marquardt LJC (1983) Quebrada Blanca, Chile: An enriched porphyry copper
deposit. Min Eng 35 :636 - 644
James DE (1971a) Andean crustal and upper mantle structure. J Geophys Res 76:3246-3271
James DE (1971b) Plate tectonic model for the evolution of the central Andes. Geol Soc Am Bull
82:3325 - 3346
James DE (1982) A combined 0, Sr, Nd and Pb isotopic and trace element study of crustal con-
tamination in central Andean lavas, I. Local geochemical variations. Earth Planet Sci Lett
57:47-62
Jensen ML (1971) Provenance of Cordilleran intrusives and associated metals. Econ GeoI66:34-42
Karig DE, Kay RW (1981) Fate of sediments on the descending plate at convergent margins. Philos
Trans R Soc London Ser A 301 :233 - 251
Laznicka P (1976) Porphyry copper and molybdenum deposits of the USSR and their plate tectonic
settings. Inst Min Metall Trans B 85:B14-32
Lehmann B (1978) A Precambrian core sample from the Altiplano/Bolivia. Geol Rundsch
67:270-278
Llaumett C et al. (1975) El yacimiento cobre porfidico "Andacollo", Provincia de Coquimbo, Chile.
Rev Geol Chile 2:56 - 66
Lowell JD (1974) Regional characteristics of porphyry copper deposits of the Southwest. Econ Geol
69:601-617
Lowell JD (1980) Metallogenesis and porphyry deposits of North America and the Pacific region. In:
Mineral resources. Bur Rech Geol Min Mem 106:40-47
McNutt RH, Caelles JC, Farrar E, Haynes SJ, Zentilli M (1975) Initial 87Sr/86Sr ratios of plutonic
and volcanic rocks of the central Andes between latitudes 26° and 29°S. Earth Planet Sci Lett
27:305 - 313
McNutt RH, Clark AH, Zentilli M (1979) Lead isotopic compositions of Andean igneous rocks,
latitudes 26° to 29°S: Petrologic and metallogenic implications. Econ Geol 74:827 - 837
Oyarzun MJ, Frutos J (1980) Metallogenesis and porphyry deposits of the Andes (southeastern
Pacific region). In: Mineral resources. Bur Rech Geol Min Mem 106:50- 62
Pacci D, Munizaga F, Herve F, Kawashita K, Cordani UG (1980) Acerca de la edad Rb - Sr
Preciimbrica de rocas de la Formaci6n Esquistos de Bei{m, Depto. de Parinacota, Chile. Rev Geol
Chile 11 :43 - 50
Plant JA, Simpson PR, Green PM, Watson JV, Fowler MB (1983) Metalliferous and mineralized
Caledonian granites in relation to regional metamorphism and fracture systems in northern
Scotland. Inst Min Metall Trans B 92:B33 - 42
250 R. H. Sillitoe: Metallogenic Implications

Ramos A, Pesce AH, and collaborators (1982) Mapa geologico de la Republica Argentina,
1: 2500000. Subsecretaria de Mineria, Buenos Aires
Shackleton RM, Ries AC, Coward MP, Cobbold PR (1979) Structure, metamorphism and geo-
chronology of the Arequipa massif of coastal Peru. J Geol Soc (London) 136:195-214
Sillitoe RH (1972) A plate tectonic model for the origin of porphyry copper deposits. Econ Geol
67:184-197
Sillitoe RH (1979) Some thoughts on gold-rich porphyry copper deposits. Miner Deposita
14:161-174
Sillitoe RH (1981) Regional aspects of the Andean porphyry copper belt in Chile and Argentina. Inst
Min Metall Trans B 90:B15-36
Sillitoe RH, Hart SR (1984) Lead-isotopic signatures of porphyry copper mineralization in oceanic
and continental settings, Colombian Andes. Geochim Cosmochim Acta 48:2135 - 2142
Thorpe RS, Francis PW, Harmon RS (1981) Andean andesites and crustal growth. Philos Trans R
Soc London Ser A 301 :305 - 320
Tilton GR (1979) Isotopic studies of Cenozoic Andean calc-alkaline rocks. Carnegie Inst Washington
Yearb 78:298-304
Tilton GR, Barreiro BA (1980) Origin of lead in Andean calc-alkaline lavas, southern Peru. Science
210:1245 -1247
Tilton GR, Pollak RJ, Clark AH, Robertson RCR (1981) Isotopic composition of Pb in central
Andean ore deposits. Geol Soc Am Mem 154:791- 816
Titley SR, Beane RE (1981) Porphyry copper deposits. Part I. Geologic settings, petrology, and
tectogenesis. Econ Geol 75th Anniversary Vol, pp 214-235
Warnaars FW (1983) Copper tourmaline breccias at Los Bronces, Chile. Soc Min Eng AIME Trans
272:1902-1911
Westra G, Keith SB (1981) Classification and genesis of stockwork molybdenum deposits. Econ Geol
76:844-873
Zimmermann JL, Kihien Collado A (1983) Determination par la methode KI Ar de I'iige des
intrusions et des mineralisations associees dans Ie porphyre cuprifere de Quellaveco (sud ouest du
Perou). Miner Deposita 18:207 - 213
Metallogenic Zoning of Volcano-Plutonic Belts and
Porphyry-Copper Mineralization
V. A. EVSTRAKHIN, A. I. KRIVTSOV, and I. F. MIGACHYOV 1

Abstract

Volcano-plutonic belts (VPB) that host porphyry-copper deposits are grouped

into basaltic and andesitic types. Basaltic (eugeosynclinal) VPB give rise to
island-arc barrier zones (fore-arc). Andesitic VPB originate in an orogenic-
activation regime and, depending on the character of the basement, are sub-
divided into epicratonic, epimiogeosynclinal, and epieugeosynclinal types. Ore
genesis products within the VPB are consistent with vertical and lateral-temporal
metallogenic zoning (MZ). In basaltic VPB, copper and gold porphyry-copper
deposits are associated with copper massive sulfide ores and are replaced in time
by veined polymetallic, and in places, skarn mineralization. MZ of andesitic VPB
is controlled by metallogeny of their basement combined with molybdenum-,
copper-molybdenum porphyry, and molybdenum-copper deposits (depending on
the substratum) as well as with later porphyry Sn, Sn - W, W - Mo, W, Au - Ag
ores. The metallogenic style of the VPB is determined by the formation of
porphyry deposits. The metallogenic zoning of the VPB is exemplified in some
provinces of the Soviet Union.
The formational approach to a specific metallogeny that is much favored now
implies that mineral deposits are natural components of certain geologic forma-
tions. Appraising and correlating the existing metallogenic concepts, Smirnov
(1981) clearly gives preference to the formational approach, pointing out not
only its prospects, but its appreciable practical significance as well. Recognition
of ore-bearing geologic formations and their geotectonic and paleotectonic
settings, serves as a basis for metallogenic models, that, in the long run, enable us
to define the position of the mineralization in the spatial and temporal series of
the products of the tectono-magmatic cycle and to outline boundaries of metal-
logenic areas.
Applied to porphyry-copper deposits, the approach was extensively employed
by such authors as Vlasov (1978), Krivtsov (1977, 1983), Pavlova (1978), Popov
(1977), and others. It was clearly demonstrated that porphyry-copper deposits
are related to specific geostructures: volcano-plutonic belts known in geosyn-
clinal folded systems of different ages (from Phanerozoic to Recent). It is
obvious that regimes that are responsible for the origin of the volcano-plutonic

1 Ministry of Geology, Moscow, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
252 V. A. Evstrakhin et al.

belts do not belong exclusively to the Kimmerian or Alpine epochs, but manifest
themselves in much earlier periods of the Earth's crust development.
At the same time, American and English authors, using the Laramide and
Pacific provinces as examples, look upon the geotectonic position of porphyry-
copper deposits from the plate-tectonic viewpoint. Spatial and temporal siting of
such deposits are analyzed, based on the position and tilting of the associated
subduction zones, the interposition of plates, and the speed and directions of
their displacements. Most models are based on the reconstruction of a subduc-
tion paleozone setting, while the formational analysis of the environments of ore
origin, which are available for direct observations, are neglected.
Independent of the time of formation, volcano-plutonic belts (VPB) hosting
porphyry-copper mineralization are looked upon as specific geostructures that
occupy certain lateral-temporal positions in conjugate geotectonic elements
(Krivtsov 1977, 1983). We subdivide VPB into the basaltic (island-arc) type that
terminates preorogenic evolution of eugeosynclines, and the andesitic type
forming in orogenic-activation regimes at the flanks of contemporaneous geo-
synclinal systems (Krivtsov 1977, 1983, Krivtsov and Ageyeva 1979).
According to Vlasov (1978) belts of both types are conjugate with other geo-
tectonic elements that differ in their regimes. The East Asian type of orthogeo-
synclines, in particular, consist of the following lateral elements: ocean, trench,
Benioff zone, barrier zone of the island-arc (fore-arc, on the ocean side) basaltic
VPB, inner zone of the island-arc with massive sulfide formations, median
massif, back-arc basin with a miogeosynclinal regime, epicontinental andesitic
VPB, and continent with local troughs with terrigenous filling (see Fig. 1). It
should be noted that rear troughs, in case they acquire specific features of the
secondary (regenerated) geosynclines, hamper interpretations of such systems.
Being different in regime, the aforesaid geostructures are conjugate in time
and share a common genesis that is usually interpreted through a single subduc-
tion zone associated with it. On the other hand, continent-ocean migration of
such series have been established for the Cimmerides and Alpinedes of the
Eastern Soviet Union, the Hercynides of the Urals, and the Hercynides and
Alpinedes of the Caucasus. These relationships make it possible to superimpose
andesitic VPB upon any of the previously evolved elements of such geostruc-
tures. Accordingly, depending on the composition of the substratum, we group
the andesitic VPB into epicratonic, epimiogeosynclinal, and epieugeosynclinal
ones (Krivtsov 1977, 1983).

I liT
1-

Fig. 1. Geotectonic elements of the East Asian type of orthogeosynclinal systems. Trench (I), Benioff
zone (II), Barrier zone (fore-arc, basaltic VPB) of island-arc (III), inner zone of island-arc (IV);
back-arc basin (miogeosynclinal) (V); median mass (VI); epicontinental andesitic VPB (VII),
continental local trough (VIII). Continent is shown by diagonal lines
Metallogenic Zoning of Volcano-Plutonic Belts and Porphyry-Copper Mineralization 253

Intracontinental orthogeosynclinal systems, well-studied through the paleo-

zoids of Kazakhstan, also involve some conjugate geostructures with a different
evolution regime. As shown in Fig. 2, within the paleozoids, axial-rift troughs ex-
posing oceanic crust, are fringed on one or two sides by the secondary geosyn-
clines produced as a result of the expansion of the geosynclinal regime onto the
sialic frame. Being most remote from the rifting axis, andesitic VPB are super-
imposed both on the secondary-geosynclinal complexes, and on their old base-
ment. In the general scheme of the system (orthogeosynclinal, secondary geosyn-
cline, andesitic VPB, foredeep, continent), there is lack of the basaltic VPB
(though they are not prohibited) . In the above scheme series, each succeeding
element is somewhat younger than the previous one and is partially overlapped in
space. The latter fact is not characteristic of marginal-continental systems.
Nevertheless, intracontinental andesitic VPB also is subdivided into these three
above mentioned types.
The differences in the positions of the basaltic and the andesitic VPB, as well
as the nature of their basements, affect significantly not only the geochemistry of
the associated porphyry-copper deposits, but also the general metallogeny of the
VPB.
As shown in Fig. 3, the basaltic VPB that evolved on the oceanic crust and
was composed of undifferentiated mafic lavas are relatively poor in ore-genesis

Fig. 2. Geotectonic elements of intracontinental orthogeosynclinal systems. Primary geosyncline (rift

trough) (I); secondary geosyncline (/I); andesitic VPB (III); foredeep (IV). Continent is shown by
diagonal lines

Fig. 3. Basic paleo-reconstructed section of the southern-central Urals. 1 Precambrian rocks; 2 Pre-
Middle Ordovician sedimentary and volcanogenic-sedimentary sequences; 3 basaltic rocks of the
undifferentiated formation associated with coaly-shaly and silicious rocks; 4 coaly-shaly and siliceous
rocks; 5 volcanics of contrasting and successive massive sulfide-bearing formations; 6 volcanics of the
basalt-andesite-basalt formation of basaltic belts; 7 sediments including flyschoids and turbidites;
8 deposits; a zinc-copper massive sulfide; biron-manganese; c porphyry copper; d veined
polymetallic
254 V. A. Evstrakhin et aI.

products. Porphyry-copper and gold-porphyry-copper deposits proper (Titley

1975), related to gabbro-diorite formations, are predominant in such belts.
Previously evolved massive-sulfide ores are localized in differentiated formations
of the inner zones of the island-arc outside the basaltic VPB-hosting porphyry-
copper ores. Younger ore-genesis products within the VPB consist of veined
polymetallic mineralization that locally can be related to the periphery of the
porphyry-copper systems. The monotony of the metallogeny and simplicity of
the basaltic VPB can apparently be attributed to the primitive composition of
their substratum that undoubtedly supplies some amount of the ore-forming con-
stituents. The fact that basaltic VPB's are incorporated into a lateral-temporal
series of geostructures makes it possible to explain the confining at such deposits
to structures similar in the nature of zoning.
Andesitic VPB that evolve on a basement varied in character and composition
are distinguished both in the geochemistry of porphyry-copper deposits and the
diversity of other ore-genesis products. Porphyry-molybdenum deposits are
common to the initial stages of epicratonic VPB; porphyry-copper-molybdenum
deposits are typical of epimiogeosynclinal VPB; and porphyry molybdenum
copper ones, of epieugeosynclinal. These deposits are related to granitoid,
monzonitoid, and granodiorite plutonogenic formations, respectively.
Porphyry-molybdenum-copper and porphyry-copper deposits of andesitic
belts are incorporated into a temporal series of metallogenic zoning of two ranks.
The first rank zoning, which has predominant lateral expression, is attributed to
the andesitic VPB that is conjugate with eugeosynclines and pericratonal basins.
In the eugeosynclines, preorogenic metallogeny with predominant massive-
sulfide ores is classical. In the second case, stratiform polymetallic deposits can
evolve. The second-rank zoning that manifests itself in VPB structures and has
mainly vertical expression is associated with deposits attributed to different ages,
viz., from ore occurrences localized at the VPB basement to deposits formed at
the late stages of VPB evolution (see Fig. 4).
Metallogeny of the VPB basement is determined by the style of its evolution.
The sialic substratum of epicratonic VPB is typically poor in minerals. Charac-
teristic granitophilic elements are inherited by metallogenic processes at the
period of VPB emplacement. Predominantly terrigenous, carbonate-terrigenous,
and carbonate complexes of the basement of epimiogeosynclinal VPB are not
infrequently hosts to stratiform polymetallic deposits. Being superimposed by
porphyry-copper systems, the latter undergo skarnization and regeneration as a
result of which veined deposits, conjugate with stratiform ones, come into being.
Volcanogenic formations of the basement of epieugeosynclinal VPB are charac-
terized by preorogenic metallogeny with typical massive-sulfide, iron, and
manganese volcanogenic ores. So far only some examples can be given to illus-
trate the involvement of this metallogenic material into the processes of forma-
tion of porphyry-copper systems, though skarn magnetite ores are known from
the periphery of porphyry-copper systems in VPB of various ages.
Later periods of VPB emplacement (after porphyry-copper systems have
formed) produced ore-genesis material ranging greatly in its composition (see
Fig. 4). All the deposits occur within the VPB, yet it has been so far difficult to
admit the possibility of the generation of vertical series in some parts of the VPB.
Metallogenic Zoning of Volcano-Plutonic Belts and Porphyry-Copper Mineralization 255

r----'

~~ J~ /~ / *0 "-..
III "---
.6. \0~ /
/ *
\~
/
T

II V
V QI
(y V ~V ~
~ ~ <JI 'if
T~\~ ~
-17~. ~
~~

I~ [3 ·
· . . .
· . . .
=

D • L

1--11 [ Zb 1I13[ZJ4 [~J.,. I ~ 16 CZI17

.8 -9 = fQ QH ~f2 0 13 OM +15

1/5 *f7 TI8 ~/9 .6.20 6 21 °22 029 024

Fig. 4. The structure and metallogenic zoning of andesitic volcano-plutonic belts (VPB).
I Substratum of the belts: 1rigid sialic (cratonic); 2 miogeosynclinal; 3 eugeosynclinal; 4 - 7volcano-
plutonic associations (VPA) of VPB. II Favorable for porphyry-copper mineralization and preceding
VP A: 4 epicratonic zones; 5 epimiogeosynclinal zones; 6 epieugeosynclinal zones; 7 late VP A III.
8 - 23 Deposits; 8 - 10 stratiform deposits of belts' substratum; 8 zinc-copper massive-sulfide; 9 iron
ore; 10 lead-zinc. VPB Deposits; 11 porphyry molybdenum (with copper); 12 porphyry-copper-
molybdenum; 13 porphyry-molybdenum-copper; 14 skarn lead-zinc; 15 iron-copper skarn; 16 veined
polymetallic; 17 gold-silver and gold-bearing; 18 tin; 19 tin-tungsten; 20 molybdenum-tungsten;
21 tungsten; 22 molybdenum-uranium; 23 niobium-tantalum; 24 deposits (of the system) with
porphyry appearance

Thus, the following geologic-economic series of deposits are recognized in

basaltic volcano-plutonic belts: copper-zinc, massive-sulfide, iron-manganese-
porphyry copper + veined polymetalllic. In andesitic belts, these are as follow:
massive-sulfide, base-metal-lead zinc stratiform, iron ore, porphyry-copper
+ skarn copper iron ore, gold, silver, tin + tin-tungsten + molybdenum-tungsten
+ tungsten, niobium-tantalum.
There foregoing lateral-temporal, vertical-temporal, and lateral-vertical-tem-
poral series of metallogenic zoning are available for observation, as a rule, in
lateral expression. This seems to be borne out of the fact that the vertical exten-
sion of zoning as well as the real depth of exposure of a deposit are not compar-
able in values. The general characteristics of the VPB metallogenic zoning is
exemplified by some provinces of the Soviet Union.
The products of early and late geosynclinal ore genesis in the region of the
Urals are incorporated into various elements of eugeosynclinal space, which rules
out their spatial conjunction (see Fig. 3). Basaltic belts in the Magnitogorsk intra-
continental eugeosyncline give rise to extended volcanogenic island-arcs of
barrier zones and fringe inner areas of contrasting and successive massive-sulfide
ore-bearing formations . Basaltic rocks of the undifferentiated formation com-
256 V. A. Evstrakhin et aJ.

monly make up the basement of the belts. Porphyry-copper mineralization here

is preceded by the formation of zinc-copper massive-sulfide deposits associated
with earlier eugeosynclinal formations. Porphyry-copper mineralization within
the studied areas of the Urals is represented by the porphyry-copper (without
molybdenum) type proper. The productive volcano-plutonic association involves
essentially sodium volcanogenic basalt-andesite-basalt and plutonogenic gabbro-
diorite-plagiogranite formations. Intrusive bodies are controlled by ruptures
bounding barrier zones, paleovolcanic vents, and these apparently belong to
"subvolcanic intrusions". Porphyry-copper deposits are related to quartz-diorite
and diorite-porphyry massifs. Ore mineralization occurs in apical, endo- and
exocontact parts of the such intrusions, viz., in albitization, silicification, chlorit-
ization, and sericitization zones of rocks regionally metamorphosed in prehnite-
pumpellyite facies. Iron-manganese deposits of the Near-Magnitogorsk group
are commonly associated with the terminations of massive-sulfide formations.
However, they can occur in the basalt-andesite-basalt formation of basaltic belts.
The formation is known with host-veined polymetallic mineralization traced in
the Irendyk and Imennovsk zones of the Urals. These areas are consistent only in
the gabbro-diorite quartz-diorite formation that hosts porphyry-copper mineral-
ization.
Metallogenic zoning of intracontinental andesitic volcano-plutonic belts
conjugate with geosynclines can be exemplified by the Late Paleozoic belt of the
northern Near-Balkhash region of Kazakhstan. As shown in Fig. 5, from south-
west and northeast, it fringes the northern Balkhash geosyncline that evolved
into a eugeosyncline in Cambrian - Late Ordovician time and into miogeosyn-
cline, in Late Ordovician - Early Famennian time. In the Famennian-
Tournasian along the periphery of the geosyncline, secondary geosynclines
evolved. Temporally (after the Visean) and laterally (southwest and northeast)
the secondary geosynclines give way to a volcano-plutonic belt. Depressions with
terrigenous sedimentation evolved along the periphery of the belt.
The substratum of the belt is known to host Early Paleozoic massive-copper
sulfide ore occurrences. Middle-Late Devonian massive-sulfide - base metal and
Later (Famennian-Tournasian) lead-zinc stratiform mineralization is related to
Famennian-Tournasian secondary geosynclines. Gold-silver mineralization
confined to the latest dacite-liparite components of effusive rock sequences is
related to the volcanic rocks favorable for porphyry-copper mineralization.
Copper-magnetite ore-bearing skarn bodies are traced at the contact of ore-
bearing intrusions hosting porphyry-copper mineralization. Molybdenum-
tungsten and niobium-tantalum mineralization is associated with plutonic rocks
of later, Permian dacite-liparite, and trachidacite-trachiliparite volcano-plutonic
association. Porphyry-copper mineralization of the region is represented by three
groups of deposits related to epicratonic (essentially molybdenum deposits of the
Karatasak area), epimiogeosynclinal (the Borlinsk area), and epieugeosynclinal
(molybdenum-copper deposits of the Kounrad area) environments, respectively.
It is important to emphasize some conservatism in the succession of metal-
logenic processes.
In eu- and miogeosynclines, typical of preorogenic ore formation processes
was volcanogenic-sedimentary ore accumulation, irrespective of the ores accu-
Metallogenic Zoning of Volcano-Plutonic Belts and Porphyry-Copper Mineralization 257

~1 bSSI2 ...
~~ ~1 ~5
~6 EJ
.':' 7 Ixxx 18 41lvlool~ ~m;a~<;lfO
1/,111 1+++112 b2]f3 ~'4 U{]fS

[9Jf6 ClJfl ~f8 a~~g ~20

Fig. S. The position of porphyry-copper deposits in metallogenic zoning of the andesitic belt in the
northern Near-Balkhash region (a fragment) of Kazakhstan. 1- 6 Caledonian and Early Hercynian
basement of the volcano-plutonic belt: 1 eugeosynclinal complexes; 2 miogeosynclinal complexes;
3 molasse; 4 gneiss; 5 granitoids; 6 ultramafic rocks; 7 volcanogenic and sedimentary sequences for
the Famennian-Tournaisian secondary geosynclines; 8 intrusions of the gabbro-plagiogranite second-
ary geosynclinal formation; 9-12 volcano-plutonic belt rocks: 9 -10 favorable volcano-plutonic as-
sociation: volcanics C I _ 2 -C2 (9a) and synchronous sediments (9b). 10 Ore-bearing granodiorite
massifs: Sodium (a); potassium-sodium (b); potassium (c). 11-12 Postore associations: 11 volcanics
C 2 - 3 - PI; 12 granitoids C 2 - 3 - P and younger; 13 mayor faults. 14 -20 mineralization; 14 copper
massive-sulfide; 15 massive-sulfide copper-zinc metal; 16 lead-zinc, stratiform; 17 skarn iron ore; 18
porphyry-molybdenum copper; 19 gold-silver; 20 molybdenum tungsten; 21 niobium-tantalum
258 V. A. Evstrakhin et al.

mulated (sulfide, iron-oxide, manganese) . Meanwhile, in the case of andesitic

VPB, preferable ore-genesis processes were those producing porphyry-copper
type systems.
As is known, ore matter is accumulated in porphyry-copper deposits as
veinlets and impregnations in intrusive bodies and at their exocontacts, breccia
pipes as well as veins, in the intrusive frame. The similarity of the style of ore
accumulation at the early and late epochs of the VPB emplacement manifests
itself primarily in the intensity of mineralization of the first two types.
For foreign VPB, Sutherland-Brown et al. (1971), Sillitoe et al. (1975), and
some other authors showed that some tin and tungsten deposits belong to the
porphyry type. All elements of such deposits are the ones common to porphyry
systems, though of a different qualitative expression.
Evstrakhin and Itsikson (1980), who purposefully studied the position and
structure of deposits related to andesitic VPB in the Soviet Union demonstrated
the constancy of the "porphyry style" of accumulation for some metals: tin, tin
and tungsten, uranium, and gold and silver.
These researchers identify porphyry deposits of the volcanogenic level with
metals being accumulated in breccia pipes, necks, and volcanic intrusions. The
group includes tin, uranium-molybdenum, and gold-silver porphyry deposits (see
Fig. 6). .
Tin-tungsten, molybdenum-tungsten, and tungsten-porphyry deposits are
related to deeper plutonic levels, and occur in hypabyssal intrusions as well as at
their exocontacts (see Fig. 7) .
Depending on the formational conditions of mineralization, on the wayan
ore-magmatic system was opened or closed, the products of early and main
stages of ore genesis can occupy different positions with respect to ore-bearing
porphyry intrusions. They can be far away from their apical parts, occurring as
veinlets or veined zones (tin deposits of a supra-intrusive zone, tungsten-tin
deposits) or, on the contrary, be concentrated in a stock or nearby one, giving
rise to veinlet-impregnated mineralization (porphyry-copper and porphyry-
molybdenum deposits, and tin-bearing greisens of an intrusive zone) . Later min-
eralization occurs in the peripheral and the upper parts of ore-magmatic systems
and controls the mineralogical-geochemical zoning of deposits.
Metallogenic specialization of the substratum of the belts plays a crucial role
in formation of some types of porphyry deposits as much as in the case of

Fig. 6. Ore zones and generalized geologic sections of a molybdenum-uranium deposit of the USSR. 1
Granite porphyry; 2 quartz porphyry; 3 clastic quartz porphyry; 4 andesite dacite lavas; 5 ore minera-
lization
Metallogenic Zoning of Volcano-Plutonic Belts and Porphyry-Copper Mineralization 259

Fig. 7. Ore zones and generalized

geologic plan of the molybdenum-
tungsten deposit of the USSR. 1 Por-
phyrites, schists, limestones Early
Cambrian; 2 ultrabasite Cambrian;
3 diorites, granodiorites, Early
Paleozoic; 4 plagiogranites Farly
Paleozoic; 5 bostonites Tri\lssic;
6 granite porphyry Triassic-Early
Jurassic; 7 quatz-molybdenite veins;
8 molybdenite stockwork; 9 granite
porphyry; 10 tungsten stockwork

I'" ",II 02 ~3 B~ ~5 U 6

8 7 ~8 G:2J9 QIO ESJII

porphyry-copper deposits. Taking the Sikhote-Alin folded system as an example,

Kulish and Usenko (1982) showed that widely spread sandy-shaly formation with
higher tin contents played a significant role in tin potential of granitoid and in tin
concentrations. Granitoids intruding these terrains are commonly tin-bearing.
The continuity of the metallogeny of substratum has been so far established
only for some VPB. Thus, in particular, epicratonic zones of VPB host copper-
molybdenum porphyry, molybdenum porphyry, tin and tin-tungsten porphyry,
uranium porphyry, and gold-silver-porphyry deposits. Epimiogeosynclinal
zones, along with porphyry copper, host tin porphyry deposits.
The foregoing series certainly do not encompass the whole metallogeny of
volcano-plutonic belts, but their identification is important for metallogenic
researches, viz, for outlining zones favorable for porphyry-copper mineraliza-
tion.

References

Evstrakhin VA, Itsikson MI (1980) Porphyry type ore deposits. Razved Okhr Nedr 1:9-16 (in
Russian)
Krivtsov AI (1977) Types of areas of porphyry-copper mineralization. Geol Rudn Mestorozhd
4:46 - 57 (in Russian)
Krivtsov AI (1983) Geological fundamentals of prognostication and prospecting for porphyry-copper
deposits. Nedra, Moscow, 254 pp (in Russian)
Krivtsov AI, Ageyeva ST (1979) Petrological features of volcano plutonic association of porphyry
copper mineralization areas. SOy Geol 7:36-49 (in Russian)
Kulish EA, Usenko SF (1982) Granites and tin ore formation in the Sikhote-Alin folded system.
Collected abstracts, Tbilisi, IAGOD VIth Symp, pp 79 - 80
260 V. A. Evstrakhin et al.: Metallogenic Zoning of Volcano-Plutonic Belts

Pavlova IG (1978) Porphyry-copper deposits. Nedra, Leningrad, 275 pp (in Russian)

Popov VS (1977) Geology and genesis copper- and molybdenum porphyry deposits. Nauka, Moscow,
201 pp (in Russian)
Sillitoe RH, Halls C, Grant IN (1975) Porphyry tin deposits in Bolivia. Econ Geol 70:913 - 927
Smirnov VI (1981) The trends in metallogenics. In: Distribution of mineral resources. Nauka,
Moscow, XIII:8-20 (in Russian)
Sutherland-Brown A, Cathro RJ, Panteleev A, Ney CS (1971) Metallogeny of the Canadian
Cordillera. Can Inst Min Metall Bull 64-709:37 - 61
Titley SR (1975) Geological characteristics and environment of some porphyry copper occurrences in
the Southwestern Pacific. Econ Geol 70:768 - 784
Vlasov GM (1978) Young geosynclines of the Pacific belt, their volcanogenic and ore formations.
Nauka, Moscow, 177 pp (in Russian)
Geological and Structural Conditions of Localization of
the High-Grade Ores of Porphyry Copper Deposits
I. M. GOLOVANOV, E. I. NIKOLAYEVA, and M. A. KHAZIKHIN 1

Abstract

A generalized geological-structural modeling of porphyry copper deposits reveals

the common features of formation of the high-grade ores of the inner parts of
stockworks. The common internal structure of the ore stockworks has been
established. The ore stockworks form a series of bodies concentric in plan, whose
copper concentration decreases outward from the centre. Studies of the Almalyk
area show that the morphology of ore stockworks and high-grade ore parts of
chalcopyrite-bornite ores is determined by the form of the stock roof, particular-
ly by the presence of ore-localizing negative structures - depressions, cavities
and trenches. The so-called structure of "an egg in a wineglass" is the prototype
of high-grade core sections of porphyry copper ores enveloped by the tongues of
porphyritic stocks at Almalyk. Low-grade ore envelopes the high-grade ores,
smoothing their outline and filling up the irregularities of the stock roof.

Porphyry copper deposits are considered to be the most important type of

copper deposit. In developed capitalist countries and in developing countries,
these deposits made up 35.50/0 of the total mineral reserves in 1963, and their
share increased to 65.1 % in 1980. For a long time porphyry copper deposits have
played the leading role in the balance of the world's reserves and output of
copper. The deposits determine the strategy of development of the world's
extractive copper industry for the near future (Krivtsov 1983, Pervago 1978, and
others).
The classical high-grade ores (copper content 3 - 5%) of porphyry copper
deposits are characteristic of the zone of secondary sulphide enrichment. The
average composition of the primary ores is notable for the low amount of copper
(0.3 - 0.6%) and accompanying elements - molybdenum, gold and silver.
Generalization of the data on the structure of deposits in the USSR, USA,
Canada and Iran and elaboration of the geological-structural models of
porphyry copper deposits show the common features of the high-grade hypogene
ore sections of porphyry copper stockworks. The most important initial elements
in a model appear to be a porphyritic ore-generating granitoid stock and an ore-

1 Central Asian Scientific Research Institute of Geology and Mineral Resources, Ministry of
Geology, Tashkent, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
262 I. M. Golovanov et ai.

bearing stockwork. In Almalyk the core structure of the latter is determined first
of all by the morphology of the roof of the porphyritic stock and by the composi-
tion and structure of the enclosing rocks. A common inner structure of ore stock-
works has been established for many ore deposits. The stockworks have been
found to form a series of bodies concentric in plan, with copper concentration
decreasing from the centre.
The localization of stockwork bodies of copper ore (especially those of high-
grade ores) is, clearly, in negative structures ("rims", depressions, cavities) of the
intrusives in the Kounrad and Aktogai deposits in Kazakhstan, Erdenetuin-Obo
in Mongolia (Krivtzov 1983), and the Almalyk group of deposits in Uzbekistan
(Golovanov 1982).
Porphyry copper deposits of the Almalyk area may be considered typical of
porphyry copper ore formation. The history of their complex investigation and
industrial exploitation numbers 50 yrs. During this period a lot of work has been
done at the deposits - from prospecting to exploitation. The deposits were ex-
plored to a depth exceeding the pinching out of mineralization. Within the limits
of the ore field there are ore bodies considerably stripped by erosion, as well as
those unstripped.
The detailed information on the vertical section of the Almalyk deposits and
the similarity of their composition to that of the majority of the world-known
porphyry copper manifestations allowed us to take these deposits as a standard
for determining the localizing conditions of porphyry copper mineralization and
for making a complex model of the ore-metasomatic and magmatic system.
The geological-structural specificity of the area is determined by the develop-
ment of two systems of faults. The first system is represented by northwest-trend-
ing faults that have no conspicuous manifestation in the modern erosion section.
These faults control the distribution of porphyritic stocks. The second system
conspicuously is represented by sublatitudinal long-lived faults. They are mani-
fested in the modern erosion section by cataclastic zones.
Effusive quartz porphyries (Lower Devonian), diorite and monzonites
(Middle Carboniferous) and the rocks of porphyritic stocks (Upper Car-
boniferous) are considered to be the main ore-bearing rocks of the Almalyk
deposits.
Judged from their petrographic and petrochemical parameters, the por-
phyritic stock rocks are quartz monzonite porphyries. They appear to be the last
link in the incomplete magmatic association gabbro-diorite-diorite-monzonite
(syenodiorite)-quartz monzonite porphyry. Hyperalkalinity of the porphyritic
stocks is ensured by an alkali sum (Na20+ K20) of 8.40/0, with K20 dominating
over Na20 (K20/Na20 = 1.56-2.13). Biotite content is higher and amphibole
content is lower in depth and also in the inner parts of stocks. Clinopyroxene
appears also in depth.
Morpho-structural modeling was shown to the rather informative for study-
ing the relationship between magmatism and mineralization. To make such
models the authors used the data from exploratory drill holes.
Three-dimensional block diagrams (models) of fields with ore concentrations
and ore-bearing stocks are dissected to show sets of parallel vertical sections pass-
ing across the strike of the ore bodies (Fig. 1). The ore bodies are correlated on
Geological and Structural Conditions of Localization of the High-Grade Ores 263

Fig. 1. Geological sections of a porphyry copper deposit. Hachures: porphyry stock, black:
stockwork of high-grade ore

the maps of various hypsometric levels and on longitudinal vertical sections. A

similar method has been used in the Mount Morgan deposit (Cornelius 1969).
The group of geologists from the Central Research Institute of Geological Ex-
ploration, under the leadership of Krivtzov (1983), also used this method.
To make the models of the bodies, various outlines of copper concentration
have been used. The external outline corresponds to low-grade ores (0.20/0 Cu)
and the internal contours represent high-grade bornite-chalcopyrite ores (up to
1% Cu) (Fig. 4) .
The analysis of morpho-structural models that included five units of the
Almalyk ore field showed that all the deposits investigated, though different in
size, had the same structure of ore bodies and porphyritic intrusives and the same
dimensional relationship between these two.
The roof morphology of the porphyritic stock is considered to be the decisive
factor in the localization of copper ores at Almalyk. The ore stockwork of any
deposit has a certain correspondence to the host body of quartz monzonite por-
phyry. It should be noted that the bigger deposits are known to have bigger
stocks. The smaller deposits are characterized by the simplest morphology of the
apical part of the stock and its copper ore bodies, respectively. The form of
stocks and ore bodies becomes more complicated in big units (Dalnee,
Kalmakyr).
The apical parts of all ore-generating intrusives in the Almalyk ore area are
characterized by negative structures, such as depressions, pits and trenches, that
are bordered by tongues (apophyses) of the intrusives. These negative structures
are represented by the subsidence domes, the latter are formed when the stock
edges have already consolidated, but the middle part reduces in size while
cooling, and then shrinks. Very frequently the tongues bordering the depressions
form solid, ribbed, multidissected walls while emerging, but sometimes they look
like the tentacles of an octopus. The simplest possible stock structure reminds
one of a crater with sloping walls.
264 I. M. Golovanov et al.

II

e
Fig. 2 a-e. Models of porphyry stocks and ore stock works outlined by various copper content. Dark
to light shading stock; dark-bordered stippling ore-bearing stockwork. I Morphologically simple
deposit; II morphologically complicated deposit. a Porphyry stock; b stockworks of rich ores in stock
depressions; c, d stockworks of intermediate copper content; e stock works of poor copper ores
 Geological and Structural Conditions of Localization of the High-Grade Ores 265

High-grade stockwork ore is situated directly in the stock roof depression,

and the shape of the ore body imitates the relief of the stock roof surface (the so-
called structure of an egg in a wineglass). Ore bodies with lower copper concen-
trations gradually expand and cover the tongues, walls and upper parts of the
stock. The form of intermediate stockworks is similar to that of the central part,
but has a smoother profile (Fig. 2, I). The approximate geometry of the ore
bodies defines them as ellipsoids of revolution or elliptical paraboloids of revolu-
tion with similar or equal relations between short axes. The porphyritic stock and
its ore bodies are sure to have a most regular and isometric form if the intrusive
was emplaced in an isotropic medium. In this case, in plan, the contour of the
stock and its ore bodies approximates a circle. The presence of enclosing rocks of
diverse composition as well as tectonic dislocation, distorts the form of the
intrusive and its ore bodies. The Almalyk deposits have several types of ore-con-
taining rocks. Stockwork mineralization is usually associated with monzonites
(syenodiorites), sometimes with diorites, and very rarely with effusive quartz
porphyries (Fig. 3). The roof morphology pf large intrusives is characterized by
the presence of sags. Sometimes these sags are localized on different hypsometric
levels. The bottom of large depressions is complicated by swells and crests. The
morphology of ore bodies, particularly that of their high-grade core areas, is
determined by the complicated relief of the stock surface. If there are several sags
in the intrusive roof, each negative structure has a corresponding section of high-
grade ore. Swells, crests and tongues in the sag bottom prevent the even develop-
ment of high-grade ore in the sags. Low-grade ores form a body of complicated
morphology that includes also high-grade cores. While expanding, this body
envelopes the upper part of the stock and its tongues (Fig. 2, II).

A \ ,

, A

. --
(
/'
,
,

,
,I
\,
\
v
., ':"
'\.. ,
'\
'"z .. ~

" v

Fig. 3. Typical geological section of a porphyry copper deposit (for legend see Fig. 4)
266 1. M. Golovanov et at.

.,-..- '. _... a. _ . .. ~....- : '

1\ v
/'/'<r-- :.~.- '1"-:"
. •....: . . _ • . .• ~

r ;
~ ~:~~~ . ~r \\'~
2'
v
~, \ ._.-f·-·\J
v\. ~ "(v
I /' I!\\ j I' \ \ \
v 1\ /'' L \N \i l\,\\,,'- f I \v \ /,

/. i/ '- t /'~ I I\v

! I
v \i __

I 'I
.
v II
:.... .....
V :.\11
.... ~I
" !
I I! \!I v
v II ......... .: \\ ! . I i
1\ II/ ?.I · ............. .. . ...... . \1
v ..... ............ ...: ~
'\ 1 / 1\
. I V I"
II
I\
\
/J . . . " , : .~, 1 , 1 .; I l\ i
"/ ? [ ~ V /' i \ \ V
/ 1 /v : / .i\",- '/ I/ i"
I \ ' 'I
V \ \
v I ." j" \\
I 1/ ,~'-. ~. 1\ \ \
I .I ~. J \ \
.I 1/

I
I/ >~
x
X

x x"
" "
. ' ),
~\\
' \ 1\

-'---\
x \\
I

1\
1\

v, \
\. v \

\I,
)
\. \
\
v

" \\ v 1\ \ \

x x X ,\ X X X x,\, v " I
\v \
I
x x t., 1\ V 1\ \ \
\\ \ \ I
x x x x x \\. v /1 Ii

Fig. 4. Typical geological section of a porphyry copper deposit in a wedge-shaped tectonic block .
1 Quartz porphyries; 2 diorites, gabbro-diorites; 3 monzonites (syenodiorites); 4 quartz monzonite
porphyries; 5 faults; 6 faults with fracture zone; 7 rock boundary; 8 drill hole; 9, 10, 11, 12 contours
of ore stock works according to copper content; 9 rich copper ores; 10, 11 intermediate content;
12 poor copper ores

The level of erosion determines to a great extent the morphology of the

deposit. The Almalyk deposits, except the Kalmakyr, appear to be practically
uneroded. The Kalmakyr deposit has been stripped only to one-third. Thus,
earlier the Kalmakyr ore zone structure was described as a ring put on a stock;
actually the stock is the bottom swell of a big downwarp in a large intrusive.
It is important to note that the negative structures of the stock roof coincide
with faults, paired to give a wedge-shaped block (Fig. 4). These negative struc-
tures were shown to be the main factor determing the location of porphyry
copper mineralization. The localization of core parts of high-grade copper ores
directly in the roof downwards of the fault-broken intrusives determines the loca-
tion of ore and the metasomatic zonation of the porphyry copper system.
The composition and zonation of metasomatites in porphyry copper deposits
(Fig. 5) vary, depending on:
1. Localization of the ore-metasomatic system relative to large faults.
2. Localization of the ore-metasomatic system in negative structures of the upper
part of the stock; i.e., morphology of subsidence dome, inhomogeneity of the
enclosing rocks, and stock configuration.
Geological and Structural Conditions of Localization of the High-Grade Ores 267

A B

Fig. SA, B. Ore-metasomatic zonation model of a porphyry copper deposit. A Section perpendicular
to the ore-bringing fault plane; B section parallel to the fault plane. Zones: I central; II inner;
IIIa,IIib intermediate; IV fringe; V peripheral; 1 porphyry stock; 2 silicic alteration; 3 K-feldspar
alteration; 4 phyllic alteration; 5 argillic alteration; 6 propylitic alteration I; 7 propylitic alteration II.
Abbreviations: Py pyrite; Cp chalcopyrite; Mt magnetite; Hm hematite; Bn bornite; Mo molyb-
denite; Of galena; Sf sphalerite; S, Te sulphides, tellurides; Au, Ag native gold and silver

3. Physico-mechanical characteristics (composition, density, jointing, porosity,

etc.) of the enclosing rocks.
The metasomatic composition naturally changes relative to an ore-generating
fault and to the quartz monzonite porphyry stock. At their intersection, the fault
and the stock determine a focal plane for metasomatic zonation. The zones of
silicic, K-feldspar, and phyllic metasomatites enclosed in a propylitic aureole
replace one another while moving away from the given plane, progressively
advancing into the enclosing rocks. The metasomatic system is pyritizing, and
pyritization is especially intense in the external phyllic zone, which is anhydrite-
bearing at the lowermost levels. The external lateral zones are more persistent at
depth than the internal ones.
Mineralization of sub economic character occupies a certain place in the
general zonation of the system. Usually it is localized much higher than pay ores.
Primary ore zonation is expressed in the change of molybdenum mineralization
to copper-molybdenum, copper, polymetallic, or gold-silver mineralization.
These changes go from the centre to peripheral parts as well as vertically from the
268 I. M. Golovanov et al.

bottom to the top. Zonation is expressed relative to the contact of the porphyritic
intrusion and ore-localizing faults.
Changes in the mineral composition of the ores are determined by the same
zonation as that in the metasomatites: ore-zonation is three-dimensional and
coaxial. Five zones of various mineral compositions can be identified.
1. Central zone - silicified barren core asymmetrically placed downwards from
the system centre to the stock body.
2. Internal zone of maximum grade ores - high-grade copper core with gold and
silver.
3. Intermediate zone - lower grade, but thicker copper-molybdenum ores.
4. External zone - low-grade copper-molybdenum ores with lead, zinc and
silver.
5. Peripheral zone with lack of copper mineralization and presence of veined
gold-silver metallization.
Dimensionally, central zone ore mineralization corresponds to certain facies
of metasomatic zoning. Impregnations of hematite and magnetite (the "bottom"
of the ore-metasomatic system plunging within a stock body) are associated with
the silicified core. Magnetite-chalcopyrite-bornite ore with gold, silver and a
number of additional minerals - sulphides, sulphosalts, tellurides and native
elements (high-grade copper ore core of internal zone) (Fig. 5) - is associated
with orthoclase-rich rocks and lowermost levels of phyllic alteration.
Judged from the amount and concentration of long-lived veinlets, the high-
grade copper ore zone may be called the zone of maximum ore genesis. Mineral
associations in this part of the stockwork are very diverse. They are characterized
by a wide range of ore-depositing conditions.
The veinlets, from early to late, vary in composition. They may include
magnetite, pyrite, molybdenite, chalcopyrite, chalcopyrite-bornite-tetrahedrite-
gold with chalcopyrite, digenite, sulphides and sulphosalts of tin, bismuth,
nickel, silver, with native copper, bismuth, tin, silver and others; sphalerite-chal-
copyrite-gold and many other veinlets (Nikolayeva 1980).
The deposition of the inner zone mineral complex started under conditions of
relatively high temperature, as evidenced by decomposition of bornite-chalco-
pyrite solid solution; then the veinlets underwent repeated opening. Thus, miner-
alization was renewed and early minerals were recrystallized and desulphurized.
At the last stage of ore formation the ores underwent the influence of low-tem-
perature solutions with high-oxidation states: the formation of hypogene
goethite, chalcocite, digenite, covelline and also low-temperature minerals -
stromeyerite, and others - took place. Recycled minerals are also found there.
The stage is characterized by the following processes: diffusion regrouping -
metamorphogenic decomposition of copper sulphide solid solutions, and dis-
sociation of electrum, resulting in the appearance of native silver and gold of
high fineness. These processes and some others indicate the extreme permeability
and active ore genesis of the so-called bowl, plunging within the deep negative
structure of the stock roof. The inner zone of the stock roof partially involves
phyllic and early propylitic zones passing outward from the stock centre. This in-
ner part has its own zonation: the lower section is represented by a hypogene oxi-
Geological and Structural Conditions of Localization of the High-Grade Ores 269

dation zone (locally absent) with bornite-bearing ores; the middle section by
associations in which chalcopyrite dominates, the external edges by enrichment
in molybdenite.
The central high-grade copper core is known to have high ratios of
copper: molybdenum, gold: molybdenum, and silver: molybdenum. It is also
distingusihed by having a maximum-positive-paired correlations coefficient for
copper with gold and silver (up to 0.7 - 0.8) and a negative one for molybdenum.
Langton and Williams (1982) described similar geological-structural condi-
tions for the localization of high-grade ores and metasomatites of the Dos Pobres
deposit in Arizona. The core sections of high-grade copper ores are represented
by bornite-chalcopyrite ores. Copper, gold, silver and K20 have common or con-
tiguous zonation centres. Molybdenum-enriched sites are situated within the lim-
its of the 0.4010 copper contour. Copper, gold, silver and K20 zonation centres
are near the Foothill fault.
The generalized model of ore and metasomatic zonation of similar distribu-
tion has been also described for the Christmas deposit by Koski and Cook (1982),
for the Bingham deposit by Einaudi (1982) and for some others. According to
Einaudi's scheme the deep core section of copper is represented by chalcopyrite-
bornite and is situated under the zone of intensive potassium feldspathization.
Approaching the peripheral parts, the ores change to chalcopyrite-pyrite and
then to pyrite-chalcopyrite.
According to John (1978), the bornite-chalcopyrite zone of the Bingham de-
posit appears to be the central part of the high-grade copper ore body containing
equal amounts of bornite and pyrite. Chalcopyrite occurs together with bornite
and sybstitutes for it. Jambor and Beaulne (1977) reported the predominance of
bornite over other sulphides in the central part of the high-grade core zone, e.g.
for the Valley Copper deposit in Canada.
Potassium metasomatism (potassium feldspar, biotite and anhydrite) and de-
velopment of the bornite-chalcopyrite-digenite association are connected with
the early-, late-magmatic stage in the Chuquicamata deposit in Chile (Soto 1981).
Rock transformation and appearance of the pyrite-bornite and pyrite-bornite-
chalcopyrite associations took place during the second main phase of the hydro-
thermal stage. The central part of the EI Abra deposit in Chile is characterized
according to Ambrus (1977), by a high content of copper and a low content of
sulphur in biotite-potassium feldspar metasomatites produced by diorites. The
high copper content is provided here by bornite-chalcocite-chalcopyrite and bor-
nite-chalcopyrite associations.
The review presented here shows that the established morphostructural pecu-
liarities of ore-bearing intrusives and stockwork bodies, and the zonation distribu-
tion of ore and metasomatic complexes, appear to be common features for a num-
ber of porphyry copper deposits. The detailed geological investigation of other
deposits (their morphostructural constructions in particular) will make it possible
to determine similar structural characteristics for high-grade core zones of
porphyry copper deposits. The foregoing data on the structure of ore-generating
porphyry intrusives, copper-ore stockwork bodies and their dimensional position,
ore zonation, and metasomatism may be recommended for application to certain
stages of geological prospecting and exploration and probably also to exploitation.
270 I. M. Golovanov et al.: Geological and Structural Conditions

References

Ambrus G (1977) Geology of the El Abra porphyry copper deposit. Chile. Econ Geol 72:1062 -1085
Cornelius KD (1969) The Mount Morgan mine - a massive gold-copper pyritic replacement deposit.
Econ Geol 64:885 - 902
Einaudi MT (1982) Description of skarns associated with porphyry copper plutons. In: Advances in
geology of the porphyry copper deposits, Southwestern North America. Univ Ariz Press, Tucson,
Arizona, USA, pp 139 - 183
Golovanov 1M (1981) Geological-genetic model of concentric zonation of a porphyry copper deposit.
Genetic models of endogenic ore formations. Abstracts of reports of All-Union conference.
Novosibirsk, p 68 (in Russian)
Golovanov 1M (1982) Conditions of ore localization in concentric zonal porphyry copper systems .
. Genesis of ore deposits. Rep IAGOD Vlth Symp (Abstr), Tbilisi, pp 202 - 203 (in Russian)
Jambor JL, Beaulne JM (1977) Exploration possibilities for porphyry deposits in the central part of
Highland Valley, British Columbia. Pap Geol Surv Can 1:101-106
John EC (1978) Mineral zones of the Utah copper orebody. Econ Geol 73:1250-1259
Koski RA, Cook DS (1982) Geology of the Christmas porphyry copper deposit, Gilta County,
Arizona. In: Advances in geology of the porphyry copper deposits, Southwestern North America.
Univ Ariz Press, Tucson, Arizona, USA, pp 353 - 374
Krivtzov AI (1983) Geological basis for porphyry copper deposit forecasting and exploration. Nedra,
Moscow, pp 256 (in Russian)
Krivtzov AI, Yudin 1M (1976) Hypogenic zonation of porphyry copper formation deposits.
Geological methods of search and exploration of ore deposits. Mem All-Union Inst Miner Res
Econ, p 44 (in Russian)
Langton JM, Williams SA (1982) Structural, petrological and mineralogical controls for the Dos
Pobres orebody: Lone Star Mining District, Graham County, Arizona. In: Advances in geology of
porphyry copper deposits, Southwestern North America. Univ Ariz Press, Tucson, Arizona, USA,
pp 335-352
Melnikova LV, Poletayev AI (1981) Mineralogical zonation of porphyry copper deposits and the
possibility of using it for search and exploration (after the example of Aktogai). Prospects of
copper ore provinces in Kazakhstan. Alma-Ata, pp 9-17 (in Russian)
Nikolayeva EI (1980) Paragenesis of gold and silver in porphyry copper deposits of the Almalyk ore
field. Notes of Uzbek Branch. Mem All-Union Miner Soc 33:16- 23 (in Russian)
Pavlova IG (1978) Porphyry copper deposits. Nedra, Leningrad, p 273 (in Russian)
Pervago VA (1978) Geology and economics of porphyry copper deposits. Nedra, Moscow, p 170 (in
Russian)
Popov VS (1977) Geology and genesis of porphyry copper and porphyry molybdenum deposits.
Nauka, Moscow, p 204 (in Russian)
Shayakubov TSh, Golovanov 1M, Rakhubenkov AT (1983) The Dalnee porphyry copper deposit.
Nedra, Moscow, p 110 (in Russian)
Soto PH (1981) Alteraci6n y mineralizaci6n primaria en Chuquicamata (Chile). Tecniterrae 8,
43:27-44
Elements Determining in the Geological-Structural Model
of the Erdenetuin-Obo Copper-Molybdenum Ore Field,
Mongolia
M. ZHAMSRANt, V. I. SOTNIKOy 2, A. P. BERZINA 2, D. GARAMZHAYt,
and Yu. A. SARYANOy 1

Abstract

Copper-molybdenum mineralization in Mongolia is localized within the limits of

regional structures referred to as the superimposed metallogenic belts. The latter
is spatially controlled by the late Paleozoic (in Southern Mongolia) and late
Paleozoic-early Mesozoic (in Northern Mongolia) volcanic belts, elongate in a
sublatitudinal trend. The through-going transverse fault system of the northwest
trend has played an important role in forming ore-concentrating structures.
Intersection of transverse faults and ones that trend along the volcanic belt
account for the development of a block fabric and highly permeable zones that
are favourable for the existence of long-lived magmatic centers that host the ore
fields.
The Erdenetuin-Obo ore field is located in the Erdenet-Tsagansuburgin ore-
concentrating structure that stretches throughout Mongolia and on into the
USSR (in the north) and into China (in the south). Distribution and localization
of mineralization within the ore field are determined by three fault systems:
northwest, north-northwest (submeridional) and west-southwest (sublatitudinal).
Of these fault systems, the first one was of the most significance. It is considered
to have provided channels for fluid flow concentration at a level necessary to
produce large volumes of ores.
Nowadays, Mongolia is among the leading copper-molybdenum provinces of
the world due to the discovery of the Erdenetuin-Obo (Khasin et al. 1977) and
Tsagan-Suburga (Sotnikov et al. 1980) deposits as well as of numerous copper-
molybdenum prospects in different regions of the country. At the Erdenetuin-
Obo deposit, the most efficient mining and concentrating plant in Asia is func-
tioning.
Copper-molybdenum mineralization in Mongolia is localized within the limits
of regional structures of the superimposed metallogenic belt type: North
Mongolian, Central Mongolian and South Mongolian. Their formation is related
to the development of extrageosynclinal tectonic-magmatic processes, and they
are spatially controlled by the late Paleozoic and late Paleozoic-early Mesozoic
volcanic belts (of the same names, respectively), elongate in sublatitudinal trend

1 Institute of Geology and Mining Industry of the MN, Ulan Bator, MNR
2 Institute of Geology and Geophysics, Siberian Branch of the USSR Academy of Science,
Novosibirsk, 630090, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
272 M. Zhamsran et a\.

u R
s

90
,
gO
,
18Q
I
km

Fig. 1. Late Paleozoic-early Mesozoic volcanic belts in Mongolia and the Erdenet-Tsagansuburgin
ore-concentrating structure. The volcanic belts (1- 3): 1 North Mongolian (PZ 3 - MZ t ); 2 Central
Mongolian (PZ 3 - MZ t ); 3 South Mongolian (PZ 3 ); 4 the fault systems (1- VI) : I Khangai; II Mon-
golo-Okhotsk; III Main Mongolian; IV Bayankhongor; V Kerulen; VI Gobi-Tien-Shan; 5 the
Erdenet-Tsagansuburgin ore-concentrating structure; 1, 2 the copper-molybdenum deposits:
Erdenetuin-Obo (1) and Tsagan-Suburga (2)

(Fig. 1). On the whole, volcanic belts inherit the orientation of the Hercynides on
which they are superimposed. Similar orientation is observed in structures with a
Caledonian or a Proterozoic basement. All these considerations suggest the con-
nection of volcanic belts with deep long-lived faults , the latter periodically
becoming active.
The North Mongolian volcanic belt is traced by a system of extensive faults of
east-north-east trend on the east and of west-north-west ones on the west (Fig. 1).
The fault system on the west of the North Mongolian belt is designated as the
Khangai lineament, and the one on the east represents the western flank of the
Mongolo-Okhotsk lineament extending from the USSR. Late Paleozoic volcanic
rocks of the South Mongolian belt are concentrated in two linear zones, one of
which is controlled by the Main Mongolian lineament, and the other by the Gobi-
Tien-Shan system of faults (Yarmolyuk and Kovalenko 1981). The Central
Mongolian volcanic belt is limited to the north by the Bayan-Khongor and
Kerulen faults, and to the south by the Main Mongolian lineament. According to
their importance, the magma-controlling faults are deep and regional ones that
governed the development of orogenic complexes (Geology of the Mongolian
People's Republic 1973). These faults, where they reached the zones of deep-
seated magma generation, where the principal magma-delivering structures.
Volcanic belts can be regarded as common regional ore-controlling structures.
Elements Determining in the Geological-Structural Model 273

Through-going transverse structures of nortwest (in Eastern Mongolia) and

northeast (in Western Mongolia) trends played an important part in forming ore-
concentrating structures. The greatest one is the Erdenet-Tsagansuburgin ore-
concentrating structure (Fig. 1) that is traced across all Mongolia from north to
south (from the USSR frontiers up to the Chinese ones) and that crossses geosyn-
clinal structures of different ages and the Proterozoic basement. This ore-con-
centrating structure includes the majority of ore prospects of the North
Mongolian, Central Mongolian and South Mongolian belts; the Erdenetuin-Obo
and Tsagan-Suburga deposits occur along them. The structure is bounded by
deep long-lived zones of basement faults. These are combination of faults, traced
over tens of kilometres, zones of crumpling, crushing and high jointing as well as
individual, usually echelonlike fractures tens to hundreds of metres long. In
many cases these tectonic dislocations are cryptic, revealed only during large-
scale mapping and structural investigations of deposits and prospects.
The combination of both transverse faults, involved in the system of through-
going ore-concentrating structures, and of others trending along the volcanic
belts, gave rise to the development of block structures as well as of highly perme-
able zones. These regions are characterized by intense occurrence of late
Paleozoic (late Paleozoic-early Mesozoic) volcanoplutonic association. The most
intense, long-developing, and diverse magmatism occurred within the limits of
the ore fields. The latter spatially fix centres of continuous endogene activity.
Evidently, an ore field can be regarded as a taxonomic unit, most completely
summing up all features of evolution of an ore-magmatic process in a concrete
geological situation. As a typical example of such an ore field, picting the condi-
tions of occurrence of copper-molybdenum mineralization in geological struc-
tures of Mongolia, one can study the Erdenetuin-Obo ore field.
Copper-bearing mineralization of the Erdenetuin-Obo ore field is clearly con-
trolled by the zone of drawn-together northwest faults, arranged in an echelon
pattern from the Tola river up to the lower and middle reaches of the Eigiin-gol
river. Localized in the Erdenet-Tsagansuburgin ore-concentrating structure, this
zone is known as the Erdenet zone (Fig. 2) and not only determines the arrange-
ment of ore-metasomatic rocks, but also influences both localization and
morphology of intrusive massifs and dikes, different in age and contained within
the framework of Lower-Upper Permian volcanic rocks.
The Erdenet zone is traced along the contact of the regional blocks (Fig. 2),
differing in composition and interval of movement (Mossakovsky and
Tomurtogoo 1976). It has been suggested (Gretskaya and Mossakovsky 1969,
Mossakovsky and Tomurtogoo 1976) that the block structure was best developed
in the Middle Jurassic period, but its elements began to form earlier. Being
formed initially in the Caledonides; the faults continued to develop during the
accumulation of the Permian volcanogenic series and were regenerated at the end
of the Permian - at the beginning of the Triassic period. The block structure is
mainly delineated by a combination of faults of three directions: northwest,
northeast and north-northeast. Of these, the fault and highly fractured NW-
trending system is most clearly defined.
To achieve efficiency in exploration the most promising areas on the whole
are those with a combination of the above fault systems showing maximum activ-
274 M. Zhamsran et al.

20 0 20 40 Km
I !

~2
1:-::::':::,:-1J

Fig. 2. Block pattern of the northern part of the Erdenet-Tsagansuburgin structure. The blocks
(J - 4): 1 Northern; 2 Central; 3 Southern; 4 Eastern; 5 faults; 6 the Erdenet ore-bearing zone

ity and highest permeability. However, those structures most responsible for the
total character of distribution of arrangement of copper-molybdenum mineral-
ization within the framework of the Erdenetuin-Obo ore field remain the
northwest dislocations (Fig. 3). The largest of them are represented by faults and
uplifts with an amplitude up to several hundreds of metres. This most clearly
manifested fault system appeared to be the main ore-concentrating structure
providing maximum drainage for ore-bearing fluids and highest metal
concentrations in relatively local areas.
The Erdenetuin-Obo ore field is subdivided into three areas: the Erdenetuin-
Obo deposit, the Central area and the Southern area (Fig. 3). In the deposit
(Fig. 4), magmatic rocks of the Selenginsk complex mainly occur. They are
represented by granodiorite, granite and gabbro-diorite. Granosyenitelike rocks
in which generation metasomatic processes played an important role are
widespread. In pendants of the granitoid massif, relicts of Riphean-Cambrian
gneisses, amphibolites and crystalline schists, often considerably K-feldspathized
survive. It should be emphasized that the occurrence of potassium metasomatites
is the most characteristic feature of the rocks exposed in the ore-bearing district,
especially close to the ore manifestations and in the highly permeable zones.
Within the deposit area, various shapes of porphyry bodies of the ore-bread-
ing magmatic complex are widespread. These consist of granodiorite and plagio-
granite porphyries, quartz-diorite porphyries, and more rarely, of granite and
Elements Determining in the Geological-Structural Model 275

v
v~
V v
v

+. '+ •
--::::; v v
0,7S
I
0 a's 1,5Km
t + '_---
,,-t- ....
It :

~~ + +
,,+ ,~~ ,
" +q::!l
' +.nr" +
~2 e±:::B5
'-, '+, ~ + ' :t
~J c;;::::::] 6 " . . . .:t'" ,,; , /
c=JI ,/
...,."
,,;

Fig. 3. Geological scheme of the Erdenetuin ore field, 1 Quaternary sediments; 2 volcanic rocks,
P I - 2 ; 3leucocratic granite, T 1 ; 4 porphyry (the ore-bearing magmatic complex), P 2 - T 1 ; 5 granitoid
(the Selenginsk magmatic complex), P; 6 faults; 7 the mineralization areas (I - III) I Erdenetuin-Obo
deposit; II Central area; III Southern area

granosyenite porphyries with transitions to dacite and rhyolite-dacite in intrusive

endocontacts. Pre-ore, intra-ore and post-ore porphyries are recognized. On the
whole, the porphyries of different intrusive phases are similar in composition.
Pre-ore porphyries are mainly represented by northwest-trending stocks. Among
intra-and post-ore porphyry bodies, dikes of the north-northwest strike predomi-
nate. Stock like porphyry bodies are continuously accompanied by near-contact
explosion breccias composed of granitoid fragments. The latter are cemented by
finely-crushed biotitized and K-feldspathized material derived from the same
granitoids.
In the central portion of the deposit, a porphyry body rather hydrothermally
altered was mapped. Being generally stretched out in a northwest direction which
is consistent with the total orientation of the Erdenet fault zone, the stocklike
body about 2100 m long and 900 to 1 000 m wide is of complex shape with
numerous embayments into the host rocks as apophyses and satellites about 250
to 500 m long and 10 to 100 m wide. The stock dimensions decrease with depth,
276 M. Zhamsran et al.

x~ -+ + + l- + r +
+ + + \ +
+
x x
x x
x X

X )(

'. ?< x )<

)(
)< ·. X
)< x x
)( x
x ;' x x
x
X
X
)<
)<
x x
x x:
x
'1< X
X
x .¥
x
x ' x'
..x X
x
x X x:
X X
>( X X X
x
X X
60 0 6Q /PO 180 X X
~
AUf.trs
)( X ;'
X X
X X
X
1-....11 E;Jz I :·~~I J G:Jq ~5 X
·.X
'. X
)<
X x )<
)<
X'.

Fig. 4. Geological scheme of the Erdenetuin-Obo deposit. 1,2 Dikes, T2 - J 1 : 1 andesitic and trachy-
andesitic porphyry; 2 syenodiorite; 3 stock of felsite porphyry, T 2- J I; 4 porphyry. The Selenginsk
magmatic complex (5 - 7): 5 granodiorite; 6 granite; 7 gabbro-diorite; 8, 9 Outline of hydrothermal
alteration: 8 strong silicification; 9 sericitization and silification; 10 outline of ore body; 11 faults

its morphology becomes somewhat simpler changing to a piplelike body with a

northeast pitch. In the east the stock is cut off by the Central fault.
Dikes and small isometric bodies of fine-grained leucocratic granites, micro-
granites and granite-aplites are less developed in the deposit. They are wide-
spread in the southwestern portion of the deposit and continue into the western
part of the Central area (Fig. 3). Leukocratic rocks belonging to the post-ore
phase are hydrothermally altered (mainly poor K-feldspathic and bleached) only
at certain places, mainly in the zones of high permeability and early mineraliza-
tion.
In the deposit area, rocks of the Late Triassic-Early Jurassic age are repre-
sented by a feldsite porphyry stock with a northwest orientation as well as by
dikes of andesitic and trachyandesitic porphyry that fill northwest and especially
north-northwest to north-northeast tectonic zones. Syenodiorite porphyry dikes
predominantly have a west-southwest strike.
The structural pattern of the deposit is determined by a combination of the
three main systems of faults: generally northwest, west-southwest and north-
northwest. In general, the most important is the northwestern system that is a
Elements Determining in the Geological-Structural Model 277

component of the large, long-lived Erdenet fault zone the main structural
element of the ore field. This fault system accounts for the localization of the
stock- and dikelike bodies of porphyries as well as for the generation of the zone
of high jointing and permeability to which ore-metasomatic rocks of the deposit
are confined.
The north-northwest fault system is represented by the two large zones, i.e.
the Central and Western ones. The former is developed most intensely, being
represented by a series of large sutures and highly jointed zones marked by the
andesite porphyry dikes. This seems to have prevented ore-metasomatic
processes from spreading to the east. The latter, together with the registered
north-northwest orientation of quartz-sulphide veinlets, testify to the
development of faults of this direction during the pre-ore period also. Most
intensive renewal of the Central zone evidently took place in the Late Triassic
when, in its eastern part, numerous dikes of andesite porphyry were involved in a
large north-northwest dike belt stretching far beyond the deposit. Along most
joints of the Central zone, post-ore movement with occurrence of mylonites and
clay gouge took place. The Western fault zone, which is similar to the Central
one, but differing from it by the lesser intensity of the tectonic processes,
together with northwest faults, controlled the localization of the Late Triassic
felsite neck in its northern part.
A west-southwest fault system is traced by apophyses of the porphyry stock, a
number of quartz-sulphide veins as well as post-ore dikes of syenodiorite
porphyry. This system is of minor significance and has influenced the formation
of the deposit structure much less than the two fault system described above.
Vein-disseminated ore mineralization is confined to the zone of hydro-
thermally altered rocks elongated northwest for over 2.8 km, its width being 0.3
to 1.3 km. Spatially, it is related to the porphyry stock that penetrates into the
granitoid stocks of the Selenginsk complex for over 300 - 500 m. Ore mineraliza-
tion is observed at the depth of 1000 m without signs of pinching out. Towards
the periphery of the ore body, the succession of vein-disseminated mineralization
is followed for more massive veins. Quartz-sulphide veinlets have strikes that
follow the three directions of jointing (Khasin et al. 1977): northwest
(305 - 330°), north-northwest to north-northeast (355 - 20°) and west-southwest
(270°). Veins of the first two directions dominate. The veinlets range in width
from parts of a mm to 2 - 3 cm and usually are traced along the same trend over
tens of centimetres and their direction often changes. The veins most persistent in
trend are characteristic of the periphery of the deposit. In the central parts of the
ore body, especially in the near-contact zone of the porphyry stock, high-grade
breccia ores, irregular in morphology, occur. Spatial zones of pre-ore explosion
breccias (including those connected with the emplacement of the porphyry stock)
as well as areas of high jointing with maximum occurrence of joints of the three
above-mentioned directions are filled by breccia ores. As a whole, the miner-
alized zone can be considered as a zone of intense pre-ore brecciation and joint-
ing favourable for development of ore-metasomatic processes and concentrated
ore deposition.
The intensity of hydrothermal alteration of the host rocks decreases from the
centre to the periphery of the deposit. An inner zone of strong silicification, an
278 M. Zhamsran et al.

intermediate one of sericitization and silicification and an outer one of chloritiza-

tion have been distinguished. In the zone of chloritized rocks, early K-feldspar
alteration is observed, whereas in other metasomatic zones the K-feldspars are
destroyed by later, superimposed metasomatic processes. Relicts of K-feldspar
granitoids create a discontinuous fringe around the zone of quartz-sericite meta-
somatites. Hydrothermally altered rocks are characterized by different ore-
bearing features. The copper and molybdenum contents in different types of
metasomatites are respectively: strongly silicified rocks (not more than 0.11 "70 Cu
and 0.009% Mo), moderately silicified and sericitic ones (0.4 - 0.8% Cu and
0.022 - 0.026% Mo), slightly silicified and sericitic ones (0.2 - 0.3% Cu ·and
about 0.01 % Mo). In fresh and poorly chloritized rocks, copper and molyb-
denum are present in negligible amounts. The ore body structure is complicated
by post-ore dikes of syenodiorite, andesite and trachyandesite porphyry.
According to Krivtsov et al. (1981), the Erdenetuin-Obo deposit by relation-
ship between the ore body and the porphyry intrusion, are assigned to the group
of disconformable deposits the formation of which is supposed to have taken
place in a subsurface environment. The ore-magmatic system is considered to
have been open and to have been developed at low pressure and temperature as
well as in a changeable deformation pattern. Here, ore deposition corresponds to
the period when the porphyry stock was practically crystallized. The authors
believe that in the present case the ore-generating portion of the ore-magmatic
system was deeper than the level of the intrusion and crystallization of a
porphyry stock. The latter give rise to the possibility of mineralization occurring
at several levels. Under these conditions the ore-controlling role of tectonic
dislocations is clearly shown to be a part of the regional structures.
Structural factors in the distribution and localization of ore mineralization
within the framework of the Erdenetuin-Obo deposit are also determinants for
most prospects in the northern part of the Erdenet-Tzagansuburgin structure.
These prospects show in miniature the structural conditions of ore formation
that are characteristic for certain parts of the deposit discussed. The degree of
development of ore mineralization very much depends on the intensity of occur-
rence of pre-ore-concentrating and ore-controlling structures. In general, the
combination of transverse northwest (north-northwest at a later stage) disloca-
tions and west-southwest faults consistent with the general trend of the volcanic
belt are the decisive factors for the formation of ore concentrations for all
prospects in the Erdenetuin-Obo ore field as well as in other areas that make up
the system of the Erdenet-Tzagansuburgin ore-concentrating structure. It should
be noted, in particular, the channeling role of transverse ore-concentrating
structures in providing loci for the upward flowing ore fluid allowing a level
favourable for the formation of large volumes of ores.

Acknowledgments. The manuscript was reviewed by Drs. Richard H. Sillitoe and K. F. Clark whose
comments have greatly improved the work and whose contributions are gratefully acknowledged.
Elements Determining in the Geological-Structural Model 279

References

Geology of the Mongolian People's Republic, vol II (1973) In: Khasin RA, Borsakovskii YuA,
Zonenshain LP (eds) Magmatism, metamorphism, tectonics. Nedra, Moscow, 750 pp (in Russian)
Gretskaya TA, Mossakovsky AA (1969) Some questions of the stratigraphy and structural relation
between the Carboniferous and the Triassic in northern Mongolia. Akad Nauk SSSR Izv Ser Geol
2:56 -71 (in Russian)
Khasin RA, Marinov NA, Khurts Ch, Yakimov LI (1977) The Erdenetuin-Obo copper-molybdenum
deposit in northern Mongolia. Geol Rudn Mestorozhd 6:3 -15 (in Russian)
Krivtsov AI, Migachev IF, Shishakov VB (1981) Morphogenetic type designs of ore bodies at copper
porphyry deposits. SOy Geol 10:28 - 41 (in Russian)
Mossakovsky AA, Tomurtagoo 0 (1976) The Upper Paleozoic of Mongolia. Sovmestnaya Sovetsko-
Mongol'skaya Geologicheskaya Ekspeditziya, Trudy, vol 15. Nauka, Moscow, 126 pp (in Russian)
Sotnikov VI, Berzina AP, Zhamsran M, Myagmar L (1980) The Tsagan-Suburga molybdenum-
copper deposit (MNP). Geol Rudn Mestorozhd 3:34-46 (in Russian)
Yarmolyuk VV, Kovalenko VI (1981) The role of largest faults in the location of continental volcanic
rocks of MNR. Akad Nauk SSSR Doklady 256,2:436-440 (in Russian)
Genetic Aspects of the Recsk Mineralized Complex,
Hungary
Cs. BAKSA t

Abstract

The following conclusions can be drawn concerning the genesis of the Recsk
deposit:
1. The mineralization belongs to the Paleogene volcanic arc along the Balaton-
Darno line.
2. The main phases of mineralization, culminating with porphyry copper
formation, are products of hydrothermal events related to a diorite porphyry
intrusion (a3 stage).
3. The two younger (post-intrusion) volcanic cycles have resulted in late-stage
near-surface mineralization, which is considered to be a product of remobil-
ization.
In comparison with other porphyry copper mineralization it may be suggested
that the Recsk deposit has an affinity with the island-arc-type magmatism. It
shows similarities with the diorite model in the zoning of the intrusive rocks and
the ore distribution. Morphogenetically it may be termed a "conformable"
deposit. Its classification was simplified by the completeness of the ore deposi-
tional sequence, although the sequence of superimposed ore-forming stages
created difficulties for interpretation.

Introduction

More than 130 yrs have elapsed since exploration and mining commenced in the
Recsk ore district. Numerous studies and papers dealing with several genetic
aspects of the Recsk and Paradfiirdo ore occurrences have been published in this
period (Panto 1951) A new chapter in the history of Recsk exploration opened in
1958 with a deep drilling program, the results of which were summarized in 1975
(Baksa 1975a, b, Cseh Nemeth 1975, Csillag 1975, Csongradi 1975) (Fig. 1).
The common factor of genetic interpretation in these papers was the ac-
ceptance of a relationship between the igneous activity and the repeatedly

1 National Ore and Mineral Mining Co., 1062 Budapest, Hungary

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Genetic Aspects of the Recsk Mineralized Complex, Hungary 281

activated Darno megastructural zone (Zelenka 1973, 1974). This ideas was
reflected by the paleogeographic reconstructions, suggesting a genetic link
between the different types of mineralization and the Eocene Priabonian tectonic
evolution along this zone. The Darno zone merges with the Balaton line south-
eastward. Several other volcanic areas are known along this line, among which is
the NE part of the Velence Mts., where the igneous complex shows age relations
and chemistry similar to Recsk. All these occurrences are situated on the
northern boundary of the Igal-Bukk eugeosyncline (Wein 1969, 1978), in the
Central Mountains belt and include different types of mineralization.
An updated synthesis of this volcanic belt was given by Csillag et al. (1980),
who postulated the uniform island-arc nature of these Paleogene volcanic occur-
rences. This volcanic arc and other earlier volcanic zones in the Carpathian
system (Banat-Timok area in Rumania and Yugoslavia, Vardar zone in
Yugoslavia) can be interpreted uniformly, based on the changes in the con-
vergence style of the Africa and European plates and the several microplates
between them (Radii et al. 1977). This approach also explains the similiar miner-
alization and petrologic features of these deposits, and the relative age dif-
ferences.

Geological Setting of the Recsk Deposit

Recently several new aspects concerning the relationship between the tectonic
evolution of the Darno zone and the Recsk ore deposit were revealed (Balla et al.
1980; Zelenka et al. 1983). Field observations in the SW part ofthe Bukk Mts., in
the Darno area and evaluation of drill core from Recsk have called into question
the earlier theories concerning the links between the Central Mountain belt and
the Recsk basement. The Mesozoic rocks at Recsk were in a similar manner
overturned and folded, and exhibit similar lithological properties to the Mesozoic
on the SE side of the Darno zone. Thus Zelenka and co-workers (1983) have
concluded that the Darno zone was not a paleogeographic boundary in the
Mesozoic period, as was previously stated.
The age of the igneous activity can be quite accurately identified from both
stratigraphic and radiometric data. The limestones and marls beneath and on top
of the volcanic sequence belong to the Nummulites fabianii horizon, of the
Priabonian (Upper Eocene) stage (determination by Baldi). The KI Ar age of
relatively fresh andesite from the Recsk Andesite Formation falls in the
34.9-35.7 (±2.5-4.9) m.y. range (determinations by Balogh).
The first volcanic products were the submarine effusive ( + to a lesser extent
explosive) products assigned to the a2-type andesites (this is a biotite-hornblende
andesite of the first stratovolcanic stage in local, informal mine terminology),
which accumulated on the Triassic basement, or on earlier Priabonian sediments
(Fig. 2). The a2 andesites are barren, no post-volcanic hydrothermal activity is
linked to this stage. The ore indications in these volcanics are later products. The
effusive centre of the a2 andesites is not known, only the extent of the volcanics
can be contoured. This coincides with the position of the diorite porphyry
282 Cs. Baksa

-------- ... - ... " ,,

,,
.,. ,,
,
/-4 " ,,
,,
,,
"
,,
,,
" ,,
'-,
...........,
,,
,,
,,
\
",,I

,- ... ~

l[[ill 4 r-::::,J
29 s[ZJ
31:;: :::::1 60 ... - .. ----------
a

Bukk MIs.

t
N

c a'-,_ 10km
_ _ _ _..JI

Fig. la-c
Genetic Aspects of the Recsk Mineralized Complex, Hungary 283

1=35 ·106 years

"Sl. sea level
1. voican ic phase

E3 and esi Ie oc::::::=:: Ii meslone

---------------------------~----
limeslooe ::::=====:::::~

Triassic

basement
sediments

,... - ..... ....

/ secondary',
/magma chamber ,

Fig. 2. The first phase of the Priabonian (Upper Eocene) biotite-amphibole andesite volcanism,
intruding and accumulating on the Triassic basement

intrusion, which comprises the second magmatic phase (Fig. 3). It can thus be
that the a3-type diorite (this is local, informal mine terminology also) porphyry
stock was emplaced by the partial or total destruction of a secondary magma
chamber and accompanying vent for the a2 effusives, and by the assimilation of
an equal volume of Triassic host rocks. The a3 intrusive body is 3 km long and
600 - 800 m wide, with an elongate shape in a N - S direction. It follows the
morphology of the uplifted basement horst. The a3 intrusive body is litho-
logically complex.
The chemical and lithological properties of the a3 intrusives are similar to the
diorite porphyries of eugeosynclinal belts (trench). This is in agreement with the
setting described in the island-arc reconstruction of Csillag et al. (1980). Thus
these characteristics (chemical and lithological properties) and mineralogical
parameters (alteration and zonation) link the Recsk occurrence to the diorite
model of Hollister (1978).
The Triassic wall rocks were both mechanically and chemically affected by a3
intrusive emplacement. Due to the destruction by later processes, only scarce
data are available for the alteration event synchronous with a3 emplacement.
Typical hornfels are not abundant. The movements within and adjacent to the
intrusion, which created the fracturing necessary for the later flow of hydro-
thermal fluids, are related to the cooling history of the system below 800°C.

Fig. 1 a-c. a The regional setting of the Paleogene volcanic arc in the Carpathian Basin. 1 Oceanic
lithospheric remnants of the Igal Biikk eugeosyncline belt; 2 the Paleogene volcanic arc; 3 Eocene
shelf and shoreline-facies marine sedimentation; 4 Cretaceous-Paleogene flysh zone; 5 Balaton-
Darno line; 6 the direction of Eocene marine transgression and volcanic activity (after Csillag et al.
1980). b Space and time relationships of the Paleogene volcanism indicating the ages of the first vol-
canic products: L = Lower Eocene; M = Middle Eocene; U = Upper Eocene (after Csillag et al.
1980). c Tertiary volcanics of the Matra Mts. 1 Basement formations; 2 Upper Eocene biotite-horn-
blende andesites; 3 Miocene volcanics
284 Cs. Baksa

Fig. 3. Idealized section of the diorite

porphyry intrusion that ascended
from the secondary magma chamber
during the second magmatic cycle

intrusion

Sr 8718~0 70 7
I

ROCK ALTERATION

(present ti me
a bove sea level)
-7oom

Ma Ma
T= 400 -600"C

Fig. 4. Alteration zoning in the diorite porphyry intrusion. Q = quartz; Ah = anhydrite; Bi = biotite;
Ma = magnetite

Localized stresses developed by the conversion of heat to mechanical energy, and

produced zones of tension, micro fracturing and brecciation. These zones
provided favourable loci for alteration and mineralization.
The most typical alteration, which is zoned, is plotted on Fig. 4. The altera-
tion was studied by Csillag (1975). A 150 - 200 m wide skarn zone has developed
on the contacts of the intrusion below -700 m (above sea level), and consists of
calc-silicate assemblages, which were formed during contact metasomatism.
Both endo- and exoskarns can be distinguished. The zonal pattern, with certain
generalizations, is similar to the diorite model of island-arcs (Hollister 1978).
Genetic Aspects of the Recsk Mineralized Complex, Hungary 285

pH ~ 7

H~ No: cl; NOj. ~5, OW

:t. Q,:;l ,,' -2
CU, Pb,Zn,Fe ,Mo,Au , Ag , Cd ,Re + HF, 5U
i" , l ' CUj , SO"

Fig. 5. Hypothetical compositions and flow paths of the fluids related to the cooling of the intrusion
and the physicochemical parameters of the system

Q 2
---------=----'- ~

Fig. 6. Vein-type and irregular hydrothermal replacement polymetallic ore deposits in the Triassic
wall rocks of the intrusion. The first stage of hydrothermal ore deposition

Ore Formation

Ore formation began with the emplacement of the a3 diorite porphyry intrusion.
The alteration processes, which were related to the cooling intrusion, first
affected its peripheral zones and the adjacent wall rocks, and progressively
invaded the inner zones of the intrusion. Later alteration phases were super-
imposed on the earlier ones, and a complex system was developed.
The space-time model of the Recsk deposit can be summarized as follows:
A. Pre-hydrothermal stage:
1. Contact-metasomatic magnetite deposits of the skarn zone (Fig. 4).
 -----
286 Cs. Baksa

---c::::::::::::::=-

Fig. 7. Hydrothermal replacement polymetallic and pyrite deposits on the outer boundary of the
skarn zone, along lithological contacts in Triassic wall rocks. The second stage of hydrothermal ore
deposition

--------------------===::~-----

Fig. 8. Hydrothermal replacement chalcopyrite-pyrite deposits in veins, lenses and pods in the skarn
zone, replacing calc-silicates. The third stage of hydrothermal ore deposition

B. Deposits related to hydrothermal activity of the diorite porphyry intrusive:

1. Hydrothermal vein-type and replacement Pb - Zn - Cu - Fe deposits (Fig. 6).
2. Hydrothermal stratabound replacement Cu - Zn - Fe and pyrite deposits
(Fig. 7).
3. Hydrothermal replacement Cu - Fe deposits in the exo- and endoskarns of the
skarn zone (Fig. 8).
4. Porphyry-type Cu - Fe - Mo mineralization of the intrusion (Fig. 9) .

C. Deposits emplaced in the late stages. Due to remobilization effects of later

volcanic phases:
Genetic Aspects of the Recsk Mineralized Complex, Hungary 287

From this sloge ~-t&

\~
- 7oom

Fig. 9. Hydrothermal disseminated and vein-type Cu - Mo deposits (porphyry copper), formed in

stock works in the intrusion. The fourth stage of hydrothermal ore deposition. Also shown are the
relative timing (t1 (2 () (4) of the formation of the main paragenetic types in the whole complex in this
stage

Fig. 10. Hydrothermal polymetallic ore deposits in silicified zones produced during the eruption
phase of the a1 q type biotite-hornblende andesite as the second phase of stratovolcano activity. The
fifth stage of hydrothermal ore formation

1. Hydrothermal metasomatic Cu - Fe - Pb - Zn - As - Ag ores related to

silicified zones in the at q phase (local, informal mine terminology of the
second stratovolcanic stage, Fig. 10).
2. Hydrothermal stockwork, massive sulphide and volcanic-sedimentary
Cu - Fe - Au - Ag - Sb - As ores related to the at phase (this is a hornblende-
288 Cs. Baksa

LAH6cA.HILL

Fig. 11. The Cu - Fe - Au - Ag - As - Sb - Se - Te volcano-sediment deposits generated by the al-

type biotite-hornblende andesite eruptions during the third phase of stratovolcano activity. The sixth
stage of hydrothermal ore formation . The post-ore ua-type pyroxene-biotite-hornblende andesite is
the fourth phase of stratovolcano activity

biotite andesite of the third stratovolcanic stage in the local, informal mine
terminology (Fig. 11). The iia phase on Fig. 11 is the fourth stratovolcanic
stage without mineralization).
These phases (al q, al and iia) are post-intrusion stratovolcanic stages.

This sequence of events indicates the temporal evolution and genetic develop-
ment at Recsk: the different stages are depicted in Figs. 4 to 11. As stated above,
the mineralization comprises a number of different, superimposed parageneses,
which developed as functions of continuously changing heat fluxes of the igneous
phases, which, in turn, caused circulation of fluids. In the hydrothermal stage
related to a3 intrusion, mineralization adavanced from the outside of the intru-
sion inward. In the subsequent volcanic phase, the formation of the mineraliza-
tion was governed by local, though similar, PTand chemical factors.
The temperatures of the various mineralization stages (Figs. 4 to 11) were
taken from the results of decrepitation studies (determinations by Csillag and
Kovach).
The last and most important mineralization event related to the diorite por-
phyry intrusion was the porphyry copper stage. During this stage, the minerals
filled stockworks in the diorite porphyry. The ore-grade contours and the altera-
tion zone boundaries closely follow the retreating path of the thermal fronts,
which generated the fluid circulation. The structural parameters of the rocks (i.e.
fracturing in the intrusion, lithological boundaries, and pre-ore fissures in the
sedimentary wall rocks) have played on important role in the shaping of these
Genetic Aspects of the Recsk Mineralized Complex, Hungary 289

contours. The grade contours of the porphyry mineralization are not related to
lithological parameters. Two types of the porphyry mineralization can be dis-
tinguished according to their shape. "Conformable" deposits are those in which
grade contours are subparallel to the shape of the intrusion, whereas "discon-
fonnable" deposits are those in which the symmetry axis of the ore body does not
fit with that of the intrusion. Cu and Mo distributions coincide in the discon-
formable type, but differ in the conformable type. The Recsk porphyry copper
mineralization shows the following characteristics:
1. The grade contours are subparallel to the geometry of the intrusion;
2. Cu and Mo show a negative correlation;
3. The centre of the Mo reserves coincides only in part with the centre of the Cu
reserves in the explored part of the intrusive body.
Comparing these data with the classification of Popov (1977), the Recsk
deposit to belongs to the category of conformable deposits in a morphogenetic
aspect. The formation of this geometry is promoted by the covered and quasi-
closed nature of the system.
The detailed description of ore-genetic data and the parageneses of each ore-
forming stage is beyond the scope of this study. Csongradi (1975) and Baksa
(1975a, b) have discussed the paragenetic sequences in the deep-level and near-
surface mineralization, respectively.

Acknowledgments. This study would have been impossible without the constructive assistance and
cooperation of geologists working on the Recsk exploration for more than 10 yrs. Special thanks are
due to J. Foldessy who helped with separation of the different magmatic cycles, and to J. Csillag for
providing laboratory results and valuable discussions concerning the study.

References

Baksa Cs (1975a) New data on the enargite-luzonite-pyrite massive sulphide deposits, North from
Lahoca-Hill, Recsk, Hungary. Foldtani Kozlony, Bull of the Hungarian Geol Soc 105:58 -74 (in
Hungarian)
Baksa Cs (1975b) The subvolcanic andesite body of Recsk and its dikes. Foldtani Kozlony, Bull of the
Hungarian Geol Soc 105:612 - 624 (in Hungarian)
Baksa Cs, Cseh Nemeth J, Foldessy J, Zelenka T (1980) The Recsk porphyry and skarn copper
deposit, Hungary. International Symposium Belgrade. In: Jankovic S, Sillitoe RH (eds) European
Copper Deposits; proceedings of an international symposium. Special Publications - Society for
Geology Applied to Mineral Deposits (SGA) 1:73 - 76
Balla Z, Baksa Cs, Foldessy J, Havas L, Szabo I (1980) The tectonic setting of the ophiolites in the
Bukk Mts., Hungary. Geologica Carpathica 31 4:465 - 493
Cseh Nemeth J (1975) Deep-seated base metal ore occurrence of Recsk: geological pattern of ore
accumulation. Foldtani Kozlony, Bull of the Hungarian Geol Soc 105:692-708 (in Hungarian)
Csillag J (1975) Rocks transformed upon magmatic effect in the Recsk area, Hungary. Foldtani
Kozlony Bull of the Hungarian Geol Soc 105:646-671 (in Hungarian)
Csillag J, Foldessy J, Zelenka T, Balazs E (1980) The plate tectonic setting of the Eocene Volcanic
Belt in the Carpathian Basin. In: Bisztricsanyi E, Szeidovitz S (eds) Proceeding of the 17th
Assembly of the European Seismological Commission. Budapest, publishing House Hung Acad
Sci 1:589-599
290 Cs. Baksa: Genetic Aspects of the Recsk Mineralized Complex, Hungary

Csongradi J (1975) Characterization of the deep-seated base metal ore mineralization of Recsk on the
basis of ore-microscopic analyses. Ftlldtani Ktlzltlny, Bull of the Hungarian Geol Soc
105 :672 - 691 (in Hungarian)
Ftlldessy J (1975) The stratovolcanic andesite formation of Recsk. Ftlldtani Ktlzltlny, Bull of the
Hungarian Geol Soc 105:625 - 645 (in Hungarian)
Hadii E, Aleksic Y, Pantic N, Kalenic M (1977) The plate movements in South-Eastern Europe
during the Alpine cycles. In: Jankovic S et al. (eds) Metallogeny and Plate Tectonics in the North-
eastern Mediterranean; proceedings of an international IGCP symposium 1:231- 248. Inter-
national Symposium, Belgrade
Hollister YF (1978) Geology of the porphyry copper deposits of the Western Hemisphere: New York,
Soc Mining Engineers AIME 219 p
Krivtcov AI (1977) Types of the regions of porphyry copper ore mineralization. Geologija Rudniih
Mesztorozsgenij XIX-4:44 - 57 (in Russian)
Lowell JD, Guilbert JM (1970) Lateral and vertical alteration-mineralization zoning in porphyry ore
deposits. Econ Geol 65:373 - 408
Panto G (1951) Ore deposit of Lahoca at Recsk, Hungary. Ftlldtani Ktlzltlny, Bull of the Hungarian
Geol Soc 81 :146 -152 (in Hungarian)
Popov YSz (1977) Geology and genetic model of porphyry copper deposits. Nauka, Moscow (in
Russian)
Sillitoe RH (1972) A plate tectonic model for the origin of porphyry copper deposits. Econ Geol
67:148-197
Sillitoe RH (1973) The tops and bottoms of porphyry copper deposits. Econ GeoI68:799-815
Wein Gy (1969) Tectonic review of the Neogene covered areas of Hungary. Acta Geol Acad Sci Hung
13:399-436
Wein Gy (1978) Development of Carpathian Basin. General Geological Review, Issued occasionally
by the Section for General Geology of the Hung Geol Soc 11:5-28 (in Hungarian)
Zelenka T (1973) New data on the Darno megatectonic zone. Acta Geol Acad Sci Hung 17:155 -162
Zelenka T (1974) Megatectonic and magmatogenic history of the Northeastern Matra Mountains.
Acta Geol Acad Sci Hung 18:377 - 385 (in Russian)
Zelenka T, Baksa Cs, Balla Z, Ftlldessy J, Ftlldessyne JK (1983) The role of the Darno line in the
basement structure of NE Hungary. Bratislava, Geologica Carpathica 31.4:53 - 69
Ore-Magmatic Systems of Copper-Molybdenum Deposits
V. I. SOTNIKOV and A. P. BERZINA 1

Abstract

Ore deposition concludes a complex cycle of transport and evolution of materials

throughout the crust (20 - 30 km thick) where magmatic and hydrothermal
associations, derivatives of a single ore-magmatic system, are being formed. Ore
formation takes place in or near porphyry stocks considered as cupolas of
intermediate magmatic chambers, the crystallization of which markedly
influenced the formation of the flow of ascending fluid as well as a total
temperature rise. The zone of magmatic cupolas (stocks) is characterized by more
intensive ascending fluid flow. Here, favorable conditions for magmatic-
hydrothermal systems exist. Conditions of mineralization in the hypabyssal
environment become more complex because of the anatectic chamber, the
crystallization of which causes an additional fluid flow into the upper levels of
the crust. The development of the ore-magmatic system results from the
interaction of subcrustal mafic magma and fluids separating from it with a
magmatic melt formed in the crust. Fluids of two types existed in forming
copper-molybdenum deposits. The magmatic system includes concrete intrusive
bodies of an ore-bearing magmatic complex and the rock volumes adjacent to
them. The meteoric-hydrothermal fluid system does not evolve in close
juxtaposition with these intrusives.
Porphyry crystallization took place within the 1230- 800°C temperature
range, explosive breccia formation from 750-600°C to 300-200°C, hydro-
thermal alteration and mineralization from 730°C to 200°C.
Nowadays complex copper-molybdenum deposits with large reserves of
relatively low-grade ores offer one of the main sources of copper and
molybdenum as well as of a number of accompanying rare and precious metals.
That is why both theoretical and practical interest in these deposits exists all over
the world.
As to the systematics of copper-molybdenum deposits, the authors suggest
the use of ore-association analysis. The concept of ore association is widely
applied in Soviet geology and is considered by investigators to be any group of
deposits related in the common mineral composition and conditions of their
formation. An ore association is characterized by specific geologic-genetic

t Institute of Geology and Geophysics, Siberian Branch of the USSR Acad. Sci., Novosibirsk, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
292 V.1. Sotnikov and A. P. Berzina

features common to given groups of deposits (Kuznetsov 1972, Konstantinov

1973, Shcheglov 1980). The main task of the analysis is to elucidate the principal
features of ore mineralization that enable one to determine the particular ore
association and its boundaries and to demonstrate that it has been correctly
established. In the USSR, for the last 20 - 30 yrs, the ore-association analysis has
developed for both the scientific and practical study of deposits and has in-
fluenced the development of theoretical principles of metallogeny and prognosis.
In contrast to the ore-association analysis, other systematics are based on
separate features of mineralization: geotectonical position (Sillitoe 1972, Pavlova
1983, Pokalov 1972, Krivtsov 1977), chemical features of ore-bearing
magmatism (Westra and Keith 1981, Mutschler et al. 1981), geochemical profile
of mineralization, and accompanying alteration (Rekharskii and Distler 1965).
One can see from these works that the deposits, different in some features, are
similar in others. That is why copper-molybdenum deposits are considered by
some researchers as one group, but are distinguished by others into porphyry-
copper and porphyry-molybdenum groups.
Revealing in some cases essential differences among each other, on the whole,
all copper-molybdenum deposits show rather stable common geologic-genetic
and physicochemical parameters. Their differences are of regional appearance
and are attributed to the specific features of development of the ore-magmatic
systems that depend on the type of geologic setting, the location of the magmatic
chamber, the relationship among the different sources of fluids and metals, the
patterns of the fluid-transporting zone, hydrothermal alteration, ore deposition,
and other parameters. That is why all deposits in question should be grouped into
one class - copper-molybdenum association that is subdivided into molybde-
num, copper-molybdenum, and copper subassociations (Sotnikov 1966, Sotni-
kov et al. 1977, 1985).
In this paper, the development of ore-forming, particularly the conditions of
formation of large-volume ore concentrations, are considered on the data
obtained in the study of copper-molybdenum deposits in the principal ore
districts of the USSR and Mongolia and also on the basis of the generalizations
advanced in many publications on these problems. The authors have taken into
account a variety of the geologic situations in which ore mineralization occurs,
different ages, and the geochemical distinctive features of the deposits. The prin-
cipal assumptions concern the copper-molybdenum association as an integral
unit.
Copper-molybdenum deposits are located mainly along regional structures
referred to as superimposed ore belts, which are related to the evolution of extra-
geosynclinal tectonic-magmatic processes and on the whole controlled by
volcano-plutonic belts. Location of ore districts within these volcano-plutonic
belts is controlled by the intersection of regional tectonic structures, coinciding
with the direction of volcanic belts, and large through-going zones of transverse
dislocation. These extend far downward (actually into the mantle), such zones
create great possibilities for the connection of the ore-forming process with deep-
seated layers of the Earth. Further, they offer very efficient permeable zones to
permit the ascent of mantle material and the accumulation of fluid flow up to a
level sufficient for intense magmatic and ore-bearing processes to develop. These
Ore-Magmatic Systems of Copper-Molybdenum Deposits 293

belts contain centers of repeated endogenous activity where magmatic rocks of

various ages, morphology, and genesis have been emplaced.
Copper-molybdenum deposits refer to the class of magma-related deposits
associated with the volcano-plutonic series. The ore mineralization shows close
spatial and temporal relation to porphyry magmatism (stocks and dikes).
Because of their small sizes, difference in age, and a certain independence with
respect to large host plutons. porphyry bodies generally cannot both energetically
and by volume of separating fluids account for the formation of large-scale
copper-molybdenum deposits. The latter seem to be possible only if porphyries
are an integral part of larger magmatic systems, developed over a long period of
time. All these points are taken into account in interpreting the relationship of
mineralization with porphyry magmatism. Ore deposition concludes a complex
cycle of transport and evolution of materials throughout the crust 20- 30 km
thick where magmatic and ore-metasomatic rocks, derivatives of a single ore-
magmatic system, are being formed (Fig. 1). Porphyry stocks are supposed
(Kalinin et al. 1980) to be cupolas of intermediate magmatic chambers, the crys-
tallization of which markedly influenced the formation of ascending fluid flows
as well as providing a total temperature rise in the zone of ore deposition.

i km -

10 km -

Fig. 1. The ore-magmatic system of cop-

per-molybdenum deposits. 1 Ore-con-
centrating structure of high permeability;
2 primary anatectic magma chamber; 3
intermediate magma chamber; 4 porphy-
ry; 5 explosive breccia; 6 fluid flow (from
mantle and magma chambers); 7 meteor-
rrTl-s ~-7 r:7l-a 1:::.::::::1 ic water; 8 exhausted hydrothermal solu-
LL.lJ ~ IL-J "' '.- -9 tions; 9 ore-bearing zone
 294 V. I. Sotnikov and A. P. Berzina

Magmatism developed in high permeability structures (under the conditions of

drastic decrease in external pressure) are considered to be extremely favorable for
the separation of late- and post-magmatic ore-bearing fluids. Conditions of min-
eralization in the hypabyssal environment become more complex because of the
presence of a primary anatectic magmatic chamber, the crystallization of which
during the regressive period of the development of the magmatic system causes
the production of additional fluid in the upper levels of the crust.
According to the mathematical model of a magmatic-hydrothermal system
with magmatic chambers at different levels (Fig. 2a), the zone of the magmatic
cupolas (stocks) are characterized by more intense ascending flows especially in
the early stages of crystallization. Here, favorable conditions for magmatic-
hydrothermal systems are being formed. Near-contact zones of the cupolas are
noted to show the maximum temperature gradients (Fig. 2b) that are suitable for
the development of the ore-bearing process. After the melt crystallized, the
stocks themselves exhibited high-contractional porosity (especially in the periph-
eral zones), and they can be treated as structures draining any ascending flows.
Metasomatic processes play an important role in forming high-potassium por-
phyries under the conditions of high-fluid saturation of an ore-magmatic system
(Artyshev and Vasil'ev 1982, Sotnikov et al. 1983, Spatz 1979). Usually such
metasomatic processes are characteristic of ore-magmatic systems with elevated
activities of potassium and fluorine.

a b
0 /0 0 1O
km /em

100·

+ + +
+ + + + '100-
+
+

++++
++- ...
+- + + +-
+- +- +- +-
+ + + + + ++-
+++-to++++
+- + +- + -to + +<»
10 +- + + +- + + + l1l
km
t t ttl t On E:::!J 1 fIIIJ z I-t +13 c::=J 4
Fig. 2a, b. Model of magma system (a) and temperature distribution in supra-intrusive zone of an
intermediate magmatic chamber (b) . Q/1 (the magnitude of mantle fluid flow) = 10- 8 g/(cm 2 s-I).
The intermediate chamber is 2.5 km high and 6 km wide. Melt crystallization temperature, 800°C.
Before formation of the intermediate magmatic chamber, the heat flow was 1.25· 10- 6 call(cm 2
S-I). The temperature distribution is shown at 10000 yrs after formation of the intermediate
magmatic chamber. 1 Intermediate chamber and its cupolas (stocks of porphyries); 2 deep zone of
partial melting; 3 granitoid pluton hosting porphyry stocks and mineralized rocks; 4 volcanic-
sedimentary and metamorphic rocks
Ore-Magmatic Systems of Copper-Molybdenum Deposits 295

At present, rather abundant isotopic data are available (Taylor 1982, Tugari-
nov et al. 1974, Ford and Green 1977, Johan and Le Bel 1980, Sotnikov and
Berzina 1979), which indicate that crust and mantle matter take part in different
quantitative combinations in the ore-magmatic processes of copper-molybdenum
deposits. These data, along with the often observed occurrence of more mafic
rocks in the early stages of development of the ore-bearing magmatic complexes,
suggest that the ore-magmatic system resulted from interaction of subcrustal
mafic magma and fluids separating from it with a magmatic melt being formed in
• of potassium enrichment in the late rocks of the ore-
the crust. A clear trend
magmatic complexes and the common development of potassium metasomatism
are caused either by a continuous flow of deep fluids or by fluid interaction and
the redistribution of mobile components (potassium in particular) in the
boundary region between the alkaline-mafic and secondary silicic melts
(Dobretsov and Dobretsov 1983). These processes are most fully realized in the
zones of elevated and lasting tectonic-magmatic activities. Due to the
combination of differently directed deep-fault structures, favorable conditions
are created for the mantle fluids to penetrate the zone of magma generation
(Korzhinskii 1952, Perchuk 1973, Beus and Golubev 1979, Banks and Page 1980)
with intensive heating of the crust basement and fluid concentration; the deep
basic magma ascent to the crust being possible in the same zones.
Such characteristics as the geotectonic, lithologic, and geochemical, among
others, of certain units of the crust involved in the tectonic-magmatic processes
can cause certain differences in the nature of the volcano-plutonic series that
affect the development of the ore-bearing magmatic system, as well as the
geologic and geochemical features of the ore districts (Kalinin and Sotnikov
1982). To the west of Southern Mongolia, the copper-molybdenum-bearing
volcano-plutonic belt is emplaced in an oceanic domain and is represented by
andesite associations with subordinate monzonite-diorite-granodiorite ones. Ore
mineralization is essentially copper-rich (often auriferous) with slight quartz-
sericite alteration. High chlorine activity was registered especially at early stages
of the magmatic process. A "diorite" model (Hollister 1978) of the formation of
copper-molybdenum deposits appear to have been realized in these regions. In
activated old structures of essentially the mafic type located in Northern
Mongolia (Sotnikov et al. 1985) and Middle Asia, USSR (Pavlova 1978), the
portion of granitoids among magmatic rocks increases. Higher Mo/Cu ratios are
characteristic of the ores in these areas. Strong quartz-sericitic alteration is wide-
spread in the better mineralized centers. Development of an ore-bearing process
is considered to conform to a "quartz-monzonite" model (Lowell and Guilbert
1970). In the zones of activated old median massifs and platform margins, in-
cluding a prominent granite-metamorphic series (Kuznetskii Alatau, Trans-
baikalia), dacite-rhyolite-granodiorite, dacite-rhyolite-granodiorite and rhyolite-
granite associations, usually with significant predominance of plutonic rocks,
enjoy preferential development within the framework of volcano-plutonic belts
(Sotnikov et al. 1977). Mineralization of copper-molybdenum and essentially
molybdenum profile with the appearance of tungsten in a number of cases is
typical. K-feldspar alteration is dominant. The magmatic-hydrothermal process
is characterized by higher fluorine activity. In a number of cases these deposits
296 V.l. Sotnikov and A. P. Berzina

begin to acquire the features of greisen in edones. A "granite" model, essentially

applicable to molybdenum deposits (Clark 1972), and partly a "quartz-mon-
zonite" model are realized in these areas. In the zones of activation of old con-
solidated structures (the Aldan Shield) copper-molybdenum mineralization is
likely to manifest itself in connection with the development of alkaline mag-
matism (Kochetkov 1982).
Fluids of two types are suggested to have existed in forming copper-molyb-
denum deposits (Sotnikov and Berzina 1979, Sotnikov et al. 1983). The
magmatic system includes concrete intrusive bodies of an ore-bearing magmatic
complex and their adjacent areas (within the scope of a given fluid system, the
processes of explosive brecciation and those of high-temperature potassium
alteration usually take place). Meteoric-hydrothermal fluid systems evolve
outside a close spatial connection with these intrusives (to a great extent it is the
region of the development of sericitization, silicification, and chloritization and,
in many cases, of the concentrated ore deposition). In the former fluid system,
magmatic sources are sharply dominant (often with preferential juvenile com-
ponent participation). The latter is provided by the components of deep-seated
(including juvenile) and meteoric sources. In this system a meteoric part for a
number of components can dominate appreciably. In particular, in cases of wide
and intensive hydrothermal alteration, the host rocks can offer one of the prin-
cipal sources of ore-forming substances. Hydrothermal alteration occurring
within the limits of the influence of the meteoric-hydrothermal fluid system and
usually characterized with an exothermal effect (Sotnikov et al. 1977, Norton
and Cathles 1982) could appreciably contribute to the energy budget of the ore-
magmatic system, facilitating in a number of cases spontaneous development of
these processes. It should be noted that the zones of hydrothermally altered
rocks, which, in contrast to the initial (unaltered) ones, are characterized by
elevated values of thermo- and temperature conductivity coefficients and also by
higher permeability, can offer good heat conductors and a suitable medium for
moving fluids.
Ore formation in copper-molybdenum deposits (occurring in connection with
the development of pulsed multiphase porphyry magmatism) is considered to
have been a complex process of multistage development characterized by
repeated occurrence of ore-hydrothermal rocks. The ore-forming process, in
general, is characterized by a regressive development with a step-to-step tempera-
ture decrease and by the replacement of the oxide mineralization that is most
typical for the period of the early explosive brecciation and high-temperature
mineral deposition, by a sulfide one. At an early stage a large-scale "metasomatic
steaming" of the host rocks occurs, with huge volumes of disseminated mineral-
ization generated. During this preliminary stage the conditions potentially
favorable for the subsequent formation of the large-volume ore concentrations
are being created. Later on, during the main productive stage, mobilization of
disseminated metals takes place. Ore components also enter in the solutions gen-
erated by deep-seated zones of the ore-magmatism, the efficiency of which
depends greatly on the activity of CI complexes. The latter considerably influence
both the extraction and the transportation of ore components by magmatic
fluids. Insignificant redeposition of ore substance also took place during the later
Ore-Magmatic Systems of Copper-Molybdenum Deposits 297

Table 1. Some data resulting from an investigation of melt and fluid inclusions in feldspar, quartz,
apatite, and zircon a

Process State of
aggregation

Crystallization of ore- 1230- 800 Melt

bearing melt
Explosive brecciation 750-250 2500-1100 Gas
K-feldspathization 730-400 1800-1600 Mainly gas
Ore-formation 470-220 1500-600 Gas, liquid
Sericitization 410-330 Liquid
ArgiIlization 360-320 Liquid

a Generalized after Berzina and Sotnikov (1979).

b Determination of temperature of melt inclusion as described in Roedder (1979).
C Syngenetic inclusions of essentially water and essentially carbon dioxide, formed by boiling of
mixed H 20 - CO2 solutions and three-phase inclusions with halite were used for estimating the
pressure. See: Roedder and Bodnar (1980).

stages of the formation of the deposits in connection with final stages of ore-
bearing magmatism. Economically polyascendent deposits where multiple
magma intrusions (accompanied by explosive brecciation) took place, are the
most favorable for prospecting.
Due to the multistage development of ore-magmatic systems, the ore-forming
process in copper-molybdenum deposits is characterized by a complex evolution
of physicochemical parameters (Piznyur 1983, Sotnikov et al. 1979). Some of
these are shown in Table 1. The formation of porphyry intrusives of the ore-bear-
ing magmatic complexes took place within the 1200 - 800°C temperature range.
Feldspar and quartz phenocrysts crystallized within the 1230-1150°C and
111 0 - 865 ° C temperature range, respectively. The lower limit of homogeniza-
tion of the primary melt inclusions in quartz phenocrysts was found to be within
a narrow temperature range, i.e., 960- 950°C in most of the dike- and stocklike
intrusive bodies studied. Only in a few cases was the temperature lowered to
865°C. The homogenization temperatures of both secondary and pseudo-
secondary melt inclusions is usually lower than the minimum homogenization
temperature or primary ones; their lowest values nearing those of the highest
temperatures of homogenization of inclusions from explosive breccias and early
high temperature metasomatites.
Fluids of an early stage of intrusive activity that were fixed in primary melt in-
clusions of quartz phenocrysts are represented by nitrogen. In the composition of
fluids coexisting with the melt during the formation of secondary melt inclusions
along with nitrogen, a high content of carbon dioxide and halides is noteworthy
(Fig. 3). The maximum effect of these components in a fluid manifests itself later
during the accessory mineral crystallization with carbon dioxide and halides
dominating during zircon and apatite deposition, respectively.
Fluid-explosive breccias form mainly pipelike bodies composed of host rock
fragments. A cement is dominated by finely crushed and grated material of these
rocks. Newly formed minerals, with biotite and magnetite being prevalent, are
298 V.1. Sotnikov and A. P. Berzina

Fig. 3. Composition of the gas

phase of fluid inclusions in the
quartz of mineralized and al-
tered rocks. The curves show a
change of the gas composition
in a trend from high- to middle-
temperature rocks (from left to
right)

observed as well. As a result of later superimposed hydrothermal processes, the

cement of the breccias (to a lesser extent, the fragments) is altered. The tempera-
ture range of explosive-breccia formation varied from 750 - 600°C (in the early
stages) to 300 - 250°C (in the late stages). The highest pressures (registered
during the thermobarometric investigations) reached 2000 - 2500 atm. On the
whole, the early explosive brecciation occurred somewhat earlier than the forma-
tion of porphyry bodies. Violent degassing of the magma was accompanied with
the separation of the fluids enriched in halogens (mainly HCI and ferro-
chlorides). Breccias were formed as a result of multiple explosions; during
relatively quiescent stages (when the feeding channel was blocked), the mUltiple
extrusion of breccias ceased, and ejections consisted of finely crushed and
ground rocks. The pressure decrease following the ejection produced the hetero-
geneity of the solutions. The conditions favorable for the inflow of meteoric
waters from the surroundings were being developed.
The important role of chlorine during the process of early explosive breccia-
tion is the reflection of its geochemical history in a magmatic process connected
with its escaping from the magmatic chamber during the crystallization of the
melts. Magmatic melts could have been degassed both during crystallization and
emplacement of intrusives, when, with the existence of high permeable zones,
melts boiling up resulted in major fluid separation. The latter cases were more
favorable for the evolution of explosive brecciation as well as for the formation
of wide zones of disseminated ore mineralization. In a number of deposits,
carbon dioxide (along with halogens) were important components in the
composition of the fluids and accounted in large part, for the formation of the
explosive breccias (Fig. 3).
Ore-alteration rocks were generated in the broad temperature range:
730 - 600°C in the early period (K-feldspathization, albitization, early silicifica-
Ore-Magmatic Systems of Copper-Molybdenum Deposits 299

tion) to 450 - 200 °C (the main ore-forming stage as well as sericitization, sili-
cification, argillization) and still lower in the final stage. The pressure ranged
from 1500 to 100 - 80 atm. The solutions were as a rule, of alkali-chloride char-
acter and, at relatively low temperatures, alkali-chloride-carbonate composition.
High-temperature fluids were characterized by a gaseous state, although due to
the high pressure obtained they had the density of liquids. The medium- and low-
temperature mineralization was deposited as if from liquids. The fluids respon-
sible for the formation of high-temperature K-feldspathic rocks were charac-
terized by a large content of halogens. At lower temperatures, the contribution of
carbon dioxide and nitrogen in the fluid composition increased, the highest in-
crease occurring after the solutions became truly liquids. The latter as compared
to gaseous fluids usually contained higher concentrations of essentially all con-
stituents. Concentrated ore deposition took place from the solutions with a rela-
tively high carbon dioxide contribution.
Under conditions of relatively shallow depths and highly permeable struc-
tures, mineral-forming solutions boiled and as a result, the salt concentration in
the liquid phase increased. In a number of cases, typical brine-melts appeared.
The regions where highly concentrated solutions existed, a more concentrated ore
deposition was observed. Less concentrated solutions that migrated to the pe-
ripheral zones of the deposits (ore field), are of an essentially carbonate composi-
tion. Under favorable circumstances they could dump their ore loads to form
mineralization of a different geochemical profile (in particular, polymetallic and
gold-silver ores).
The development of ore-magmatic systems in ore fields of copper-molyb-
denum deposits is accompanied by a considerable concentration of ore substance
in a relatively limited volume; this predetermines the following most important
conditions of ore formation: the concentration of a fluid (hydrothermal) flow in
certain structures, responsible not only for the morphology, but also for the very
possibility of the formation of ore bodies; the presence of a mechanism of ore
deposition within a relatively narrow depth range.
It has already been emphasized that draining structures of different types and
orders play a significant role in determining the concentration of fluid flows.
Analyzing the conditions of the formation of large-volume veinlet-disseminated
ores, which mostly characterize copper-molybdenum deposits, it is necessary to
take into account that ore-forming solutions must be delivered to the region of
the widestspread jointing (to microjoints often not connected directly to each
other). Along with the explanation of the fluid mechanism developed by
Burnham (1982), one can use capillary transport, the significance of which
markedly decreases at temperatures above 300°C. Special conditions arising in
the process of the formation of stockwork copper-molybdenum deposits that fix
the elevated tectonic activity sites are also likely to stimulate the solutions
transfer into the region of relatively high temperatures and pressures. Here, the
relaxation of tension results in discompaction of the rocks and in the increase of
their volume through the appearance of tiny cracks and pores. The process of
mass relaxation of tectonic tensions results in the creation of an ultrasonic field
in which the capillary transfer of fluids rises drastically since it is no longer
limited by high temperature and pressures.
300 V.1. Sotnikov and A. P. Berzina

Thermo-osmotic transfer of the substance could widely develop in the region

of a meteoric-hydrothermal system under conditions of nonhomogeneous
heating, considerable humidifying of huge volumes of rocks, their variable
porosity, and due to the presence of the two thermodynamic phases of interstitial
water - the free and bound ones (Dudarev and Sotnikov 1982). Thermo-osmosis
presents the process of formation of heat slip flow of bound (film) water in a
temperature gradient field (Lykov 1954). This film liquid flow is usually turned
to a heater. Here, as result of evaporation, the thickness of the water film is de-
creased and the liquid can transfer from a thicker film into a thin one. Thus,
under nonuniform heating conditions in a porous environment, migration of a
substance from a bound water zone into a thermo-osmotic zone and further to a
heater takes place. Experimental work suggests· that along with film liquid
(water), alkalies and metals, including copper and molybdenum, are transferred
effectively due to this thermo-osmotic phenomenon. Transfer effect rises with
increase of porous medium dispersion. In this case, higher-temperature bound
water, when compared with lower temperature, is characterized by enhanced
transporting capability. A continuous thermo-osmotic flow (and what is of
paramount importance, the flow to the source of heating) can reach 500 - 600°C
even at a distance far enough from the region of heating granted a certain
amount of free fluid is available. This is comparable with the maximum tempera-
tures of the process of ore formation in copper-molybdenum deposits.
Ore deposition on the geochemical temperature barrier in the region of
maximum temperature gradients is considered as a possible mechanism of ore-
substance deposition, thus providing for the formation of large copper-molyb-
denum deposits. Computation of a similar model (Kalinin et al. 1981, Sotnikov
et al. 1983) revealed that the pattern, on the whole, is consistent with peculiarities
of ore formation in copper-molybdenum deposits. During the progressive stage
of development of a temperature field in the intrusive, the geochemical tempera-
ture of a temperature field in the intrusive, the geochemical temperature barrier
situated in the supraintrusive region at a distance from the contact (the ore-
substance concentration in the contact zone is fixed in case of highly concen-
trated solutions only) is gradually shifting upward and simultaneously growing
narrower. This is the time when the area (ore bodies) rich in copper and molyb-
denum are being formed and, on the whole, instead of being scattered in a large-
volume mineralization (by "metasomatic steaming") occur in a locally concen-
trated one. During the regressive stage, the geochemical temperature barrier is
lowered, and here late mineral associations are depositing. The latter seems to
account for superimposition of the late polymetallic and gold-silver mineraliza-
tion onto the early copper-molybdenum one of higher temperature. It has been
verified that economic concentrations in the geochemical temperature barrier
zone can be formed in the presence of .the fluid flow only. The presence of a
lateral flow of meteoric waters introduces certain complications into the pattern
of ore deposition. The lateral flow acts as an additional temperature barrier
facilitating the formation of a new zone of ore deposition in the lower portion of
the horizontally permeable structure in its point of intersection with the vertical
one. On the whole, the zone of lateral flow of meteoric waters showing minimum
solubility of ore components can offer a region of maximum ore deposition.
Ore-Magmatic Systems of Copper-Molybdenum Deposits 301

To conclude, it is necessary to point out that some elements of the ore-

magmatic system discussed have been investigated and evidence has been
produced to a lesser degree; for others, the evidence is much better, which is why
they are not equally reliable. The greatest difficulties naturally arise in the genetic
model for deep portions of the ore-magmatic system. And here we anticipate the
analysis of the dynamic models of endogenous processes on the basis of recent
geologic and geophysical evidence on deep-seated structures of the crust,
experimental and theoretical developments of both modern petrology" and the
theory of ore formation and use of the latest computational equipment.

Acknowledgments. The authors are grateful to Drs. Richard H. Sillitoe and Richard L. Nielsen for
their helpful reviews of the manuscript.

References

Artyshev SA, Vasil'ev BD (1982) Quartz-albite metasomatites of magmatic stage of median depth in
Kuznetsk Alatau. In: Metasomatism and ore formation. Tezisy dokladov 5th Vsesojuznoi Konf,
23 - 25 November, Leningrad, pp 33 - 34 (in Russian)
Banks NG, Page NJ (1980) Some observations that bear on the genesis of porphyry-copper deposits.
In: Ridge JD (ed) Proc 5th Q IAGOD Symp, vol I. Stuttgart, pp 49 -73
Berzina AP, Sotnikov VI (1977) Physicochemical conditions of endogene processes in copper-molyb-
denum deposits in Central Asia. Econ Geol 72:25 - 36
Beus AA, Golubev VS (1979) Geochemical model of evolution of continental lithosphere. Akad Nauk
SSSR Izv Seriya Geol 3:108 -121 (in Russian)
Burnham C Wayne (1982) Magmas and hydrothermal fluids. In: Barnes HL (ed) Geochemistry of
hydrothermal ore deposits. Mir, Moscow, pp 71 -121 (in Russian) (original edition in English)
Clark KF (1972) Stockwork molybdenum deposits in the Western Cordillera of North America. Econ
Geol 67:731 -758
Dobretsov GL, Dobretsov NL (1983) On problem of genesis of alkaline-salic rocks. Geol Geofiz
2:3 -11 (in Russian)
Dudarev AN, Sotnikov VI (1982) Thermo-osmotic filtration of solutions in rocks. Nauka, Novo-
sibirsk, 108 pp (in Russian)
Ford JH, Green DC (1977) An oxygen and hydrogen isotope study of the Panguna porphyry copper
deposit, Bougainville, Papua New Guinea. Geol Soc Aust J 24:63 - 80
Hollister VF (1978) Geology of the porphyry copper deposits of the Western Hemsiphere. Amer Inst
Mini Metall Petrol Eng, New York, 219 pp
Johan Z, Le Bel (1980) Geochimie isotopique du soufre et sa contribution a la comprehension de la
genese des mineralisations type porphyry cuprifere. Mem Bur Rech Geol Min 99:151-161
Kalinin AS, Sotnikov VI (1982) New aspects of associational analysis of endogenic ore deposits. SOy
Geol Geophys (Geol Geofiz) 23, 5:29 - 37 (in English)
Kalinin AS, Sotnikov VI, Berzina AP (1980) Hydrodynamic and temperature model of magmatogene
hydrothermal system of molybdenum deposits (modelling by means of electronic computer). Akad
Nauk SSSR Doklady 252, 5:1230 -1234 (in Russian)
Kalinin AS, Sotnikov VI, Kolonin GR, Berzina AP (1981) A model of geochemical temperature
barrier of copper-molybdenum deposit (modelling by means of electronic computer). Akad Nauk
SSSR Doklady 259, 4:962 - 965 (in Russian)
Kochetkov AYu (1982) On new type of porphyry copper mineralization (Aldan shield). Akad Nauk
SSSR Doklady 267, 2:443 - 445 (in Russian)
Konstantinov RM (1973) Principles of associational analysis of hydrothermal ore deposits. Nauka,
Moscow, 215 pp (in Russian)
Korzhinskii DS (1952) Granitization as magmatic replacement. Akad Nauk SSSR Izv Ser Geol
2:56 - 69 (in Russian)
302 V.1. Sotnikov and A. P. Berzina: Ore-Magmatic Systems of Copper-Molybdenum Deposits

Krivtsov AI (1977) Types of porphyry copper areas. Geol Rudn Mestorozhd 4:3 -20 (in Russian)
Kuznetsov VA (1972) Ore associations. Geol Geofiz 6:3 -14 (in Russian)
Lowell JD, Guilbert JM (1970) Lateral and vertical alteration-mineralization zoning in porphyry ore
deposits. Econ Geol 65:373 - 408
Lykov A V (1954) Transfer in capillary-porous bodies. Gostekhizdat, Moscow, 269 pp (in Russian)
Mutschler FE, Wright EG, Ludington S, Abbott J (1981) Granite molybdenite system. Econ Geol
76:874-897
Norton D, Cathles L (1982) Thermal aspects of ore deposition. In: Barnes HL (ed) Geochemistry of
hydrothermal ore deposits. Mir, Moscow, pp 481-496 (in Russian)
Pavlova IG (1978) Porphyry copper deposits. Nedra Leningrad, 275 pp (in Russian)
Pavlova IG (1983) Geologo-genetic models of porphyry molybdenum-copper deposits. In: Kuznetsov
VA (ed) Genetic models of endogene ore associations, vol I. Nauka, Novosibirsk, pp 127 -134 (in
Russian)
Perchuk LL (1983) Thermodynamical conditions of deep petrogenesis. Nauka, Moscow, 318 pp (in
Russian)
Piznyur AV (1983) An evolution model of ore-bearing fluids as the basis of local evaluation and
prediction of ore mineralization. In: Kuznetsov VA (ed) Genetic models of endogene ore associa-
tions, vol I. Nauka, Novosibirsk, pp 148 -153 (in Russian)
Pokalov VT (1972) Genetic types and prospecting criteria of endogene molybdenum deposits. Nedra,
Moscow, 271 pp (in Russian)
Rekharskii VI, Distler VV (1965) On feldspar-quartz association in molybdenum deposits. Geol Rudn
Mestorozhd 4:19-26 (in Russian)
Roedder E (1979) Origin and significane of magmatic inclusions. Bull Miner 102:487 -510
Roedder E, Bodnar RJ (1980) Geologic pressure determination from fluid inclusion studies. Annu
Rev Earth Planet Sci 8:263 - 301
Shcheglov AD (1980) Principles of metallogenic analysis. Nedra, Moscow, 430 pp (in Russian)
Sillitoe HR (1972) A plate tectonic model for the origin of porphyry copper deposits. Econ Geol
67:184-197
Sotnikov VI (1966) An experience of systematics of tungsten and molybdenum mineralization in the
Altai-Sayan geosyncline region. In: Kuznetsov VA (ed) Endogene ore associations in Siberia and
the Far East. Nauka, Moscow, pp 115 - 123 (in Russian)
Sotnikov VI, Berzina AP (1979) Fluid and metal sources of copper-molybdenum deposits. In:
Kuznetsov VA (ed) Nature of fluids and metal sources of endodene deposits. Nauka, Novosibirsk
pp 15 - 32 (in Russian)
Sotnikov VI, Berzina AP, Nikitina EI, Proskuryakov AA, Skuridin VA (1977) The copper-molyb-
denum ore-association. Nauka, Novosibirsk, 422 pp (in Russian)
Sotnikov VI, Berzina AP, Shugurova NA, Motorina IV (1979) Physicochemical parameters of the
processes of the generation of the copper-molybdenum association. In: Kuznetsov VA (ed) Main
parameters of nature processes of endogene ore formation, vol I. Nauka, Novosibirsk, pp
209 - 220 (in Russian)
Sotnikov VI, Kalinin AS, Berzina AP (1983) Genetic model of the copper-molybdenum association.
In: Kuznetsov VI (ed) Genetic models of endogene ore formation, vol I. Nauka, Novosibirsk, pp
112 -127 (in Russian)
Sotnikov VI, Berzina AP, Zhamsran M, Garamzhav D, Bold D (1985) Copper associations in
Mongolia. Nauka, Novosibirsk
Spatz DM (1979) Potassium-argon age of the Cerro Colorado porphyry copper deposit, Panama.
Econ Geol 74:693 - 695
Taylor HP Jr (1982) Oxygen and hydrogen isotopes of hydrothermal mineral deposits. In: Barnes HL
(ed) Geochemistry of hydrothermal ore deposits. Mir, Moscow, pp 200 - 237 (in Russian)
Tugarinov AI, Voinkov DM, Grinenko LN, Pavlenko AS (1974) Isotope composition and sulphur
sources of copper-molybdenum prospects in Mongolia. Geokhimiya 2:171 -178 (in Russian)
Westra G, Keith SB (1981) Classification and genesis of stockwork molybdenum deposits. Econ Geol
76:844-873
Part III
Deposits of Volcanic-Hydrothermal Association
Base Metal Deposits in the Iberian Pyrite Belt *
G. K. STRAUSS and K. G. GRAY!

Abstract

The Spanish-Portuguese Pyrite Belt covers a large area in the SW part of the
Iberian Peninsula. Known reserves of massive pyrite and cupreous stockwork ore
exceed 1000 million tons.
The stratiform sulphide deposits and accompanying manganese mineraliza-
tion are of synsedimentatry-exhalative origin and occur in a Lower Carbonifer-
ous, geosynclinal, volcanic-sedimentary rock sequence, strongly folded during
the Hercynian Orogeny. Epigenetic stockwork ores are in close relationship with
volcanism and often represent the feeder channels of the massive sulphide sedi-
ments.
Most important ore types mined are massive "crude" pyrites, lead-zinc-rich
"complex" pyrites, high-copper "banded" ores, copper-rich stockwork ores, dis-
seminated "porphyry"-type copper ores and gold-silver-rich gossans. The main
active mines are described and beneficiation of these ores is discussed.

Introduction

The Iberian Pyrite Belt represents one of Europe's most prominent metallogenic
provinces, as well as one of its oldest mining districts. It occupies the SW part of
the Iberian Penisula, extending approximately E - W over 250 km, from the
Province of Seville, Spain to the west coast of Portugal, covering some 8000
km 2•
Proved mineral reserves are in excess of 1 000 million tons and since the
middle of the last century some 250 million tons of massive pyritic ores have
already been mined. Other ore types, geologically closely associated in time and
space, include low-grade copper and copper-zinc mineralizations and gold-silver-
bearing gossans, both of great economic importance. Numerous small-sized
manganese deposits were worked in the past.
Mining and smelting were carried out by the Tartessians, Phoenicians, and
Romans. Following this period most of the mines lay dormant until their sudden

1 Compania Espanola de Minas de Tharsis, S.A., Minas de Tharsis (Huelva), Spain

* Manuscript submitted 12-2-1984.

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Base Metal Deposits in the Iberian Pyrite Belt 305

resurgence in the middle of the 19th century when extraction of copper from the
secondary enriched cementation zones of the rediscovered pyrite ore bodies be-
came the principal objective. Copper was produced either by direct smelting of
high-grade ores or by open-air roasting of cupreous pyrites and subsequent leach-
ing and recovery of copper via cementation of scrap iron. In the second half of
last century and during some 15 yrs. Spain was thus the leading copper district in
the world - until the much bigger producers in the USA and Chile started to
gather momentum.
At the same time, and with the depletion of the Sicilian sulphur deposits, the
enormous Spanish pyrite reserves became an alternative source as raw material
for the sulphuric acid production demanded by the ever-growing European
chemical industry. The base metals contained in the "cinders" (residues from
pyrite roasting), mainly copper and zinc, were recovered by newley developed
hydrometallurgical processes in several Northern and Central European plants.
The remaining "purple ore" (iron ore, freed of its base metal contents) was used
as feedstock for blast furnaces.
The Spanish-Portuguese pyrite industry, the world's third largest after the
USSR and P.R. China, nowadays produces some 2500000 tons per year (Tpy) of
crude pyrites. Some 2600000 Tpy of complex pyrites are mined to produce
copper, lead, and zinc concentrates with flotation pyrites as a by-product.
Another 6000000 Tpy of low-grade copper ores are mined and gold and silver are
produced from widespread gossans.

General Geologic Setting

The Iberian Pyrite Belt forms the central part of a Devonian-Carboniferous Geo-
syncline which represents the southernmost tectonic segment of the Hercynian
"Iberian Meseta". It consists of Upper Paleozoic eugeosynclinal rocks, pyritic
ores and manganese mineralizations and stretches as an arcuate belt across south-
western Spain and southern Portugal (Fig. 1).
A general description of the Pyrite Belt geology has been given among others,
by Strauss (1965, 1970); Schermerhorn (1971) and Strauss et al. (1977). Accord-
ing to Strauss (1965, 1970), the rocks of this belt are grouped into three units
(from bottom to top):
a) The basal Slate-Quartzite Group ("Phyllite-Quartzite Group", Schermerhorn
1971) is very uniform and consists principally of phyllitic slates and quartzites.
At its top occur rare quartzite conglomerates and some limestone lenses which
have been dated by conodonts to be of Upper Devonian (Famennian) age
(Hollinger 1958, Van den Boogaard 1967, Van den Boogaard and Schermer-
horn 1975).
b) The Volcanic-Sedimentary Complex ("Volcanic-Siliceous Complex", Scher-
merhorn 1971) overlies conformably the Slate-Quartzite Group. Its thickness
ranges from less than 50 m up to some 800 m and its lithology is strongly
variable. Out of the three, this is the only group with a wide variety of felsic to
 306 G. K. Strauss and K. G. Gray

BEJA-ARACENA
""·""NI~ICl.IN"

-ARACENA

Z
<C.
w
U
o

U
I- ~vl·"-n ll.< U"- COMPLEX
Z
oc{
--I

-
FARO

D POST-PALEOZOIC COVER ~ FERREIRA-FICALHO GREYWACKES (SUB-CULM)

~.::.::.:'.' VARISCAN GRANITES

I:!:0J
G2J
/.
%
VOLCANIC-SEDIMENTARY COMPLEX
SLATE-QUARTZITE GROUP (DEVON IAN)

D CULM GROUP [IJ PULO DO LOBO GROUP (SILURIAN-DEVONIAN)

Fig. 1. Geology of the Pyrite Belt

mafic volcanics and it exclusively contains all the stratabound, volcanogenic

sulphide and manganese mineralizations. Its age ranges from Tournaisian to
Lower Visean (Lower Carboniferous).
c) The Volcanic-Sedimentary Complex is conformably overlain by Culm, a
thick, well-stratified, monotonous succession of alternating slates and grey-
wackes. It represents the flysch phase of the Pyrite Belt Geosyncline and
reaches from Upper Visean to Lower Westphalian. Its thickness probably
changes from only some 500 - 1000 m in Spain, to several thousands of metres
in the western, Portuguese part.
During the Hercynian Orogeny (Middle Westphalian, Schermerhorn 1971)
the whole Paleozoic rock sequence was intensively folded, with a prevailing gen-
eral E - W direction in Spain, changing into a SE - NW strike in Portugal. Most-
ly narrow, isoclinal, and S - SW overturned fold structures were developed with
accompanying thrust faults and, at the same time, a coaxial slaty cleavage was
produced. Following the main folding phase, transverse faults were developed,
striking preferentially NE - SW and with occasionally important displacements.
A low-grade Hercynian regional metamorphism took place during the final
stages, somewhat later than the main folding phase.
Post-Paleozoic cover rocks are predominantly Tertiary sediments, mainly
sands and conglomerates.
Base Metal Deposits in the Iberian Pyrite Belt 307

Volcanic-Sedimentary Complex

The Volcanic-Sedimentary Complex (VS) is a heterogeneous rock sequence with

a rapidly varying thickness and quickly changing volcano-sedimentary facies in
both lateral and vertical directions. Felsic and mafic volcanics occur in lenses and
"beds" up to several hundred metres thick and are closely related to the volcano-
genic sulphide and manganese mineralizations, and a sedimentary framework
consisting predominantly of slates (Fig. 2).

AGE LITHOSTRATIGRAPHY

~~,~~~,tJ'~:~ _,~ ,~ ~ ~ I
QUATERN. SAND AND GRAVEL

PLIOCENE MARLS AND ARGILLACEOUS SEDIMENTS

MIOCENE LIMESTONES. SAI\OSTONES AND CONGLOtJERATES

.;.::.;.;.:.,... :,~ ; ;:.:~
SLATES WITH THICK GREYWACKES BEDS
~ ::J
a.. ~
-c:::
-10
::Ja:: --' SLATES (PARTLY CARBONACEOUS)
U<.9 --r
WITH Mltm GREYWACKE LENSES
..................
.................. SUBMARINE ACID VOLCANICS :
.................. LAVAS. BRECCIAS. AGGLOMERATES•
X .. ...............
UJ .................. MASSIVE FELSPAR TUFFS
-I -
a.. PURPLE SLATES
U)
::J 0
~ ~; ..........
,...... .... .. TUFFACEOUS SLATES. FINEGRAINED TUFFS
0 u WITH CHERTS. ARGILLACEOUS SLATES
a:: '::::::' ~1
.................. SUBMARINE ACID VOLCANICS: LAVAS. BRECCIAS.
UJ >-
a:: ......... .. ......
LL AGGLOMERATES MASSIVE FELSPAR TUFFS
z ~ ........
.... .... .... .. -... .. .-. ,-!. . SUBMARINE BASIC VOLCANICS:
0 z . . . ....... SPILITIC LAVAS AND PYROCLASTICS
co :r. .......?
~

UJ
a::
---
~ DIABASE SILLS
<! (5 -= -
u UJ
.""".
PURPLE SLATES WITH LENSES OF JASPERS

.. MASSIVE SULPHDE OREBootES h'II CARBONACEOUS

U)

U
I
~
BLACK SLATES AND TUFFACEOUS ROCKS
Z :~
<! BLACK CARBONACEOUS SLATES
U
.......'.
~
-I ..... TUFFACEOUS SLATES AND CHERTY TUFFS
..........
0
.................
> ........... ... .... ~
V SLBMARINE ACID VOLCAt-JCS: LAVAS. BRECCIAS,
AGGLOMERATES. MASSIVE FELSPAR TUFFS
......
... . ..... ~"fkt~~Rl;'Ss[Hl~OUS AND
z UJ
i==:':..::::::::> ~
<! I-
N
Z I- -====
= ~
ARGILLACEOUS AND SILTV SLATES
QUARTZITES AND QUARTZITIC
0
>
UJ
a:: a..
<! ::J
== ~=::::;
CONGLOMERATES
0 ::J 0
9UJ a::
a:: <.9
UJ BASE UNKNOWN
a..
~
a.. -I
::J U)

Fig. 2. Lithostratigraphy of the VS Complex

308 G. K. Strauss and K. G. Gray

Lithology

At the beginning of the VS, the sedimentation of the Slate-Quartzite Group was
abruptly terminated by the widespread eruption of felsic tuffs produced by acid
explosive volcanism at many different centres forming thick accumulations of
pyroclastics of all grain sizes and frequently interbedded with shales, thus
indicating a submarine deposition. Massive acid pyroclastics are grouped around
eruptive centres, lined up along fissural volcanic lineaments parallel to the
general strike. These eruptive centres are outlined by volcanic breccias,
agglomerates and isolated lava flows and volcanic chimneys have occasionally
been recognized. Laterally these volcanics interfinger rapidly with sediments such
as black shales, radiolarian cherts, etc. and/or decrease in grain size to form well-
bedded medium- to fine-grained tuffs or tuffaceous shales (Strauss and Madel
1974, Strauss et al. 1977). The felsic volcanism is predominantly of soda- or
potash-rich quartz-keratophyre or sodic rhyolite composition with albite being
the principal felspar phase. Potash-felspar megacryst-bearing pyroclastics are
rare.
The basic volcanics consist mainly of intrusive albite diabases, together with
less common spilitic pyroclasts, pillow lavas and tuffs, representing submarine
effusions. Their silica contents vary between 40 and 660/0 Si0 2 , indicating inter-
mediate compositions (Schermerhorn 1970). Ultrabasic cumulates are known in a
few localities of the Spanish part of the Pyrite Belt.
Extrusive spilites occur always in the immediate vicinity of acid volcanic
lineaments. Their lithostratigraphic position is, almost in every case, on top of a
preceding acid volcanic eruption cycle (Bernard and Soler 1974, Strauss and
Madel 1974), although the stratigraphic position of the extrusive mafics differs
between the eastern and western part of the Pyrite Belt.
The intrusive albite diabases, either as dykes or, more frequently, as sills, are
spaced at greater distances from acid pyroclastics and have intruded all forma-
tions of the VS Complex.
The sedimentary Jacies of the VS is largely dominated by slates, typically dark
to black shales often with a high content of carbonaceous matter and, in part,
with finely disseminated iron oxides and sulphides that weather characteristically
in pink to reddish tints.
Cherts and slaty cherts represent phases of stronger siliceous precipitation.
They often contain radiolaria and when abundant, even pass into radiolarites.
Dark grey or red jaspers are another typical sedimentary rock of the VS Com-
plex, usually occurring above felsic volcanite series and often associated with
manganeses ores.
Coarse detrital sediments occur locally in the upper part of the VS, while
limestones form a very minor part of the VS assemblage and occur locally as thin
layers or small lenticular bodies, often in a sideritic-dolomitic composition.
These do, however, often occur close to the sulphide horizons and can form
useful geologic marker horizons.
The strata-bound massive sulphide deposits are always connected, in time, to
the different submarine explosive acid volcanisms and specifically to the final,
waning stage of each eruptive cycle. In space they are confined to the vicinity of
Base Metal Deposits in the Iberian Pyrite Belt 309

the volcanic centres, either deposited directly on massive acid pyroclastics (au-
tochtonous ores) or sometimes deposited at some distance and interbedded with
black shales and tuffs (allochthonous ores) (Schermerhorn 1970, Strauss and
Madel 1974, Strauss et al. 1977).
The stockwork ores are of epigenetic origin and were generally the feeder
channels of the overlaying massive sulphide sediments.
The manganeses ores are always associated with grey and red ferruginous
jaspers and purple-violet tuffaceous shales that make the "manganese
formation" (Strauss 1965, 1970). This ore facies (silicates, carbonates and
oxides) is either contemporaneous with the sulphide deposits or, as in most cases,
somewhat later. It was formed during greater time spans and occupies more
extensive areas than the sulphide facies.

Ore Deposits

The geology and metallogeny of the Iberian pyrite deposits have been described
and discussed in a great number of papers (Strauss 1965, 1970, Strauss and
Madel 1974, Strauss et al. 1977 and references cited therein) and will not be
repeated at length here. Several types of mineralizations are known, of which the
best known and economically most important ones are the syngenetic-sedimen-
tary, massive, polymetallic pyritic ores. They always occur in massive, lenticular-
shaped ore bodies of varying dimensions and can vary in size from less than
100000 to more than 100000000 tons. Their principal mineral is always pyrite,
together with varying amounts of chalcopyrite, sphalerite and galena, and lesser
ainounts of arsenopyrite, pyrrhotite, tetrahedrite, lead sulfosalts, cassiterite and
many other minor minerals. Principal gangue consists of finely disseminated or
banded intercalations of the host rocks, together with silica, carbonates and to a
lesser extent barytes (Strauss 1965, 1970).
The most important facts with respect to the geologic environment and
formation of these ore deposits can be summarized as follows:
1. The massive sulphide ores, together with disseminated sulphides and stock-
work-type mineralizations occur exclusively in the Volcanic Sedimentary
Complex (Fig. 3).
2. Massive sulphide ores and manganese mineralizations are both sedimentary
and are closely linked with submarine, acid-alkaline volcanism.
3. In the case of multiple acid eruptive cycles along one volcanic lineament, only
one seems to be "productive" with regard to economical sulphide mineraliza-
tions.
4. The sedimentary sulphide concentrations, both massive and disseminated, are
always located near the borders of the volcanic extrusive centres and at a geo-
logically well-defined distance, where black shales are interbedded with acid
pyroclastic rocks. (The manganese mineralization, both laterally and vertical-
ly, is distributed more widely than the much more restricted sulphide facies).
5. The concentration of sulphides into massive ore lenses is explained by the
inflow of sulphide mud and/or detrical sulphide material into topographical
depressions around submarine volcanos.
310 G. K. Strauss and K. G. Gray

J
3

PROXIMAL TYPE
0. . ACID VULCANITES

2 TRANSITIONAL TYPE 0 MEDIUM SERICITIC ALTERATION

3 DISTAL TYPE
~ HEAVY SERICITIC ALTERATION

J JASPERS WITH Mn HEAVY CHLORITIC ALTERATION

1"\\\\\\[
WITH CUPRIFEROUS STOCKWORK

Fig. 3. Types of mineralization

6. Sulphide ores with massive, fine-grained textures, were generally deposited on

coarse porphyritic pyroclastics in the immediate vicinity of the corresponding
extrusive centre. Sulphide ore bodies with a layered, internal structure and
detrital sedimentary textures were deposited in a more shaly environment at
greater distances from the eruptive centres.
7. Clastic sedimentary ore textures indicate resedimentation of sulphide
material. This implies that previously formed, already solidified pyritic
deposits were submitted to submarine erosion and/or flows, their material
being transported into newly formed depressions.
8. The close relationship between massive sulphide ores, manganese ores and
submarine acid-alkaline volcanism clearly implies a genetical link between this
type of volcanic activity and the supply of sulphur and metals: The volcanism
was associated with hydrothermal activity venting on the seafloor, where iron
and base metal sulphides and manganese silicates and carbonates were
precipitated as chemical sediments.

Synsedimentary disseminated sulphide mineralization is widespread and occurs

usually in the hanging wall of massive sulphide ore bodies, sometimes as isolated
ore bodies. They consist mostly of pyrite, chalcopyrite and sphalerite, finely dis-
seminated and sometimes layered, within a sedimentary environment, consisting
mostly of black shales, tuffs and/or tuffites. Their origin can be interpreted as
the waning phase of volcanic supply of metals and sulphur into basins where
sedimentation of shaly and tuffaceous material was more intensive than that of
the sulphides.
Closely related to the synsedimentary massive sulphide ores is the epigenetic
stockwork-type sulphide mineralization occurring underneath autochthonous
Base Metal Deposits in the Iberian Pyrite Belt 311

massive pyritic deposits, predominantly in their highly chloritized, sericitized and

silicified footwall rocks. Their sulphide grade varies usually from less than 100/0
to over 50% pyrite equivalent «5 - 25% S), but a slow transformation into
massive pyritic ores is known at several mines. They consist mainly of hydro-
thermally altered volcanics and their sulphides occur within the brecciated host
rock (breccia ore), in scattered veins and veinlets (stockwork ore) and finely dis-
seminated (disseminated ores). By far the most widespread sulphides are pyrite,
chalcopyrite and sphalerite. This type of mineralization is interpreted as of
epigenetic origin with respect to the host rocks, and as the hydrothermal feeder
channels of the sedimentary sulphides (Williams 1967, Rambaud and Perez 1969,
Bernard and Soler 1974, Strauss and Madel 1974, Garcia Palermo 1980, Williams
et al. 1975).
The syngenetic, volcano-sedimentary manganese mineralization forms an
integral and most typical part of the VS Complex and represents together with
accompanying jaspers, cherts and purple slates one of the most outstanding
marker formations within the Pyrite Belt. Over 500 small manganese ore bodies
are known, none of which, however, ever reached any economic importance. The
primary ores consist mainly of rather low-grade silicates and carbonates and only
the near surface oxide cappings were worked in the past.

Pyritic Ores

Ore Types

The great variety of pyritic ores can be subdivided due to their chemical and
mineralogical composition and industrial use, on the one hand, and to their
formation, on the other.
Massive Pyrite is a typical polimetallic sulphide ore with pyrite as the predom-
inant mineral component, always accompanied by a great variety of other base
metal sulphides, mainly chalcopyrite, galena and sphalerite, together with minor
amounts of arsenopyrite, tennantite, complex Cu-, Pb-, As-, Sb-, Hg-, Ag-sul-
phides, cassiterite, gold and many other ore minerals. Main gangue components,
which as a rule do not exceed 5% of the total, are finely distributed or layered
host rock, quartz, carbonates and sometimes baryte. A common feature of all
massive pyritic ores is that the mineral components are extremely fine-grained
and with intense intergrowth (Strauss and Gray 1982).
The massive pyrite ore is known on the international market as "crude pyrite"
or "standard pyrite" and contains typically: S 45 - 48%; Fe 40 - 43 %; Cu
0.6-1.0%; Pb 0.5 -1.0%; Zn 1.0-2.5%; As 0.3-0.4%; Au 0.5-1.2 g/t; Ag
20 - 30 g/t. This ore type has been mined for more than a century at Rio Tinto,
Tharsis, La Zarza and many other smaller operations.
Since the early 1970s complex pyrites are of growing importance. They are
found in most pyrite mines and occur usually as interbedded layers and lenses
within crudes pyrites, or in their hanging and footwalls. The typical features of
312 G. K. Strauss and K. G. Gray

these massive, "complex" pyrites, compared to the "standard" ones, are their
somewhat lower sulphur content, higher lead and zinc values and sometimes
rather high contents of arsenic, antimony, mercury, silver and tin. The extremely
fine intergrowth of their mineral components as well as the occurrence of com-
plex sulphide minerals make their separation and beneficiation by flotation very
difficult. Typical values of these complex pyrites are: S 40- 450!0; Fe 34-41 %;
Cu 0.4-1.5%; Pb 1-2%; Zn 2.5 - 5%; As 0.4-0.8%; Sn 0.01-0.06%; Au
0.5 -1.0 g/t; Ag 40 - 60 g/t.
Banded copper ores were discovered at Neves-Corvo, Portugal and are locally
known as "banded ores". Compared to normal pyrites they are very rich in
copper (8% Cu) and silver (44 g/t), low in zinc (0.40% Zn) and contain only
traces of lead.
Typical and widespread are massive low grade pyrites, known in Spain as
"azufrones". They consist of massive pyritic ores which either contain a large
amount of finely disseminated host rock as gangue or occur as thin layers inter-
bedded with the country rock. These ores are economically of little importance,
except when they contain higher amounts of base metals, usually copper.
A special type of massive, low-grade pyrite is the carbonate ore, principally
found in Tharsis where it occurs interstratified with normal crude pyrites. A
typical feature is the abundance of brecciated pyrite-clasts in a siderite matrix as
gangue.
Closely related to these snygenetic-sedimentary, massive pyritic ores occur
epigenetic, disseminated and/or stockwork mineralizations, often underneath
autochthonous massive pyrite ore bodies and sometimes as isolated units, but
always in hydrothermally altered, chloritized, sericitified and silicified, acid
pyroclasts.
A typical example of this group are the disseminated copper ores of Cerro
Colorado/Rio Tinto which have in the last decade become one of the most im-
·portant copper sources of Western Europe. Pyrite-chalcopyrite- and quartz-
chlorite veinlets and stringers in strongly, hydrothermally-altered felsic tuffs
contain about 0.64% Cu.

Massive stockwork copper ores are known and have been extracted for many
years at Alfredo/Rio Tinto. Thick chalcopyrite-pyrite veins and veinlets cross the
strongly chloritized and silicified tuffaceous host rock which sometimes "is so
shattered that it forms a breccia within a sulphide matrix. Copper and zinc values
both average about 1.5% and sulphur amounts to 20-300!0. Very similar is the
"siliceous copper ore" of La Zarza. The "stockwork" ores and "brecciated" ores
of Neves-Corvo, Portugal seem to differ from the other ones only by their much
higher copper values (1.5 -4.0% Cu).

Mines

The Spanish-Portuguese Pyrite Belt hosts a total of eleven operating mines with
another two being developed (Fig. 4).
Base Metal Deposits in the Iberian Pyrite Belt 313

LOUSAL BEJA ...•••••... / IBERIAN PYRITE BELT

o • •.•.. (with important Mines)
ALJUSTREL • LOMERO CONCEPCION
GAVIAO OO / SAN TELM0 8 0 0 0
/ LAZARZ 06 RIO TINTO

THARSISO AZNALCOllAR
6 ,/ 0 C SOTIEl 60
SALGADINHO NEVES CORVO
\
o ( HERRERIAS
\ •
\ $.EVILLA

HUELVA

FARO
•
ORE TYPES RESERVES
o CRUDE PYRITE "0 0 < 5,000.000 T,
D PYRITE BELT
o COMPLEX PYRITE {;,. 0 0 5,000,000-20,000000 T.
A COPPER ORE 6 0 0 > 20,000.000 T.

Fig. 4. Mines of the Pyrite Belt

The biggest and in all ways the most important ore field is Rio Tinto, original-
ly formed by a single stratiform sheet of massive pyritic ores, 4500 m long, 1 500
m wide, and up to 80 m thick, which was later folded into an anticline with most
of the pyrite in the flat crest of the fold being gossanized (Cerro Colorado)
(Fig. 5). The massive pyrite ore bodies all lie in a thick sequence of quartz-kerato-
phyric and rhyolitic lavas and tuffs and show widespread pyritic and cupriferous-
pyritic stockwork mineralizations in their foot walls. These consist of felsic pyro-
clasts, which are hydrothermally altered (chloritized, sericitized and silicified).
Over 140 million tons of crude and cupreous ores have been mined with 50
million tons remaining in the San Dionisio ore body, where nowadays 500000
Tpy are extracted by open pit. At "San Antonio" 12 millions tons of complex ore
have been proved by underground workings, with 1.60/0 Cu, 1 % Pb and 2% Zn.
In the" Alfredo" stockwork ore-deposit reserves of 15000000 tons grading 1.7%
copper are mined underground at a rate of 600000 Tpy. Reserves of disseminated
copper ores at the "Cerro Colorado" open pit amount to over 150 million tons,
grading 0.60% Cu and are mined at a rate of over 4 million Tpy. Total produc-
tion of copper in concentrates is over 30000 Tpy from a 20000 Tpd (tons per day)
capacity mill. Up to 4 tons of gold and 45 tons of silver are produced every year
from a cyanide mill treating 2 million Tpy of gossan which contains 2.0 glt Au
and 50 glt Ag.
At Tharsis several big ore bodies are known, the most important ones being
North Lode, S. Guillermo and Sierra Bullones, with a combined strike extension
314 G. K. Strauss and K. G. Gray

of 1 500 m, average thickness 80 m and reaching to over 400 m depth (Figs. 6 and
7). Proved reserves are over 100 million tons, consisting mainly of massive crude
pyrites with 46.511,10 S, 0.7% Cu, 0.8% Pb, 1.8% Zn and rather high gold values
(1.2 g/t Au). The stratiform pyrite lenses are enclosed by black slates with some
tuffaceous interbeds and thin layers of dolomitic-ankeritic limestones and spilites
in their hanging wall, overlain by a thick sequence of felsic tuffs. Approximately
800000 Tpy of crude pyrites are extracted from the North Lode open pit.
La Zarza consists of a large single ore body, 3000 m long, locally over 100 m
thick and over 400 m deep (Fig. 8). Some 50 million tons of massive pyrites were
extracted since modern mining started with more than the same amount remain-
ing in this 400000 Tpy underground mine. The massive pyrites average 47% S,
0.7% Cu, 0.6% Pb and 1.5% Zn and are conformably interbedded within felsic
tuffs, tuffites, black slates and cherts. They are locally underlain by "siliceous
copper ores", a stockwork mineralization of yet unknown dimensions, averaging
over 1.5% Cu, 1.4% Zn and 28% S. Disseminated pyrites and low-grade
stockwork ores with only 0.1 - 0.6% Cu occur in the same formation, as well as
chloritized pipelike structures with no mineralizations at all.
Aznalcollar mines a large ore body containing 45 million tons of massive pyritic
complex ores, grading 0.44% Cu, 1.77% Pb, 3.33% Zn, 1 g/t Au and 67 g/t Ag.
It is overlain by "pyroclast ore", finely disseminated and thin stockwork veinlets
of pyrite, chalcopyrite and minor sphalerite in chloritized, felsic tuffs. Reserves
of 47 million tons average 0.58% Cu, 0.40% Zn and 10 g/t Ag. Both ore bodies
are interbedded in a thick series of volcano-sedimentary rocks, consisting mainly
of acid tuffs, tuffites and black shales (Fig. 9). The open pit operation started in
1980 and present production is at a rate 2000000 Tpy complex ore and 1600000
Tpy of "pyroclastics".
At Sotiel Coronada a massive complex pyrite ore body is enclosed in black
shales with some limestone beds and tuff intercalations. Approximately 42
million tons of complex ore grading 0.44% Cu, 1.36% Pb, 4.25% Zn and 39 g/t
Ag are now prepared for underground operation. Eventually 600000 Tpy of ore
are expected to yield some 38000 Tpy of zinc concentrates (50% Zn), 11 000 Tpy
of copper concentrates (23% Cu) and 500000 Tpy of fine-grained pyrite concen-
trates as a by-product, half of which will be used locally for sulphuric acid pro-
duction.
At Lousal, Portugal, known workable ore reserves are limited and only 40000
Tpy of low sulphur pyrites are extracted. Relatively high cobalt values are typical
for this ore body (0.1 % Co) (Fig. 10).
At Aljustrel, six ore bodies contain over 200 million tons of crude and com-
plex pyrites (Fig. 11). They occur at the top of a sequence of coarse-grained and
fine-grained tuffs, siliceous shales and jaspers. In the northern part, a single ore
body, over 60 m thick and containing more than 100 million tons of pyritic ores,
was cut and displaced by a Hercynian wrench fault, forming the separate masses
Feitais and Estacao. The southern ore zone includes the four remaining deposits,
extending over a strike length of over 4000 m. These ores are more cupreous than
the ones of the northern zone, averaging 1.1 % Cu in the Moinho deposit and
 t:C
III
~

RIO TINTO MINE ~

(1)

§:
t:1
(1)
'0
\
t-l D TRANSPORTED GOSSAN o
~.
SLATES. '"
S·
D CABe.QNJFEROUS:
GREYWACKES. FLYSCH ;.
(1)

D TRANSITION SERIES:
CT
TUFFS. LAVAS ~.
III
ID JASPER AND CHERT LENSES P
"tI

D MASSIVE SULPHIDE (1)

a·
t:C
D "IN SITU" GOSSAN &
D ACID COMPLEX: TUFFS. LAVAS.
. . 'BRECCIAS, AGGLOMERATES
o sec "D ~ Om

SAN DIONISIO
D... BASIC COMPlEX: SLATES, LAVA~
AGGLOMERATES BASIC LAVAS
CERRO COLORADO
D DEVONIAN: SLATES WITH SOME
F. NORTE QUARTZITE BEDS
D STOCKWORKS Pl'R1TIC

D STOCKWORKS CUPREOUS

SURFACE PROJECTION OF
ANTONIO- PLANES MASS
':;;~:> SAN
8\···.M~~!'1if;~;·i: SECTION C- D
o e Dam --- FAULT

SECTION A-B
o 2 0~ m
- \\::: 1 LEACHED
Fig. 5. Rio Tinto D SECONDARY w
-'"
 ~
'"

THARSIS OREBODIES

rrl CULM GROUp,

tL.cz.J SLATES AND GREYWACKES
• ACID VULCANISM
SLATES. TUFFACEOUS SLATES. BEDDED
D PYROCLASTICS. CHERT. JASPERS. SPILITE
LAVAS AND DIABASE INTRUSIVES
1_ 1MASSIVE AND DISSEMINATED
SULPHIDE OREBODIES

o SLATES AND QUARTZITES

P
311m. ~
VI
~
I»
en
en
=
I»
::;
0.
Fig. 6. Tharsis ore bodies
~
P
(;)
..,
I»
'<
 to
I»
'"
"~
~
0
't:I
"0
:1 '"
~.

N s·
Er
"cr
"I»::!.
::s
THARSIS MINE '"tl
'<
::!.
t>
D TIP HEAPS to
~
[§;] THRUST FAULTS
~I -~
A ~,, ACID LAVA FLOWS
, " AND PYROC L AS TICS

~ SPILITE PILLOW LAVAS

,',', AND TUFFS

D LAYERED ACID TUFFS

AND TUF FITES

D BLACK SLATES

[~im MASSIVE SULPHIDE ORE

Q SILT SLATES

TYPICAL CROSS- SECTION ~ SLATES AND OlJARTZITES

Fig. 7. Tharsis Mine (Fil6n Norte) ~

.....
 ~
<Xl
LA ZARZA MINE

AN
LA ZARZA OPENCAST
I
~~~
~
f~~
\\
~
)~ PLAN
I) ,
,~

LONGITUDINAL SECTION
A
o BLACK SLATES

D TUFFS AND TUFFACEOUS SLATES

p
~ RED JASPER LENSES imll ORE TO BE MINED ~
14th FLOOR ' "'-._ __" ~ ; ... -r. ',- :4'
D BEDDED GREY GREEN AND RED TUFFS • CUT AND FILL STOPING ...
,t:lth FL.OOR ~
Q MASSIVE PYRITE ORE D FILLED STOPES '"'"
S»
o
p.
CROSS SECTION D BLACK TUFFACEOUS SLATES -~ .
~"""""""""'iiiiiI!!'!"",,"",O=' ~
COARSE PORPHYRITIC FELSPAR-TUFFS p
Fig. 8. La Zarza o WITH OCCASIONAL STOCKWORK MINERALIZATIONS
...C'l
.el
Base Metal Deposits in the Iberian Pyrite Belt 319

AZNALCOLLAR MINE
CROSS SECTION
s N . II.COO N
\, ..... .
~t . · ......... ' .. '. ~

o RHYOLITE
o PYROCLASTICS
r::zl MINERALISED PYROCLASTICS
D PYRITE
o CLAYEY PYROCLASTICS
D GRAPHITIC BLACK SHALES
o FELSITE

Fig. 9. Aznalcollar

1.5070 Cu at Gaviao. Moinho and Feitais are worked underground and some
250000 Tpy of complex pyrite ores are produced, with an average grade of
45% S, 0,86% Cu, 1.42% Pb, 3.18% Zn, 0.7070 As and 40 glt Ag and are used
for sulphuric acid production.
The most spectacular Neves-Corvo ore deposits were discovered in 1977
40 km SE of Aljustrel and under 300 to 500 m deep bedrock. Since then four flat
lying ore bodies have been outlined, Neves, Corvo, Graca and Zambujal, with
estimated total reserves of 100 million tons of massive pyritic and stockwork ores
(Fig. 12). The ore deposits are interbedded in a thick sequence of acid volcanics,
tuffites, black shales and beds of greywackes, siliceous shales, jaspers and
limestones and overlain by Culm greywackes.
The most striking discoveries were 34 million tons of high-grade copper ores.
Banded ores, massive, synsedimentary copper ores, form a 1.5 to 4 m and locally
up to 10-cm-thick layer on top of the massive pyrites and acid pyroclastic rocks,
grading usually about 8% Cu and 44 glt Ag, but only 0.4% Zn and traces of Pb.
The felsic pyroclastic rocks in the foot wall of the massive pyrite ores are strongly
chloritized, sericitized and silicified and show an epigenetic stockwork mineral-
 ....
~
LOUSALMINE
GEOLOGICAL PLAN (380 LEVEll

o BASIC TUFFS MASSIVE PYRITE

~ CARBONACEOUS SHALES D PORPHYRITIC TUFFS

50 100m
-------,
o--
~ STOCKWORK MINERALISATION
~ IN CHlORInSED PORPHYRITIC
TUFFS

p
~
...~
~
en
en
I»
.........
..... ....
....... -,
........
.....
.... .... " 5.
. ......... ........... . . ~
p
o
...
Fig. 10. Lousal ~
 t:1:l
I»
~ FEITAIS '"(\)
~
(\)

"'\''\~~
100 -\ . '.'. , "
-t ff --: [
' l},... " ..k~ ALJUSTREL ORE BODIES t;I
• jl~J' \ •
200 - , 'I" ':," ~
'1/. \, 'I: ' PLAN 200 LEVEL ~.
300 \ '"
5'
o !!!.o. s:-
400 "'V (\)

500 CT
(I)

o 300m ;.
::I
-- ."
SECTION A-B r--;:(~ - .&..\ ~.
~O~~ t:1:l
"
!¥:

D CULM SHALES AND GREYWACKES

D PARAISO SILICEOUS FORMATION PYRITE

• JASPER D LOWER TUFFS

D UPPER TUFFS ~g;:';.,:l MESSEJANA DOLERITE

Fig. 11. Aljustrel ~

~
322 G.K. StraussandK . G. Gray

NEVES- CORVO MINE

(-~~'~.
. . ..=-~

GRACA

A NEVES CORVO B
Roo~ 1 . ~~~~~-:---r~__~~~~___
.=.,••~

_<IIDo l
[J CULM GROUP ~ MASSIVE SUL.PHIDE

o SILICEOUS SCHISTS AND TUFFITES I: :: :1AC ID VOLCANICS

D GREYWACKES ~ STOCKWORK

o BLACK SCHISTS
a
~
eOOm
.iiiiiiiiiiiE==-

Fig. 12. Neves-Corvo

ization consisting mainly of chalcopyrite and pyrite. These massive stockwork

ores average 1.50 - 4.0070 Cu and 10 g/t Ag and similar breccia ores grade
2.5-3.5% Cu and up to 2% Zn (Albouy et al. 1981). An underground mining
operation is being prepared and should be on stream in 1988/89 on a basis of
1 000000 Tpy of ore producing some 60000 Tpy of copper in concentrates.

Ore Beneficiation

The Iberian pyrite mining industry is based on many different types of ores and,
similarly, several ways to use their contents of sulphur, base metals, gold and
silver have been used or are being investigated.
Base Metal Deposits in the Iberian Pyrite Belt 323

Crude pyrites was until a few years ago the most important ore type produced
in the pyrite belt and, for many years, constituted the back-bone of the European
sulphuric acid industry. At present some 2.5 million tons per year of massive,
crude pyrites are produced, crushed and screened, and sold mostly to the local
sulphuric acid industry together with some exports. After roasting there remain
0.7 tons of residues ("cinders") per ton of pyrite, some of which are then treated
for the recovery of base and precious metals and the production of iron ore
(purple ore), suitable as a blast furnace feed.
Although there are several approaches to the treatment of complex ores, the
only one being used industrially at this time is differential flotation with the pro-
duction of copper, lead and zinc concentrates together with large quantities of
very fine residual pyrite concentrates (Aznalcollar and Sotiel mines/Spain). The
complexity of the mineral association and the extremely fine intergrowth require
an ultrafine milling (800/0 < 5/20 J.1) and even so make the production of prime
quality base-metal concentrates difficult.
Alternative routes for the beneficiation of these complex ores, which are still
under investigation, can be subdivided into hydrometallurgical and pyrometal-
lurgical approaches and the following are representative (Strauss and Gray 1982):
1. Flotation of bulk concentrates followed by recovery of the base and precious
metal values by hydrometallurgicalleaching processes under varying physical
and chemical conditions. Examples of these are the Sherrit Gordon and
Comprex pressure leaching processes, on the one hand, and the Elkem Spiger-
verket and Minemet ferric chloride leach processes, on the other.
2. Roasting of the ore under sulphatizing conditions allowing the recovery of
mainly, copper, zinc and silver by a simple water or dilute sulphuric acid
leach. The roasting can be carried out in two steps with the production of
elemental sulphur in the first, if so desired. Examples of this route are
Outokumpu's Kokkola operation (Finland), and the Auxini (Spain) and Sapec
(Portugal) processes under investigation.
3. Direct pyrometallurgical (smelting) routes such as the Kivcet process are also
under consideration.
Another important source of base metals, mainly copper, are the massive
stockwork copper ores and disseminated copper ores. Being both of epigenetic-
hydrothermal origin, although also fine-grained, they are less so than the massive
ores, with less intimate mineral intergrowth characteristics and minor occurrence
of complex Cu - Pb - As - Sb - Hg minerals. As a result, milling to 80% less 325
Mesh is usually sufficient and relatively good liberation and clean concentrates of
copper, zinc and pyrite are achieved.
The gold-silver gossan ores are submitted to cyanide leach processes and
precious metals are recovered by zinc-dust precipitation.
324 G. K. Strauss and K. G. Gray: Base Metal Deposits in the Iberian Pyrite Belt

References

Albouy L, Conde LN, Follerini F, Leca T, Morikis A (1981) Les gisements de sulfures massif poly-
metalliques de Neves-Corvo (Baixo Alentejo, Sud Portugal) Chron Rech Mines (Paris) 460:5 - 31
Bernard AJ, Soler A (1974) Apercu sur la province pyriteuse Sud-Iberique. In: Bartolome P (ed)
Gisements stratiformes et provinces cupriferes. Liege Soc Geol Belg, pp 287 - 315
Boogaard M van den (1967) Geology of the Pomaro region (South Portugal). Thesis, Univ
Amsterdam
Boogaard M van den, Schermerhorn LJG (1975) Conodont faunas from Portugal and southwestern
Spain. Script Geol Leiden 28:1-43
Carvalho D (1979) Geologia, metalogenia e metodologica da investigacao de sulfuretos polimetalicos
do Sui de Portugal. Comun Serv Geol Portg (Lisboa) 65:169-191
Garcia Palermo F (1980) Caracteres geologicos y relaciones morfologicas y geneticas de los
yacimientos del Anticlinal de Rio Tinto. Thesis, Univ Salamanca
HOllinger R (1958) Beitrag zur Kenntnis der Geologie im SW der Provinz Huelva. Thesis, Univ
Munster
Kinkel AR Jr (1962) Observations on the pyrite deposits of the Huelva district, Spain, and their
relation to volcanism. Econ GeoI57(7):1071-1080
Rambaud Perez F (1969) EI sinclinal carbonifero de Rio Tinto (Huelva) y sus mineralizaciones
asociadas. Mem Inst Geol Min Esp (Madrid) 71:1-229
Routhier P, Aye F, Boyer C, Lecolle M, Moliere P, Picot P, Roger G (1979) La ceinture sub-iberique
a amas sulfures dans sa partie espagnole mediane. Mem BRGM 94:265 pp
Schermerhorn LJG (1970) The deposition of volcanics and pyritite in the Iberian pyrite belt. Min
Deposits 5 (3):273 - 279
Schermerhorn LJG (1971) An outline stratigraphy of the Iberian pyrite belt. Bol Inst Geol Min Esp
(Madrid) 82(3/4):239 - 268
Schermerhorn LJG (1980) Copper deposits of the Iberian peninsula. In: Jankovic S, Sillitoe RH (eds)
European copper deposits. Fac Min Geol, Belgrade, SGA Spec Publ 1:295 - 303
Solomon M, Walshe JL, Garcia Palermos F (1980) Formation of massive sulphide deposits at Rio
Tinto, Spain. Trans Inst Min Metall London Sect B 89:B16-B24
Strauss GK (1965) Zur Geologie der SW-Iberischen Kiesprovinz und ihrer Lagerstatten, mit besonde-
rer Berucksichtigung der Pyritgrube LousallPortugal. Thesis, Ludwig-Maximilians-Univ,
Munchen
Strauss GK (1970) Sobre la geologia de la provincia pyritifera del Suroeste de la peninsula iberica y de
sus yacimientos, en especial sobre la mina de Lousal/Portugal. Mem Inst Geol Min Esp (Madrid)
77:266 pp
Strauss GK, Gray KG (1981) Complex pyritic ores of the Iberian peninsula and their beneficiation,
with special reference to Tharsis Company Mines. Inst Mine Metall London 79 - 87
Strauss GK, Gray KG (1982) The international outlook for Spanish pyrites. Ind Min London (Abstr)
175:29
Strauss GK, MadeJ J (1974) Geology of massive sulphide deposits in the Spanish-Portuguese pyrite
belt. Geol Rundsch 63(1):191-211
Strauss GK, Madel J, Fernandez-Alonso F (1977) Exploration practice for strata-bound volcanogenic
sulphide deposits in the Portuguese-Spanish pyrite belt. In: Klemm DD, Schneider HJ (eds) Time-
and strata-bound ore deposits. Springer, Berlin Heidelberg New York, pp 55 - 93
Vazquez Guzman F (1978) Depositos minerales de Espana. Bol Inst Geol Esp (Madrid) 1958 pp
Williams D (1962) Further reflections on the origin of the porphyries and ores of Rio Tinto, Spain.
Trans Inst Min Metall (London) 71 (5):265 - 266
Williams D, Stanton RL, Rambaud F (1975) The Planes-San Antonio pyritic deposit of Rio Tinto,
Spain: its nature, environment and genesis. Trans Inst Min Metall London Sect B 84:B73 - B82
Copper-Pyrite and Pyrite Base Metal Deposits of the
Caucasian Region
N. K. KURBANOV 1

Abstract

This paper analyzes metallogenic specialization of two pairs of pyrite-bearing

structural-formational zones 2 belonging to the Caucasian segment of the
Mediterranean fold belt. These zones evolved during the Alpine tectono-
magmatic cycle along the northern and southern framing of the Transcaucasian
microcontinent. In the northern pair (the Great Caucasus meganticlinorium) the
pyrite-bearing zones are in the central trough and the south continental rise, and
in the southern pair (the Lesser Caucasus meganticlinorium), they are in the rift
ophiolitic zone and the island arc. A clustering concentration pattern of base-
metal mineralization with centrifugal and centripetal types of localization of
pyrite, porphyry and vein deposits has been revealed.

Introduction

The present-day classification of the pyrite-bearing provinces represented by

various types of geosynclinal systems is based on the character of volcanic or
volcano-sedimentary rocks filling in the geosynclines and paleotectonic setting,
which determines the types and evolution of ore-bearing formations, their suc-
cession and the ore grade (Borodayevskaya et al. 1979, Kurbanov 1982,
Hutchinson 1973).
The Caucasian segment of the Mediterranean fold belt is noted for a variety
of pyrite deposits associated with geosynclinal systems of different age and
tectonic style. In the long period of polycyclic evolution (from Riphean to
Neogene) the major recent structures of the Caucasian region - Great and
Lesser Caucasus meganticlinoria separated by the Transcaucasian median mass
(microcontinent) - underwent progressive rejuvenation of the geosynclinal
regime, folding, and metallogeny in a southward direction (Fig. 1).

I TsNIGRI, Varshavskoje Shosse, 58, Moscow, M-430, USSR

2 A structural-formation zone lies within the fold area and differs from the neighboring zones in
features of sedimentation, structure, magmatism, determined by tectonic regime and a number of
physicogeographical factors (such as climatic) characteristic of the zone during its evolution. The
zone is usually bounded by deep faults. The term has been used extensively of late.

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
326 N. K. Kurbanov

Fig. 1. Location of copper-lead-zinc deposits in the Alpine zones of the Caucasus. 1 Scythian craton;
2 Transcaucasian median mass; 3 Iranian median mass; 4 structural-formational zones of the Great
Caucasus terrigenous eugeosyncline: a Pshekhi-Dagestan zone of early stabilization; b Bzyb-Tfan
trough zone (central); c Chkhalta-Sarybash zone of the continental rise. 5 Adjar-Trialeti and Talysh
superimposed troughs; 6 cretaceous volcanogenic island arc of Pontid; 7 Jurassic Somkheti-Kafan
island arc; 8 Sevan-Akerin ophiolitic zone; 9 activated areas of the Iranian median mass: a Ordubad
volcanogenic trough; b Vedi ophiolitic zone of the Cretaceous. 10 Boundaries of structural-forma-
tional megazones (a) and zones (b). 11 Alpine-type regional (a) and local (b) trusts. 12 Commercial-
genetic types of deposits: a stratiform pyrite-base metal; b pyrite-chalcopyrite-pyrrhotite; c copper-
pyrite; d pyrite-barite base metal; e sulfur-pyrite; f vein base metal; g copper-porphyry

An analysis of the geodynamics and formational composition of the

Mesozoic rocks of the Great and Lesser Caucasus shows that in the Alpine
tectono-magmatic cycle, which was most prolific in pyrite ores, that these were
independent systems, i.e. terrigenous (Great Caucasus) and volcanogenic (Lesser
Caucasus) eugeosynclines that evolved on a consolidated subplatform base as
secondary linear tension zones (Kurbanov 1982, Tvalchrelidze 1977, Khain
1975). The essential difference in the tectono-magmatic evolution of the above
megazones predetermined in many ways the variety of commercial-genetic types
of base metal mineralization within them.

Genetic Models and Formative Conditions of Pyrite Deposits of the

Great Caucasus Terrigenous Eugeosyncline

The Alpine terrigenous eugeosyncline of the Great Caucasus meganticlinorium,

which hosts the base metal-bearing massive sulfide province, emerged in the
north of the region, along the margin of the craton (Russian platform) and
microcontinent (Transcaucasian median mass) at the onset of the Lower Jurassic
within a narrow strip of the oceanic crust. It consists of three structural-forma-
Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 327

tional zones: (1) the northern, early stabilized Pshekhi-Dagestan zone adjacent to
the craton; (2) the central Bzyb-Tfan trough zone of the rift type; (3) the southern
Chkhalta-Sarybash zone of the continental rise, which developed along the active
margin of the microcontinent and was transformed subsequently into a volcano-
terrigenous island arc. These structural-formational zones are separated by longi-
tudinallong-lived synsedimentary fault zones (the Great Caucasus and Kekhna-
medan zones) related probably to the Benioff paleozones.
All the known pyrite deposits of copper, lead, and zinc are confined exclu-
sively to the central and southern structural-formational zones that are com-
plicated by allochthonous overthrust sheets and display complex structure.
Within the central BzYb-T/an trough zone (see Table 1) the ore-bearing rocks
are the slate, slate-graywacke, and terrigenous-flysch sedimentary formations, as
well as two unimodal volcanic formations: the early (Jtopl) siliceous-picrite-
spilite-basalt, and the late (J2 t-a) spilite-diabase ones.
The ore-bearing terrigenous and volcanic formations within the trough zone
compose two polycyclic regressive rhythmostratigraphic complexes 3: the lower
(Jto pi - J 2 ) 4600-m-thick complex, and the upper (J2 a - bt), 3900-m-thick
complex. Both complexes are noted for clay (with sulfide disseminations and
concretions) and volcanic units that accumulated in the linear synsedimentary
paleobasins.
Volcanic formations are accompanied by belts (the Kazbek and Kakhetian) of
diabase stocks and dikes, bounding the trough zone across the strike.
In terms of petrography and petrochemistry, basaltic rocks of siliceous-
picrite-spilite-basalt formation are comparable with oceanic tholeiites (similar to
tholeiites of the present-day rift of the Red Sea), whereas those of spilite-diabase
formation are correlatable with their subvolcanic equivalents similar to basaltic
rocks of unimodal formations of volcanogenic eugeosynclines (Fig. 2).
The above-described formations of the trough zone are crumpled into linear
and isoclinal folds showing an S-vergence and complicated by co-folded reverse-
thrust and overthrust faults. A thick crush zone with abundant dikes and minor
intrusions of volcano-plutonic associations, namely, basalt-dacite-rhyolite-
plagiogranite association (T2 bt), is traceable in the present-day structures along
the southern margin of the trough zone.
Three rhythmostratigraphic complexes are recognized in the southern
Chkhalta-Sarybash zone (see Table 1): (1) the Sinemurian-Pliensbachian
(1 300 m) complex that incorporates a layer of basal conglomerate - extrusions
and tuff of andesite-rhyodacite calcareous-alkaline potassium-sodium formation
(Sinemurian) and the ore-bearing slate-greywacke formation of the continental
rise (Toarcian); (2) the Toarcian-Aalenian (1500 m) a complex composed of ter-
rigenous-flysch (Toarcian) and an ore-bearing slate-greywacke (Aalenian) unit
containing lava flows and extrusive-tuffaceous facies related to the contrasting
basalt-rhyolite suboceanic formation of sodium series; and (3) the Bajocian-
Bathonian (2600 m) complex, comprising volcanics and hypabyssal intrusions of
potassium-sodium basalt-andesite-dacite-rhyolite-plagiogranite volcano-plutonic

3 Rhythmocomplexes in the Soviet geologic literature mean alternating sets of volcanic and sedimen-
tary formations that evolved at the same stage in a similar tectonic setting.
Table 1. Ore-bearing structural-formational zones of the Great Caucasus
w
tv
Rhythmo- Southern zone of active continental rise (Chkhalta-Sarybash) Central trough (rift) zone (Bzyb-Tfan) 00

strati-
graphic Magmatic Sedimentary Localization Magmatic Sedimentary Localization
complex formations formations levels of formations formations levels of
stratiform stratiform
Composition Thick- Composition Thick- ore bodies Composition Thick- Composition Thick- ore bodies
ness (m) ness (m) ness (m) ness (m)

Bajocian- Basalt-ande- 500- Terrigenous- 700- Gabbro- Terrigenous- 1300

Bathonian site-dacite- 1400 flyschoid 1200 plagio- carbonate-
rhyolite granite flyschoid
gabbro-plagio- (dikes and
granite minor
intrusions)

Toarcian- Basalt- 200- Slate-gray- 1500 Upper - in Splite-dia- 300- Terrigenous- 3900 Over spilite
Aalenian rhyolite 700 wacke shales of base (Lower 500 flyschoid: flows, in
(Aalenian) the upper Aalenian) rhythmic shales of
and ter- parts of alternation the lower
rigenous- Aalenian; of sandy- parts of the
flyschoid lower - in Splite-dia- 550- shale flysch- Upper Aale-
(Toarcian) shale of the base 600 oid members man.
upper part (Toarchian) (50-10 m) Over spilite
of Toarcian and shale flows, in
beds shales of
(100-150m) the lower
parts of
Toarcian

Sinemurian- Andesite- 150- Slate-gray- 1300 In the up- (Siliceous)- 350 Slate: shale, 4600 In the up-
Pliensh- rhyodacite 200 wacke: alter- per parts of picrite- phyllite with per parts of ;z:
achian nation of shales basalt graywacke shales, over ~
graywacke ~
members basalt flows ~
..,
sandstone (30-50 m) cr"
I\>
members and rhythm- ::s
0
(50-70 m) ically thin- <:
and shale bedded silty
beds sandstone
(100-150m) (25-30 m)
Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 329

association and a sandy-siltstone terrigenous-flysch formation developed mainly

in the southeastern segment of the southern slope of the Great Caucasus.
Magmatism in both zones was terminated by the emplacement of dikes and
minor intrusions of gabbro-diorite formation (J 3 -K t ). In terms of petrography
and petrochemistry, the products of magmatism are related to calcareous-alka-
line series of ensialic island arcs (Fig. 2).
Typical of the ore-bearing formations of rhythmocomplexes is the accumula-
tion of clay facies (enriched by C org up to 5 - 6070 against 0.1 - 0.9% background
value, as well as by numerous impregnations and concretions of pyrite and py-
rite-siderite) in local synsedimentary basins frequently controlled by magma- and
ore-feeding faults and bounded laterally by vast sand uplifts ("sand barriers").
The zone are noted for large anticlines (with recumbent southern, relatively
gently inclined northern limbs and heavily block-faulted cores), and narrow
synclines. The cores and northern limbs of the anticlines are frequently com-
plicated by local injection structures, high-order folds with steep hinges and co-
folded reverse-thrust faults. Many of them emerged as synsedimentary inversion
uplifts in place of the above-mentioned local paleobasins.
The majority of investigators (Borodayevskaya et al. 1979, Kurbanov 1982,
Tvalchrelidze 1977) relate copper-lead-zinc mineralization of the Great Caucasus

50 fO 70
/-19 0

1®2@J®.®S@
K10~fr'aZO

10 0 0 0 2 3 4

'®7e'@9@'@ S06()1@
Fig. 2A, B. Volcano-plutonic associations of the Great and Lesser Caucasus. A Variation diagram
showing potassium to siliceous acid ratio. Volcanic formations of the Great Caucasus: 1 unimodal;
2 bimodal; 3 trimodal; 4 gabbro-diorite-plagiogranite Volcanic formations of the Lesser Caucasus; 5
uni modal; 6 bimodal basalt-rhyolite; 7 bimodal andesite-basalt; 8 trimodal; 9 gabbro-plagiogranite;
10 cretaceous volcanic trimodal. B The ratio of the sum of alkali, ferrum, and magnesium, Great
Caucasus: 1 unimodal; 2bimodal; 3 trimodal, Lesser Caucasus: 4 unimodal; 5 bimodal; 6 trimodal; 7
field of undifferentiated basalt magma
330 N.K. Kurbanov

terrigenous eugeosyncline to two major ore types - stratiform pyrite and hydro-
thermal vein. Two subtypes, namely, pyrite-chalcopyrite-galena-sphalerite and
chalcopyrite-pyrrhotite are distinguished within the pyrite ore type. The former
subtype developed in two stages: an early, stratified hydrothermal-sedimentary
(essentially pyrite or pyrrhotite, with sulfides of copper, lead, and zinc) and a
late, hydrothermal-metasomatic (the most abundant in base metals) stage that
emerged after early geosynclinal deformations of stratified ores and the emplace-
ment of minor intrusions of plagiogranite bodies. The later ores of the hydro-
thermal-vein type, which form after the intrusion of late geosynclinal gabbro-
diorite formation, are represented by two paragenetic associations of different
ages: the early, quartz-chlorite-chalcopyrite-pyrrhotite association and the late,
quartz-carbonate-pyrite base-metal one. The combination of ores of the pyritic
subtype with those of the hydrothermal-vein type gave rise to polygenetic com-
bined mineral deposits of different ages within the province.
Localization of polygenetic and monogenetic deposits known from the region
follows a zonal pattern. Thus, the majority of deposits in the Bzyb-Tfan zone are
made up of pyrite ores of chalcopyrite-pyrite-pyrrhotite composition, whereas
those of the Chkhalta-Sarybash zone are represented by pyrite base-metal ores.
Superimposed veinlet-disseminated ores of quartz-chlorite-chalcopyrite-pyr-
rhotite composition are widespread in the trough zone, whereas quartz-car-
bonate-pyrite base-metal ores are common in the southern zone of the active con-
tinental rise. Such a localization pattern of the mineralization makes it possible
to identify two metallogenic zones: the Bzyb-Tfan and Chkhalta-Sarybash zones
whoses geological boundaries coincide with the structural-formational zones of
the same names.
The second localization of stratiform pyrite ores is determined by their con-
finement to three stratigraphic levels (see Table 1): Pliensbachian, Toarcian, and
Upper Aalenian, and within these levels - by their restriction to stratified ore-
bearing layers (with relatively high concentrations of iron sulfides ranging from
1% to 15070 and more). These layers occupy certain regular positions in shale
units of the transgressive and regressive series of rhythmostratigraphic com-
plexes.
The analysis of the localization of the deposits shows that the commercial con-
centration of the stratiform pyrite mineralization within the ore-bearing layers is
associated with clay facies that accumulated in the above-mentioned local
paleobasins under the conditions of a reducing regime. The central parts and active
sides of the basins are complicated by magma- and ore-feeding faults and the
products of submarine hydrothermal exhalations, such as sulfide ores of solid-
massive, bedded, mottled, and other structural types. Pyrite is the main mineral in
these ores; the amount of sphalerite, chalcopyrite, and galena is varied.
As has been established during prospecting for stratiform ores, the massive
deposits change laterally and vertically to sulfide rhythmites represented by
alternating sulfide and clay interbeds ("ore flysch"). The latter interbeds are, in
their turn, replaced laterally by facies composed of lenticular-streaky pyrite accu-
mulations, and then by stratified layers of pyrite concentrations.
Owing to the emplacement of subintrusions associated with a persistent early
island-arc ore-generating formation, the stratiform pyrite ores were subjected to
 Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 331

the most essential alterations of the first stage, which consisted of the introduc-
tion (and, possibly, regeneration and redeposition as well) of sulfides of the pro-
ductive associations that replaced metasomatically the sulfides of the earlier ores.
In time, this process coincided with the early geosynclinal stage of folding and
faulting (development of linear folds, reverse-, and overthrust faults, and flow
cleavage) that were responsible for boudinage and recrystallization of ores
accompanied by regeneration and redeposition of the ore matter along conjugate
faults and resulted in vein and veinlet-disseminated systems.
The most intense metamorphism of the stratiform deposits is connected with
hypabyssal intrusions of the late island-arc gabbro-diorite formation that were
accompanied by considerable regeneration of the ore matter and transformation
of chalcopyrite-sphalerite-pyrite ores from many mineral deposits of the Bzyb-
Tfan zone into chalcopyrite-pyrrhotite ores with halos of veinlet-disseminated
quartz-chlorite-chalcopyrite-pyrrhotite and quartz-carbonate-pyrite-sphalerite-
galena ores.
b

~I ~~ B~ ~ B~ ~6 B[J1

(TIe ~, E:} mil [SJrL D I) I ---r l"

Fig. 3. Type models showing various settings in which the Great Caucasus copper-pyrite deposits
were formed. Accumulation of hydrothermal-sedimentary ores in the local basins: A above ore-gen-
erating source; B near the source; C at a distance from the source. Location in the present-day struc-
tures: a in the fold zones; b in axial zones of linear folds; c in synclinal folds; d in inversion uplifts; e
in inversion uplifts complicated by injection structures. 1 Sandstone beds (a), pockets of intercalating
silty sandstone and shale (b); 2 sandy (a) and clayey (b) facies; 3 spilite and basalt lava flows and their
feeding channels; 4 minor intrusions and dikes of gabbro-diorite formation; 5 subvo1canic intrusions
of andesite, rhyodacite, and rhyolite; 6 beds of stratiform pyrite ores of mottled-breccioid structure;
8 veinlet-disseminated ores; 9 rhythmites (ore flysch); 10 sulfide concretions; 11 magma- and ore-
feeding channels: a main, b feathering. 12 Schistosity and crush zones; 13 reverse-thrust and thrust
faults; 14 direction of outpouring of mineral solutions
332 N.K. Kurbanov

The analysis of the localization of the pyrite and hydrothermal-vein base-

metal mineralization has shown that against the background of linear metal-
logenic zones, the stratiform mineral deposits occur as clusters. These ore dis-
tricts within the longitudinal metallogenic zones are confined to relatively up-
lifted inversion cross-cut blocks characterized by greater thickness of the ore-
bearing sedimentary formations. Major uplifts, in their turn, consist of several
conjugate local ore-enclosing inversion uplifts of various types; these are the
most promising ore fields. The latter frequently form compact ore clusters con-
trolled by regional longitudinal long-lived magma- and ore-feeding faults.
Taking all this into account, the type model of the setting in which stratiform
monogenetic deposits were formed, for the region under review, would be the
accumulation of these deposits in local paleobasins: (1) directly above the ore-
generating source; (2) near the source, where they are controlled by long-lived
magma- and ore-feeding faults; and (3) at a distance from the ore-generating
source. It is noteworthy that with the distance from the ore-generating source the
role of biogenic-sedimentary sulfides (Kurbanov 1982) increases significantly
(Fig. 3 A, B). However, the present-day models of the targets for prognosis and
exploration are determined not only by initial conditions of their formation, but
also by settings that depend on a combination of these and on the intensity of
their alteration during subsequent tectonic-magmatic processes that produced
polygenetic deposits of different ages.

Genetic Models and Formative Conditions of Pyrite Barite Base-Metal Deposits

Hosted by the Lesser Caucasus Volcanogenic Eugeosyncline

The volcanogenic eugeosyncline of the Lesser Caucasus occupies a geotectonic

position similar to that of the Great Caucasus, since it is situated between the
Afro-Arabian craton in the south and Transcaucasian microcontinent in the
north. Initiated as a secondary geosyncline only in the Middle Jurassic
(Kurbanov et al. 1978, Tvalchrelidze 1977, Khain 1975), this metallogenic
province differs considerably from the above-described one in its tectonic style,
role of magmatism and, therefore, metallogenic specialization and the set of
commercial-genetic types of base-metal deposits.
Three structural-formational zones, namely: (1) the northern Somkheti-
Kafan zone; (2) the central Sevan-Akerin zone; and (3) the southern Miskhan-
Zangezur zone are recognized in the Alpine volcanogenic eugeosyncline of the
Lesser Caucasus (Kurbanov et al. 1978, Tvalchrelidze 1977, Khain 1975).
The Miskhan-Zangezur zone consists of rigid blocks (Daralagez, Miskhan-
Zangezur) activated in the Jurassic, Cretaceous, and Paleogene and confined to
the northern framing of the Iranian median mass. These blocks include the
Cretaceous allochthon of the Vedi ophiolitic zone (Khain 1975). The Alpine base-
metal metallogeny of the zone is characterized by syngenetic, frequently strati-
form, pyrite ores associated with extrusions and volcanic domes of the Eocene
basalt-andesite-dacite-rhyolite calcareous-alkaline volcanic formation.
Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 333

The Sevan-Akerin zone which is an eastern continuation of the Tavr-Anatolia

ophiolitic zone, represents a linear rift tension zone that emerged (most probably
in the Bajocian) on the oceanic crust and was transformed subsequently into a
melange suture structure with numerous allochthonous slabs of the Cretaceous
limestone, ophiolitic rocks, and olistostrome units of Oxfordian-Tithonian age.
As to the endogenic mineralization, most typical of the early geosynclinal stage is
the development of chromite ore occurrences and hydrothermal-sedimentary
copper-pyrite mineralization of the Ergani-Maden type hosted by volcanics of
the unimodal tholeiite-siliceous-spilite-basalt formation.
The set of the igneous rocks, their geotectonic evolution, and their metal-
logenic specialization make it possible to identify the Jurassic Somkheti-Kafan
zone and the Cretaceous Pontid zone, replacing it an echelon in the west, as a
single belt of the ensialic volcanogenic island arc that formed on a thick and
active southern margin of the Transcaucasian microcontinent and that is a
surface expression of an old Benioff zone (Fig. 1).
Three volcano-plutonic associations are recognized within the island arc.
Each of these is characterized by its own metallogenic specialization. The
earliest, Middle Jurassic association is represented (Fig. 2) by the Early Bajocian
unimodal spilite-basalt formation developed upon the Toarcian-Aalenian sandy-
shale unit and reaching its maximum thickness in the Southern subzone; here, it
grades vertically into volcanic products of the suboceanic contrasting basalt-
rhyodacite formation of the sodium series (...rex = 15 - 200/0) 4. In the Northern
subzone these products are replaced by facies change to thick (up to 2900 m)
central-type volcanics (...rex = 75 - 80%) of calcareous-alkaline basalt-andesite-
dacite-rhyolite formation of the sodium series, filling there separate volcano-
tectonic uplifts and depressions. This Middle Jurassic volcanism is terminated by
the Bathonian basalt-andesite formation of the potassium-sodium calcareous-
alkaline series and by the emplacement of hypabyssal polyphase intrusions of
gabbro-plagiogranite formation.
The successive, Upper Jurassic volcano-plutonic association is represented by
two formations of a transitional sodium and sodium-potassium calcareous-alka-
line series: the earlier, andesite-dacite, and the later, gabbro-diorite-granodiorite-
adamellite-banatite formations (Fig. 2) developed exclusively in the super-
imposed troughs.
The Upper Cretaceous early island-arc volcano-plutonic association (Figs. 1
and 2) identified in the Bolnisi volcano-tectonic depression of the extreme south-
east termination of the Pontid branch of the island arc (Figs. 1 and 4) is composi-
tionally referred to calcareous-alkaline magmatites of the sodium and potassium-
sodium series (Fig. 2); products of trimodal volcanic formation are concentrated
in the depression proper, while those of plutonic gabbro-plagiogranite-grano-
diorite formation - along its periphery, in uplifts of the old basement (Khram
and Lok massifs) that terminate the depression (Fig. 4).
It should be stressed that the geotectonic evolution regime of the above
volcano-plutonic associations was centripetal in the volcano-tectonic uplifts,
being centrifugal in the depressions (Fig. 4). This being so, the volcanic activity in

4 L'ex = Coefficient of explosiveness.

334 N.K. Kurbanov

Fig. 4A - D. Type models for ore districts of the Lesser Caucasus island arcs. A Ore districts with
centripetal tectonic regime of ore-bearing-volcano-plutonic associations and the related base metal
deposits: A in plan, C in the section. B The same, with centrifugal tectonic regime: 8 in plan, D in the
section. 1 Basement rocks: 0 schists; b gneiss. 2 Andesite-basalt and andesite volcanics; 3 terrigenous
(0) and volcano-terrigenous (b) sediments; 4 reef limestone; 5 andesite-dacite volcanics; 6 rhyolite
and rhyodacite extrusions; 7 rhyolite and rhyodacite volcanic domes; 8 rhyolite lava and tuff; 9
rhyolite and rhyodacite subvolcanic intrusions; 10 hypabyssal intrusions of gabbro-plagiogranite for-
mation; 11 trachyandesite, trachydacite, trachybasalt dike belts; 12 Upper Jurassic-Lower Creta-
ceous undifferentiated volcano-sedimentary formations; 13 quaternary plateau basalt; 14 boundaries
of island arc (0), subsidence-calderas (b), and synvolcanic faults (c); 15 commercial-genetic types of
deposits: 0 pyrite, b pyrite-chalcopyrite-enargite, c copper-porphyry, d vein-quartz base metal

the volcano-tectonic uplifts started with the formation (Fig. 4) of a vast caldera
filled with andesite-basalt and andesite volcanics that subsequently were replaced
along the periphery by semiring structures of volcanic domes and multi vent ex-
trusions of rhyolite-dacite and rhyolite. The successive rhyolite subintrusive
facies intruded the central parts of the uplifting caldera, the formation of which
was terminated by the intrusion of gabbro-plagiogranite into the apical part of
the uplift.
In contrast, despite the similar character of the volcanism in the volcano-
tectonic depressions, their tectonic style is reversed: the evolution of silicic and
moderately silicic volcanism was centrifugal and was terminated by the emplace-
Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 335

ment of major polyphase intrusions along the periphery of a vast volcano-tec-

tonic depression showing a mosaic structural pattern.
The known base-metal ores of the Lesser Caucasus province are attributed to
three main commercial-genetic types, viz., pyrite, porphyry, and vein types that
are noted for higher concentrations of tungsten, tin, molybdenum, gold and
silver. All the listed base-metal ore types are related spatially and genetically to
the Bajocian-Bathonian and Upper Cretaceous volcano-plutonic associations of
the Somkheti-Kafan and Pontid island arcs (Fig. 1). The pyrite deposits include
(Kurbanov 1982, Tvalchrelidze 1977): (1) iron-pyrite and pyrite-chalcopyrite
hydrothermal-sedimentary deposits (the Cyprus type), associated with lavas and
jaspers of unimodal mafic rock formation; (2) iron-pyrite synvolcanic deposits
(the Urals type) associated with contrasting formation; and (3) synvolcanic cop-
per-pyrite deposits associated with the Bajocian basalt-andesite-dacite-rhyolite
sequences and localized in subsidence calderas, vents and slopes of extrusions,
and volcanic domes (the Lesser Caucasus type); (4) chalcopyrite-enargite (with
luzonite) pyrite postintrusive deposits linked with copper-porphyry ores; (5)
polygenetic copper-pyrite and barite base-metal deposits of various located in the
Bolnisi area and attributed to the Kuroko type (Tvalchrelidze 1977, Kurbanov et
al. 1978, Borodayevskaya et al. 1979) that are localized in subsidence-calderas,
volcanic domes, and volcano-tectonic depressions filled with volcanics of Upper
Cretaceous andesite-dacite-rhyolite formation (Fig. 5).
Permanently, following the emplacement of polymodal intrusions of gabbro-
granodiorite-plagiogranite sequences and their porphyry series, within the fields
of high-temperature (with cordierite) quartzites, stockwork zones of island-arc
(Kurbanov et al. 1978) copper-porphyry ores were developed.
According to the above-described evolution regime of volcano-tectonic up-
lifts and depressions, which correspond to ore districts (ranging in size from
350 km 2 to 400 km 2), the localization pattern of the various types of deposits is
centripetal and/or centrifugal. Pyrite deposits are concentrated in the volcano-
tectonic uplifts along the periphery of the ore districts (Fig. 4); closer to the
center of the uplifts, they are replaced by pyrite ores superimposed on subintru-
sive rhyolite facies. These, in turn, are replaced in the endocontact halo of the
gabbro-plagiogranite intrusions by pyrite-chalcopyrite-enargite ore deposits.
Finally, copper-porphyry deposits are concentrated in the center of the uplifts.
These deposits are localized directly among plagiogranites (Fig. 4). In the vol-
cano-tectonic depressions, the localization patterns of the above deposits within
the ore districts are reversed (Fig. 4).
Taking into account the above-stated facts the following conclusions can be
drawn:

1. Pyrite deposits of the Caucasian region are associated genetically and spatially
with conjugate pairs of structural-formational zones, viz., frontal troughs and
island arcs that emerged along the southern and northern bordes of the Trans-
caucasian microcontinent as ensimatic linear tension zones and activated
ensialic heterogenous blocks of the Earth's crust.
2. Hydrothermal-sedimentary copper-pyrite deposits are associated with uni-
modal mafic rock formations in rift trough zones, while in the island arcs they
336 N.K. Kurbanov

It

Fig. 5. Formative conditions of the Lesser Caucasus copper-lead-zinc deposits of various commercial-
genetic types. 1 Plagiogranite; 2 granodiorite-porphyry; 3 rhyolite and rhyodacite; 4 rhyolite tuff (a)
and ignimbrite (b); 5 andesite-basalt; 6 andesite-dacite and dacite; 7 tuffstone. tuffite and tuffmud-
stone; 8 copper-porphyry ores of the outer zone; 9 copper-porphyry ores of the inner zone; 10 copper-
pyrite ores of the breccioid structure; 11 veinlet-disseminated ores of quartz-pyrite-chalcopyrite and
quartz-pyrite-enargite composition; 12 ore clasts (fragments of pyrite ores) in tuff; 13 veinlets of late-
quartz base metal ores; 14 barite ores; 15 stratiform and metasomatic copper-pyrite ores; 16 massive
pyrite-chalcopyrite-enargite ores; 17 barite-base metal ores; 18 faults

are connected with bi- and trimodal types, manifesting a well-pronounced

polygenetic nature, different ages, and a copper-pyrite section.
3. The clustering and zonal localization pattern of the copper-pyrite deposits is
associated with the tectonic style of individual transverse blocks of the ensialic
island arcs. Two types, viz., centripetal (in the volcano-tectonic uplifts) and
centrifugal (in the depressions) are recognized in the volcanogenic arcs; in
active continental rises only one type is developed - vast inversion uplifts
consisting of local uplifts that appeared in the place of compensated paleo-
basins.
Copper-Pyrite and Pyrite Base Metal Deposits of the Caucasian Region 337

References

Borodayevskaya MB, Gorzhevsky DI, Krivtsov AI, Tvalchrelidze GA, Yakovlev GF (1979) Geotec-
tonic and paleotectonic types of pyrite-bearing provinces. In: World massive sulfide deposits.
Nedra, Moscow, pp 15 - 26 (in Russian)
Hutchinson RW (1973) Volcanogenic sulfide deposits and their metallogenic significance. Econ Geol
68:1223 -1246
Khain VE (1975) The main stages of the Caucassus tectono-magmatic evolution: geodynamic inter-
pretation practice. Geotektonika 1:13 - 27 (in Russian)
Kurbanov NK (1982) The main stages of formation of combined copper-pyrite deposits and their cor-
relation with evolution stages of the Alpine Great Caucasus terrigenous geosyncline. Tr TsNIGRI
Moscow Iss 168:3-18 (in Russian)
Kurbanov NK, Kulakov VV, Zaryanov YuP, Antonov VA (1978) Jurassic-Cretaceous igneous forma-
tions of the north-eastern Lesser Caucasus. SOy Geol 5:99 -113 (in Russian)
Tvalchrelidze GA (1977) Metallogeny of the main volcanic belt types. Nedra, Moscow (in Russian)
The Hidden Mineralogical and Geochemical Zoning and
the Ore-Forming Conditions of Copper and Copper-
Lead-Zinc Massive Sulfide Deposits
N. I. EREMIN 1

Abstract

The hidden mineralogical and geochemical zoning (HMGZ) of the repeated and
nonrepeated structural types is characterized by examples of different massive
sulfide deposits of Karelia, South Urals, Rudni Altai, and Japan. This zoning is
detected in different scale orders by means of microanalysis through the compo-
sition variations of the ore-forming sulfides: pyrite, sphalerite, and tennantite-
tetrahedrite. Is is shown that in the case of primary origin of such zoning its
nonrepeated type corresponds to volcanic-sedimentary ore deposition and
repeated type - to a hydrothermal replacement one. It is established that the
variations of Co/Ni values or Co-content in pyrite, Fe-content in sphalerite, As-,
Sb-, Ag-content in tennantite-tetrahedrite in cross-sections of the individual ore
bodies (ore intervals, rhythms) are mainly due to alteration of temperature and
sulfur activity during ore formation.

Introduction

Hidden mineralogical and geochemical zoning (HMGZ), as a regular space varia-

tion in composition of minerals within a geological unit, is extremely charac-
teristic of volcanic-associated massive sulfide deposits. This group of deposits is
very important economically and genetically; it includs pure pyrite, Cu-pyrite,
Cu - Zn-pyrite, and polymetallic-pyrite (i.e., Cu - Pb - Zn-pyrite) types of min-
eralization. In contrast to sediment-hosted massive sulfide deposits, which are
mainly Pb - Zn in composition, almost all the types given above include
economic contents of copper (in addition to a general, but fluctuating content of
zinc and sometimes of lead). An exception is the highly deformed and metamor-
phosed group of deposits which are characterized by morphological transforma-
tion and newly formed, mainly Zn and Pb - Zn ore segregations (i.e., meta-
morphism differentiates in the transformed ore bodies as a result of ore material
redeposition by hydrothermal-metamorphic solutions).

I Department of Mineral Deposits, Geological Faculty, Moscow State University, Moscow, 119899,
USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 339

A substantial part of the ores of the group of deposits considered here usually
forms massive stratiform bodies and consists predominantly of pyrite (or pyr-
rhotite) with variable content of base metal sulfides and various gangue minerals.
The scale of HMGZ in such bodies may be of four orders: in ore bodies as a
whole (first order); in individual intervals composed of a single mineralogical
type of ore (second order); in hand specimens of specific texture (third order);
and in individual mineral grains (fourth order). Nonrepeated and repeated struc-
tural types of HMGZ are the usual ones in the first and second orders. The non-
repeated type is characterized by a polarity of mineral composition from the bot-
tom of the ore body (ore interval) to its top, with the formation of a layerlike
structure. The repeated type reflects a symmetrical or asymmetrical concentric
change of mineral composition from the central part of the ore body (ore inter-
val) to its peripheral, marginal parts. According to geological data, nonrepeated
and repeated types of HMGZ in weakly metamorphosed deposits are charac-
teristic of volcanic-sedimentary and hydrothermal-replacement ore bodies, re-
spectively. The identical structural types of HMGZ in strongly metamorphosed
and dislocated deposits are of metamorphic origin.
In most cases the third order HMGZ is indicated in layered ores by a rhythmic
variation of chemical composition of some specific mineral corresponding
usually to the size of its grains. Sometimes rhythmic structure is determined for
ore intervals (HMGZ of the second order) in weakly metamorphosed ore bodies.
The fourth order HMGZ corresponds to the concentric phase heterogeneity iden-
tified sometimes in weakly metamorphosed volcanic-sedimentary ore bodies. It
indicates diffusion, interrupted growth of crystals, exsolution, and other phe-
nomena accompanying crystallization of ores with the participation of hydro-
thermal and metamorphic solutions.
The indicator minerals for HMGZ are pyrite, sphalerite, and tennantite-tetra-
hedrite typical of the ores; the variations of chemical elements in these minerals
are established by means of microanalysis.
For the reader's convenience all deposits considered here are listed in Table 1,
as well as the types of HMGZ exhibited in each of them.

HMGZ in the Pore Pyrite and Co-Pyrite Ores

In the massive, pure pyrite and Cu-pyrite ore bodies that are typical of the
Karelian, Cyprus, and some Urals types of deposits, the HMGZ is normally
established through the Co/Ni value or Co-content of the pyrite.
The steep lens of the Parandovskoe deposit (East Karelia) is composed of
recrystallized, metamorphosed (amphibolite grade), primary volcanic-sedi-
mentary pyrite ores of Proterozoic age containing hydrothermal-metamorphic
veinlets of pyrite-pyrrhotite composition with scattered grains of chalcopyrite
and sphalerite. Abundant crystalline-granular pyrite, resulting from metamor-
phic recrystallization of an early colloidal variety of the mineral, shows the reg-
ular fluctuation of its composition in the cross-section of the ore body. The mean
Co/Ni values of 39 individual analyses were calculated separately for presumably
340 N.!. Eremin

Table 1. List of deposits considered, their location, type, and HMGZ exibited

Types Deposits HMGZ a

Name Location A B C D E

Cu-(Zn) Parandovskoe East Karelia, pp nrp py Co/Ni

(Cyprus, USSR
Karelian) Shimokawa Hokkaido, Cu nrp py Co/Ni
Japan Cu cnc 4 py Co

Cu-Zn Osennee South Urals, Cu nrp py Co/Ni

(Urals) USSR
Uchali South Urals, Cu nrp 3 py Co/Ni
USSR (rhm)
Gai South Urals, Cu rp py Co
USSR
Oktjabrskoe South Urals, Cu-Zn rp sph Fe
USSR

Cu-Pb-Zn Tishinskoe Rudni Altai, pp nrp 3 py Co/Ni

(Altai, USSR (rhm)
Kuroko) pm rp sph Fe
pm rp tt As, Sb,
Ag
Gusljakovskoe Rudni Altai, pm rp sph Fe
USSR pm rp tt As, Sb,
Ag
Nikolaevskoe Rudni Altai, pm nrp 2 tt As, Sb
USSR pm cnc 4 tt As, Sb,
Ag
Shakanai North Honshu, Cu nrp 2 py Co/Ni
Japan (rhm)
Hanawa North Honshu, Cu rp py Co/Ni
Japan

a A Ore: pp pure pyrite; Cu copper pyrite, yellow; Cu - Zn copper-zinc pyrite; pm polymetallic

pyrite, Kuroko. B Structural type: rp repeated; nrp nonrepeated; rhm rhythmic; cnc concentric. C
Order; D indicator mineral; E indicator variable (value) in the mineral

upper (top), middle, and lower (bottom) parts of the ore body; they are as
follows: 2.5; 1.9; 1.2 respectively, and the gradient (change of this value for 1 m
of thickness) of this ratio is no more than 0.07. Such a noncontrasting and
nonrepeated HMGZ of the first order is a primary one, reflecting volcanic-sedi-
mentary origin of the ore body; its weakening occurred during metamorphic
recrystallization accompanied by decrease of Co- and Ni-content in the newly
formed crystalline-granular pyrite compared with the initial colloidal one
(Sergeeva 1973).
The ores of the Devonian Cu-pyrite Osennee deposit (South Urals) form mas-
sive subhorizontal beds and consist of pyrite with chalcopyrite, rare sphalerite,
and magnetite. The deposit is of volcanic-sedimentary origin (Potapenko et al.
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 341

Fig. 1. A section of the ore bodies,

Osennee deposit (nonrepeated HMGZ
of the first order through the variation
of the Co/Ni value in pyrite, see Table
2). Co/Ni values: 1 more than 2.0;
22.0-0.5; 3 0 .5 -0.2; 4 up to 0.2

Table 2. Variation of the Co/Ni value in pyrite, Osennee deposit, Cu-pyrite ore (data of laser micro-
analysis, Co- and Ni-content in n x 10- 3070)

Borehole Depth Co Ni Co/Ni Gradient

(m) (m) of Co/Ni

2630 89.2 78.0 5a 15 .6

89.5 20.0 5 4.0
91.0 11.0 5 2.2 1.2
91.9 2.5 5 0.5
102.1 2.0 5 0.4

2622 89.0 18.0 5 3.6

91.0 33.0 10 3.3 0.1
97.5 19.0 6.8 2.8
148.0 32.0 8.5 3.8
153.0 18.5 5 3.7 0.6
154.0 9.0 21.0 0.4

2624 89.0 27.0 5 5.4

90.5 18.0 5 3.6
92.7 23.0 5 4.6 0.3
101.0 9.5 5 1.9
102.0 9.0 5 1.8

2627 88.0 79.5 5 15.9

96.0 14.0 5 2.8 1.2
101.0 2.0 5 0.4

a In all these cases (when the Ni values are set at 5) the density of a selected analytical line (Ni 1
3414.77) measured by high resolution microphotometer (G-11 Karl Zeiss, Jena, GDR) on the
sensitive photoplate (ZU-2 ORWO) was distinctly and stably somewhat higher as compared with
neighboring background

1973), and its metamorphic level corresponds to greenshist-facies. In vertical drill

holes of the two ore bodies (Fig. 1) there is a marked decrease of the Co/Ni value
in pyrite from the top of each body to the bottom. The gradient of this ratio in-
creases sharply toward the flank margins. In general, the cross-section of each
ore body shows nonrepeated, marked HMGZ (Table 2) which reflects the vol-
canic-sedimentary genesis of this deposit.
The Mesozoic Cu-pyrite Shimokawa deposit (Japan) of the Cyprus type
(Eremin 1983) is an inclined, flat bed of volcanic-sedimentary sulfides (Bamba
342 N.!. Eremin

1977) located among the spilitized basalts. In some places it is complicated by

subsequent hydrothermal replacement mineralization and wall rock alteration
(both in the foot wall and hanging wall); the metamorphic transformation is a
very weak one, manifested mostly in dislocations. The principal ore minerals are
pyrite, chalcopyrite, pyrrhotite, and sometimes sphalerite. Cobalt is of economic
value and most of it is concentrated in the pyrite. The distribution of this metal in
the grains themselves is irregular within the range of O.OX to O.X wtOJo; in some
cases it forms concentric HMGZ of the fourth order. The high Co- and Ni-con-
tent and the irregular distribution of these metals in the pyrite grains have caused
a significant, often irregular variations of the Co/Ni value in the cross-sections
of the bedded ore body. Nevertheless in several traverses from the top to the
bottom of the ore body, the smoothed values of this ratio show a distinct mono-
tonous increase in the range of 7.2 - 25.0, i.e., here the HMGZ of the first order
is revealed (Eremin 1983). It should be pointed out that the degree of contrast of
this zoning is very high and its gradient in different traverses is identical
(2.8 - 2.9). Together with the volcanic-sedimentary nature of pyrite, the latter
confirms the primary origin of this HMGZ which was preserved in some parts of
the ore body.
For the three deposits mentioned above, the nonrepeated HMGZ of the first
order is a primary one and corresponds to volcanic-sedimentary deposition of the
ore bodies; its gradients decrease with increasing metamorphic grade and increas-
ing geological age.
Intervals of pure pyrite and Cu-pyrite ores occur also in the complex Cu - Zn-
pyrite and polymetallic-pyrite ore bodies of the Urals, Altai, and Kuroko types.
In such cases it is possible to find HMGZ through Co/Ni values and Co-content
in pyrite, but now they are of the second and third orders.
At the southern flank of the Cu-pyrite ore body of the weakly metamorphos-
ed Uchali deposit (South Urals) there are rhythmically-layered, volcanic-sedi-
mentary textures in the pure pyrite ores. This layering is due to the difference in
the grain size of pyrite and the total quantity of the mineral (graded bedding).
This picture is complicated by discordant and rare concordant diagenetic veinlets
of pyrite 1 - 2 mm thick. In some places the layers are distorted with atectonic
dislocations of block movement and slump folding. Each rhythm combines two
neighboring layers with a distinct border. The bottom layer, 0.5 - 0.6 cm thick,
contains up to 60 - 65% of pyrite and the average size of its grains is 0.1 - 0.2
mm. The top layer up to 0.5 cm in thickness contains an apparently lesser
quantity of pyrite (40 - 50070) and the size of its grains ranges between O.OOX and
0.1 mm. The principal gangue mineral is quartz. Though the fluctuations in Co-
and Ni-content for individual grains of pyrite are irregular, the decrease of
Co/Ni values from bottom to top of each rhythm (with the greatest range in one
of them 36.0 -1.0) is clearly seen, i.e., the nonrepeated rhythmic HMGZ of the
third order is recorded.
Similar pure pyrite layered ores are recorded at the highly metamorphosed
and dislocated Tishinskoe deposit (Rudni Altai). In addition to intensive folding
of varied scale, shearing and cleavage of the host rocks, and the elongate form of
the ore body, the coarse-grained, mainly sphalerite-galena layers located in the
margins of the massive ores are here very clearly seen (as a result of remobiliza-
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 343

Fig. 2. A specimen of pure pyrite

rhythm-layered ore, Tishinskoe
deposit: I-IV rhythms; a bottom
iY parts of rhythms; b top parts of

J.~~~~~~"'9~i~~a!~~~~~, rhythms
third (rhythmic
order HMGZ
through the of the
variation of
the Co/Ni value in pyrite; see Table
ill 3, the sample numbers in this Table
correspond to those in figure)

I
cmO~. ________~__________~~________~3cm

a~b~
~ t=.::j

tion of ore material). The impregnated (semimassive) pure pyrite ores under con-
sideration form thin, elongated, and folded concordant lenses outside the main
ore body among the quartzite and quartz-sericite-chlorite rocks. The layering in
these ores is due to the alternation of sericite-chlorite and quartz-sericite-chlorite
layers of differing pyrite content and regular change in the pyrite grains. Each
single layer is 0.5 - 3.0 em thick with a regular decrease both in grain size in the
range of tenths to hundredths of a millimeter (graded bedding) and in total
quantity of pyrite from the bottom to the top. This is the cause of the rhythms
(Fig. 2).
In this case, as with the Uchali deposit, the distinct decrease in Co/Ni values for
pyrite grains from bottom to top of each rhythm is also established (Table 3); this
corresponds to HMGZ of the third order. The disseminated mode of occurrence of
pyrite among the hydrothermal-sedimentary rocks is the reason for preservation of
both the pyrite grain size and the content of metals in them, notwithstanding their
recrystallization during metamorphism. This is the reason for the almost equal
degree of contrast of the zoning in question in these two deposits.
Very impressive nonrepeated HMGZ of the second order is shown in the Mio-
cene Shakanai deposit (Japan) of the Kuroko type. The fourth ore body of this
deposit is a sub horizontal layer of black (polymetallic-pyrite) ores in its upper
part and yellow (Cu-pyrite) ores in the bottom part. Bore hole No. 756 cuts this
body and reveals four intervals of yellow ore. The regular variation of Co/Ni
values in individual grains of pyrite is apparently shown; for each interval this
ratio sharply increases from top to bottom (Table 4). The thickness of each inter-
val of neighboring, massive pyrite-chalcopyrite ores is up to 1.5 m. These inter-
vals are separated from each other by layers of highly altered acid tuffs. This
nonrepeated HMGZ corresponds to (Le., agrees with) the volcanic-sedimentary
origin of these ores; this idea is supported by the majority of investigators of the
Kuroko-type ore deposits (e.g., Matsukuma and Horikoshi 1970, Tatsumi and
Watanabe 1971, Lambert and Sato 1974, Smirnov 1976, Large 1977).
344 N.I. Eremin

Table 3. Variation of the Co/Ni value in pyrite, Tishinskoe deposit, pure pyrite ore (data of laser
microanalysis, Co- and Ni-content in arbitrary numbers)

Sample Co Ni Co/Ni Sample Co Ni Co/Ni Sample Co Ni Co/Ni

Nos. Nos. Nos.

7 4 3 1.3 4 3 3 1.0 1 2 2.0

8 5 2 2.5 5 3 3 1.0 2 2 2.0
9 5 3 1.7 6 5 2 2.5 3 5 5.0
22 4 2 2.0 19 5 1 5.0 16 8 8.0

20 3 3 1.0 17 3 5 0.6
14 3 2 1.5 21 2 2 1.0 18 4 3 1.3
12 4 2 2.0 10 3 2 1.5
15 5 5.0 13 5 2 2.5 11 5 2 2.5

25 7 6 1.1 37 2 2 1.0 30 4 9 0.5

26 7 6 1.1 38 2 2 1.0 31 4 9 0.5
27 5 2 2.5 39 3 1 3.0 32 3 2 1.5
40 4 1 4.0 33 3 1 3.0
28 8 2 4.0 34 4 1 4.0

29 4 2 2.0 23 3 2 1.5 35 4 3 1.3

24 3 1 3.0 36 6 4 1.5

Table 4. Variation of the Co/Ni value in pyrite, Shakanai deposit, intervals of yellow ore in bore hole
No. 756 (data of laser microanalysis, Co- and Ni-content in n X 10- 3 070)

Intervals Sample Distance from Co Ni Co/Ni Gradient

(from the Nos. the bore hole of Co/Ni
stratigraphic collar in the
top to the (m) interval
bottom)

244 17.7 4.7 5.9 0.8

245 17.2 5.8 4.1 1.4 2.3
246 16.8 5.2 1.8 2.9
II 247 16.4 4.0 8.8 0.4
'248 15.9 4.0 2.7 1.5 1.3
249 15.5 5.0 3.2 1.6
III 253 6.9 4.0 7.8 0.5
254 6.4 3.4 1.9 2.8
1.8
IV 256 5.2 4.3 9.2 0.5
257 4.8 4.3 3.2 1.3 2.6
258 4.4 3.4 1.3 2.6

The repeated structural type of HMGZ established through Co/Ni values or

Co-content in pyrite is uncommon for the massive pure pyrite and Cu-pyrite ores
referred to.
The steep dipping Cu-pyrite bodies of the deep horizons at the Gai deposit
(South Urals) are of the hydrothermal replacement origin as compared with the
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 345

uppermost volcanic-sedimentary Stergnevaja lens. They mark the ore-controlling

structure (Eremin 1983). The mode of the Co-content distribution (Ni is not
detected) in pyrite (up to 900/0 in the ore) corresponds to HMGZ of the first
order, which is seen in the cross-section of three ore lenses. In one lens this
zoning is a relatively simple, concentric one with the maximum Co-content (more
than 30 units) in the form of an elongated, slightly curved stripe along the central
axis of this lens. Each of the two other lenses has two parallel maxima (com-
plicated concentric zoning). The obtained temperatures of crystallization in the
range 300-120°C for these ores (using the distribution of Co-content between
coexisting pyrite and chalcopyrite, Bezmen et al. 1978) mark the concentric-
zoned structure of a paleothermal field. The elongated maxima of this structure
are traced concordantly in the general direction of the ore lenses coinciding with
the maxima in distribution of the Co-content discussed above.
A similar picture of the space distribution of Co/Ni values in pyrite has been
established in the Motojama ore body of the Kuroko-type Hanawa deposit
(Japan). This is a subvertical pipelike body of massive yellow ore consisting of
two neighboring shoots named Motojama Shinko Sho and Higashi Daijiti.
Detailed sampling of these shoots along the walls and the roofs of workings at
four horizons has been carried out. At the planes of each shoot the complicated
concentric structure of HMGZ (Eremin 1983) is indicated by one to two isometric
maxima, with several isolines of the Co/Ni value (5; 10; 15; 20; 25). Bearing in
mind the primary origin of the analyzed crystalline-granular pyrite, we conclude
that the genesis of this HMGZ of the first order is primary. It is suggested that
the yellow ore of the Motojama ore body was formed by hydrothermal replace-
ment of an ore-controlling vent structure just below the seafloor.
Thus, the repeated structural type of HMGZ established through the distribu-
tion of Co-content or Co/Ni values in pyrite of the Gai and Hanawa deposits
corresponds to hydrothermal replacement deposition of these ores.

HMGZ in the Cn - Zn-Pyrite and Polymetallic-Pyrite Ores

In the Cu - Zn-pyrite and polymetallic-pyrite ore bodies HMGZ of both struc-

tural types is established on the basis of the Fe-content in sphalerite and the fluc-
tuation in composition of tennantite-tetrahedrite. It is manifested most impres-
sively in some deposits of the Kuroko, Altai, and, rarely the Urals types.
The regular space fluctuation of Fe-content in sphalerite during the forma-
tion of the repeated type of HMGZ is characteristic of the first ore body of the
Oktjabrskoe deposit (South Urals). This body consists of massive sulfide ore and
is located in volcanogenic rocks of Paleozoic age. Its top contact is concordant
with the layering of host rocks, but its bottom one is complicated, with steep
dipping apophyses. In vertical cross-section, the body is of mushroom shape, and
in longitudinal ones it is inclined 15 - 20°. The principal ore minerals are pyrite,
chalcopyrite, and sphalerite. The maximum Fe-content in sphalerite (more than
1.5%) is found in the central, thickest part of the body, whereas the minimum
Fe-content (up to 0.25%) is found on its flanks and marginal parts. The picture is
346 N.!. Eremin

clearly seen both in the cross- and longitudinal-vertical sections (Eremin 1983).
According to geological data sphalerite mineralization is the latest one at the
deposit, being formed as a result of the reaction between Zn-containing hydro-
thermal solution and Cu-pyrite ores (Malachov et al. 1974).
The increase in Sb- and Ag-content in tennantite-tetrahedrite toward the top
of the ore bodies provides evidence of a distinct nonrepeated HMGZ in the black
ore of a majority of Kuroko deposits (Matsukuma and Horikoshi 1970). A de-
crease in Fe-content in sphalerite (HMGZ of the first and second orders) is noted
in the same direction for these ores at the Shakanai deposit (Kajiwara 1970). It
was found also (Urabe 1974) that the Fe-content in sphalerite decreased toward
the top of complex bodies composed of several types of sulfide ores (HMGZ of
the first order).
The principal ore constituents at the Gusljakovskoe deposit (Rudni Altai) are
syngenetic pyrite followed by polymetallic mineralization with sphalerite, galena,
chalcopyrite, and tennantite-tetrahedrite. Tennantite-tetrahedrite compositions
across the subvertical ore zone of mainly impregnated sulfides are concentrically
zoned, i.e., the more Ag-rich tetrahedrite varieties of the mineral are located in
the central part, while tennantite varieties occur in the marginal parts of the ore
zone (the Sb-content fluctuates in the range of 26.37 - 4.23070; the Ag-content, in
the range of 5.98-0.15%; and the As-content, in the range of 4.67-18.36%).
The distribution of the Fe-content in sphalerite is also characteristically zoned:
the most Fe-rich varieties of the mineral are located in the axial part of the ore
zone. Concentric HMGZ of the first order, defined by the composition of
tennantite-tetrahedrite and sphalerite, definitely corresponds to the structure of
the paleothermal field identified by the decrepitation of sphalerite. The axis of
temperature maximum (300 - 360 0 C) coincides also with the central part of the
ore zone (Zolotarev et al. 1975). According to geological and geochemical data,
the polymetallic mineralization followed hydrothermal alteration which in turn
succeeded pyrite impregnation (pyrite is recrystallized in the zone of alteration).
The volcanic-sedimentary Kreshenskoe ore body (Nikolaevskoe deposit,
Rudni Altai), which is weakly metamorphosed and, hence, likely to be preserved
in its original position, is comparable to the Kuroko deposits in the mode of oc-
currence of different ore types. The dominant massive ores of this body are
divided (mainly according to the data of chemical sampling) into the following
types (from the bottom to the top): pure pyrite and Cu-pyrite, Cu - Zn-pyrite,
and polymetallic-pyrite. There are no distinct variations in the composition of
tennantite-tetrahedrite, which is all of tennantite variety at this deposit (in con-
trast to the Kuroko deposits). Nevertheless, in the polymetallic-pyrite interval (as
in the Kuroko ores) nonrepeated HMGZ of the second order is marked. Here we
have a decrease in Sb-content (from 0.11 to 0.04%) and a corresponding increase
in As-content (from 14.02 to 20.32%). In these ores there are also single grains of
tennantite-tetrahedrite with concentric-zoned structure inside each grain due to
the internal chemical heterogeneity (HMGZ of the fourth order).
The main ore body of the Tishinskoe deposit mentioned above consists of
dominant veinlet-impregnated ores and massive polymetallic-pyrite ores. There is
a regular fluctuation in composition of sphalerite and tennantite-tetrahedrite in
vertical cross-section (Eremin 1983). The maximum Fe-content in sphalerite (up
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 347

to 2.40070) is in the central part of the body at the level of the thickest massive ore.
This content decreases to 0.10% toward the flanks and both (upper and lower)
ends of the body. Thus, in this cross-section of the vertical body the elongated
concentrically-zoned distribution of Fe-content in sphalerite is concordantly
outlined with maxima in the upper horizons, which are richer in the massive ores.
In this cross-section the same structure is outlined by the variation in com-
position of the tennantite-tetrahedrite: the maxima of Sb- and Ag-contents are
found in those grains of the mineral that are located in the thickest ,portion of the
massive ore as mentioned above. The isolines of mineral composition form a
concentric asymmetric pattern extending along the axes ofthe ores body. Toward
the flanks the composition of the mineral becomes more arsenical. The range in
mineral composition is very broad: As 1.00-18.33%; Ag 7.05 -0.15%; Sb
31.01 - 0.94%. This is responsible for the strong zoning. In many respects this
HMGZ resembles that described for the Gusljakovskoe deposit. This possibly
reflects the similar condition of ore formation at both deposits. But it is quite
possible that at the Tishinskoe deposit this zoning may result from metamorphic
transformation of the ore body.

The Ore-Forming Conditions

In the cases when the HMGZ is primary, it reflects the conditions of deposition
of the ore bodies. Relicts of such zoning could be found even in highly meta-
morphosed deposits.
Keepling in mind the more common occurrence of the nonrepeated, some-
times rhythmic HMGZ of the different orders as compared with the repeated
one, it could be concluded that most of the pure pyrite and Cu-pyrite ore bodies
have been formed in a volcanic-sedimentary manner on the seafloor. In this case
(Shimokawa, Uchali, Tishinskoe, and Shakanai deposits) we see more often a
decrease in the Co/Ni values in pyrite toward the top of such bodies (intervals,
rhythms) and more seldom - an increase in this ratio (Parandovskoe, Osennee
deposits). A decrease of temperature toward the top of the ore bodies (intervals,
rhythms) is assumed to occur during volcanic-sedimentary ore deposition. This
has been proved particularly for the Kuroko deposits (Large 1977).
Following Bezmen and Tichomirova (1975) it was shown by the author
(Eremin 1983) that a decrease in temperature of chlorine-bearing solutions of
constant composition causes a decrease in the Co/Ni values in coexisting pyrite
below the vaesite-millerite equilibrium line in the diagram log 0S2 - T, or an
increase in this ratio above this line. In other words, the volcanic-sedimentary ore
deposition at the Parandovskoe and Osennee deposits took place under condi-
tions of relatively high activity of sulfur.
One can always a decrease in the Co/Ni values (or Co-content) in pyrite from
the center to the margins in the hydrothermal replacement Cu-pyrite ore bodies
(the deep horizons of the Gai deposits, the Motojama body of the Hanawa
deposit); this is also in accordance with similar lowering of the temperature of
mineralizing solutions during their diffusion away from the axis of the ore-con-
348 N.1. Eremin

trolling fissure structure. For the Gai deposit this is confirmed by the paleo-
thermal zoning structure. Consequently, in the case of both these deposits the
formation of pyrite, as a basic part of the ores, occurred at the comparatively low
sulfur activity below the equilibrium line vaesite-millerite in the diagram
log as2 - T.
The successive volcanogenic-sedimentary deposition of the black ore, which
was deposited after the yellow ore at the Kuroko deposits, is confirmed by geo-
logical data and correlated with the nonrepeated HMGZ established through the
decrease of Fe-content in sphalerite and increase of Sb- and Ag-content in ten-
nantite-tetrahedrite toward the top of the ore bodies. Since the deposition of
these ores occurred during a decrease in temperature, the observed decrease in
the Fe-content in sphalerite, as it follows from the diagram log aS2 - T (Barton
and Skinner 1967), points out the increase in the sulfur potential (although the
sulfur activity was lowered, the vector of the ore formation was shifted toward
equilibria with participation of more sulfide sulfur). Deposition of black ore in
this manner conforms with the HMGZ identified by the compositional fluctua-
tions of the tennantite-tetrahedrite. In the diagram log aS2 - T, in the area of
relatively low sulfur activity, tennantite and Sb-sulfosalts are the stable phases,
whereas under increasing sulfur activity, the antimony combines more readily in
the form of tetrahedrite and arsenic in the form of enargite. The direct correla-
tion between antimony and silver is due to the crystallochemical peculiarities of
the mineral.
In the hydrothermal replacement Cu - Zn-pyrite and polymetallic-pyrite ores
(Oktjabrskoe, Gusljakovskoe and, possibly, Tishinskoe deposits), decrease in the
Fe-content in sphalerite related to the lowering of the temperature of the solution
diffusing away from the ore-controlling fissure structure is also seen. But the
change of the tennantite-tetrahedrite composition here is opposite the change
observed in the volcanic-sedimentary black ore. The more tetrahedrite-rich
varieties of the mineral with high Sb- and Ag-content are located in the high-tem-
perature (central parts of the bodies; the tennantite varieties, in the low-tempera-
ture (marginal) parts. It is possible that in this case the determining factor of the
HMGZ formation was likely to be more the different mobility of As and Sb ions
during metasomatism than the change in temperature and sulfur activity in the
diffusing solution. It is interesting to note that for these ores the heterogeneity of
the tennantite-tetrahedrite grains themselves (HMGZ of the fourth order) is not
typical. But this phenomenon is more or less ordinary for the volcanic-sedimen-
tary black ores in Kuroko deposits, and is found in individual cases in the poly-
metallic-pyrite ores of the Nikolaevskoe deposit.

Acknowledgments. The author is grateful to Academician v. I. Smirnov and to all the staff members
of the Department of Mineral Deposits (Moscow State University) for support in this work. Special
thanks to Mrs. T. S. Maximova for assistance in translation of the manuscript into English. The
author also wishes to acknowledge Dr. A. D. Genkin and all foreign colleagues for helpful sugges-
tions and criticism.
The Hidden Mineralogical and Geochemical Zoning and the Ore-Forming Conditions 349

References

Bamba T (1977) Metallogenic province of Hokkaido. J Fac Sci Hokkaido Univ Ser IV, 17:695 -708
Barton PB Jr, Skinner BJ (1967) Sulfide mineral stabilities. In: Barnes HL (ed) Geochemistry of
hydrothermal ore deposits. Holt, Rinehart and Winston, Inc, New York, pp 236 - 333
Bezmen NI, Tichomirova VI (1975) The effect of temperature on the distribution of Co and Ni
between the Fe-sulfides and the solutions of different composition. Geochemia 11: 1691 -1697 (in
Russian)
Bezmen NI, Eremin NI, Narazauly IG, Pozdnyakova NV, Sergeeva NE (1978) Pyrite-chalcopyrite
geothermometer: cobalt distribution. Geochemia 3:384-389 (in Russian)
Eremin NI (1983) The differentiation of the volcanogeneous sulfide ore-formation. MGU, Moscow,
256 pp (in Russian)
Kajiwara Y (1970) Some limitations on the physico-chemical environment of deposition of the
Kuroko ore. In: Tatsumi T (ed) Volcanism and ore genesis. Univ. Tokyo Press, Tokyo, pp
367-380
Lambert IB, Sato T (1974) The Kuroko and associated ore deposits of Japan: a review of their
features and metallogenesis. Econ Geol 69:1215 -1236
Large RR (1977) Chemical evolution and zonation of massive sulfide deposits in volcanic terrains.
Econ Geol 72:549 - 572
Malachov AA, Vorobjev VV, Malachov DA et al. (1974) On the complex processes of formation of
the Oktjabskoe Cu-pyrite deposit (South Urals). Geol Rudn Mestorogd 3:40 - 50 (in Russian)
Matsukuma T, Horikoshi E (1970) Kuroko deposits in Japan. A review. In: Tatsumi T (ed)
Volcanism and ore genesis. Univ Tokyo Press, Tokyo, pp 153-179
Potapenko BP, Trebuchin VS, Ageeva ST (1973) Dombarovskii ore region. Trudi ZNIGRI
105:178-188 (in Russian)
Sergeeva NE (1973) Pyrite from the massive sulfide deposits of Karelia. Vestn MGU GeoI5:100-105
(in Russian)
Smirnov VI (1976) The geology of mineral deposits, 3rd edn. Nedra, Moscow, 688 pp (in Russian)
Tatsumi T, Watanabe T (1971) Geological environment of formation of the Kuroko-type deposits.
In: Proc IMA-IAGOD Meet, Tokyo-Kyoto, 1970, IAGOD Vol Spec Iss 3. Inst Assoc Genesis Ore
Deposits, Tokyo, pp 216-220
Urabe T (1974) Iron content of sphalerite coexisting with pyrite from some Kuroko deposits. In:
Ishihara S (ed) Geology of Kuroko deposits. Min Geol Spec Iss 6. Soc Min Geol Jpn, Tokyo, pp
377-384
Zolotarev VG, Eremin NI, Dergatchev AL, Polenkov AI (1975) The correlation of metasomatic,
geochemical and thermal zoning at the polymetallic-pyrite Gusljakovskoe deposit (Rudni Altai).
In: Magakjan IG (ed) Metasomatism and massive sulfide ore-formation. Erevan, pp 111-113 (in
Russian)
Zoning of Massive Sulphide Deposits and Their Origin
N.S. SKRIPCHENKO, V.1. SHCHEGLOY, G.V. RYABOY, and V.L. ANDREYEy 1

Abstract

Vertical and lateral zoning of massive sulphide deposits is well known. Vertical
zoning is characteristic of volcanic-hosted deposits (e.g. Kuroko), whereas lateral
zoning occurs in sediment-volcanic-hosted ore (e.g. Zhairem). A dynamic model
of such zoning would belong to the "barrier reaction" type. An oxide-acidic
model is herein proposed for the volcanic-hosted and a reduced-carbon model
for the sediment-volcanic-hosted deposits. The first model relates to the
boundary of the metalliferous solutions/oxidized water system. Oxidized sulphur
components in this system range from S02 to S2-. The acidity and ratio of
S2- /SO~- are increased and decreased, respectively, from the base to the upper
part of the ore section. The result of these variations is the following sequence
(from bottom to top): sulphide-silicate metasomatites, pyrite-, copper- and lead-
zinc-pyrite massive ores. The form and size of these ore types is caused by the
dynamics of the hydrothermal flow and the size of the recycling zone. According
to the reduced-carbon model, the ore is deposited from the hydrothermal brine
into depressions on the seafloor. Deposition is controlled by the brine/seawater
contact, where sulphate-reduction takes place. The laterally concentric emplace-
ment of lead, zinc and pyrite ores is caused by depletion of the sulphidic ion and
concentration of the metallic ions in the brine pool.

Massive sulphide copper, zinc and lead deposits are hosted by basalt-carbona-
ceous-argillaceous megaformations. Ferrous disulphides and some baryte are the
dominant minerals of their ores. Ferric and manganese oxides are minor
constituents. As a rule, the classification of massive sulphide deposits reflects the
composition of their ores and host rocks (Borodaevskaya et al. 1979). If not
taken into consideration the morphology-combined multimodula ore bodies, the
inner structure of the massive sulphide bodies may be classified by only two types
of zoning: the vertical and lateral (horizontal). This zoning is the basis for
subdivision of the massive sulphide deposits into two genetic classes: vertically
and laterally zoned deposits. Examples of the vertical zoning type are the Kuroko
deposits (Fig. 1), and of the horizontal type, the ore bodies of the Zhairem
deposits (Fig. 2). The genetic models of both types of zoning are examined
below.

1 The Novocherkassk Polytechnica1 Institute, Mining-Geological Department, Novocherkassk, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Zoning of Massive Sulphide Deposits and Their Origin 351

- / - \ - / - • - /- , - I - \ - I - 0 - \ -1 - _I - \

......_
/ 1- _' 1, -\ _/ 1-_\ \-_
I -
1
\ r- I - \ - / - \ ~
_ " _ 1 _ \ __ 1_ \ 1 / _' _1_ \ _' . . ' & ~OO
\ -I - \ - I - ~­
100 100
I

- -.....: . .

: + '<~Li'-I · i .·.· • .•..•. ;. ..:-

\\
+ -t- +

+
~
-+- ~- 1
'-t-~' ~ ~4HlO~~S :,3~2SU;~
':=-'-'-- •

.SOzfe&u. In.ps.
b~j1t~2ITIJ3~4~5CJ6[J]7[Q]3 f
Fig. 1. Structure and genetic model of vertically-zoned massive sulphide deposits of the Kuroko type.
1 Unaltered rocks of the post-ore time; 2 volcanic rocks; 3, 4, 5,6 massive lead-zinc, copper, pyrite
and disseminated silicate-pyrite ores, respectively; 7 descending exogenic flow saturated with oxygen;
8 ascending ore-bearing flow. The diagram shows the limits of pH changes and the limits in mineral
formation temperatures (T 0c), and also suggests trend curves for concentrations of sulphide (Sr) and
sulphate (So) and for oxygen fugacity (f02)

Fig. 2. Characteristic lateral zoning in Zhairem-type ore deposits. The ore-mineral composition of
zones is shown on the horizontal projection of the deposit (be/ow). The curves of the upper diagram
indicate variations in metal content in a cross-section of the ore body
352 N. S. Skripchenko et a1.

Vertical Zoning

The complete zoning of this type is represented by a column where pyritization

metasomatite, essentially pyrite, copper- and lead-zinc-pyrite ores, occur in suc-
cession upwards (Fig. 1). The data characterizing the structural and genetic
features of the vertical zonal ore deposits are given below.
The thickness of sulphide mineralization in sections through the deposits is
very small (from a few to 100 m) in comparison to the distance from the source of
ore-bearing solutions. The lithological varieties in ore-bearing sections do not
control the ore zone location or the degree of differentiation. The ore-hosted
rocks are represented by lava, subvolcanic, pyroclastic-sedimentary and sedimen-
tary complexes. The local ore-controlling forms of paleorelief are defined as
deeps in basins with characteristic iron-oxide and, more rarely, manganese syn-
genetic mineralization accompanying the massive sulphide ores. The data
confirm the suggestion that massive sulphide ores were deposited in basinal
depressions saturated with oxygen.
Stratiformlike mineral-component differentiation of the ore bodies and
asymmetric location of various genetic types of ores (metasomatic in the lower
and hydrothermal-sedimentary in the upper parts of the deposit) define the ore
deposits as a genetically combined type (Skripchenko 1966, Smirnov 1968). The
ratio of thickness of hydrothermal-sedimentary to hydrothermal-metasomatic
types of ores varies widely. Hydrothermal-sedimentary ore is characterized by a
high sulphide: lithogenic mineral ratio (usually from 1 to 10). The degree of
lithogenic components leached and the accompanying sulphide deposition,
especially pyrite, in the metasomatic part of section increases from bottom to
top.
Vertical zoning belongs to the facial zoning type according to Smirnov (1982).
The origin of this zoning is explained by ore sedimentation at the time of dis-
charge of hydrothermal solutions into exogenic basins. The projections of the
pyrite metasomatite of the ore footwall defines the discharge area. A number of
workers (Skripchenko 1972, Smirnov 1960,1968) prove the convergent character
of massive sulphide ores. Isostructural sulphide aggregates generated under dif-
ferent P - T conditions are associated in cross-sections of ore deposits. Mineral
thermometric measurements in massive sulphide ores established the limit from
40 0 to 350 0 (Bogysh and Trufanov 1983, Smirnov et al. 1968). These data
demonstrate that temperature fluctuations are not the only factor in the process
of massive sulphide ore formation and differentiation, although it may have an
indirect influence (Tvalchrelidze and Yaroshevich 1981, Large 1977, Large and
Both 1980). The main cause of vertical differentiation and sulphide deposition,
in general, is the existence of intensive chemical reactions at the contact of
concentrated hydrothermal solutions and exogenic water. Reactions become
active as a result of the high oxidizing potential of exogenic water.
Sulphide profiles of ore formation, characterized by deposition of large
masses of iron disulphide, raise a question about the nature of the sulphide ion.
Data from Holland and Malinin (1982), Stolajarov (1980) and Barsukov et al.
(1981) show that the main, and possibly the only, form of sulphur present in
endogenic hydrothermal systems during their generation is S02. Consequently,
 Zoning of Massive Sulphide Deposits and Their Origin 353

A-8

0,0

Eh -0,1
-0,2

-0,

-0,4 1 2 3 4 5 6 7 pH
pH
Fig. 3. Change in S2- concentrations depending on pH and Eh of aqueous solutions (T = 25°C,
P = 1 atm, S2- +SO~- = 0.1 m). The diagram on the right is the section AB of Eh-pH diagram
(Skripchenko 1966)

the massive sulphide deposits genetic model may be based on the assumption that
the primary ore-bearing solution contains S02 as the main sulphur component.
In accordance with Holland and Malinin's data (1982), cooling of hydrothermal
solutions is accompanied by the reaction: 4S0 2+4H 20 = H 2S+3H 2S04 • A re-
sult of this reaction is an increase in the fugacity of the sulphide ion S2-(Sr) as
well as the sulphate ion S6-(SO)' This reaction may be supported by the thermo-
dynamic calculation of the S2- /SO~- ratio on the Eh-pH diagram (Skripchenko
1966). As indicated in Fig. 3, S2- concentrations increase with decreasing pH of
the aqueous solution. At a constant isopotential (Eh) level, the S2- /SO~- ratio
increases from pHmax to pH min (Fig. 3).
Decrease in pH in the discharge zone of ore-forming hydrothermal solutions
results from interaction of these solutions and the surface oxygen-saturated
waters. The possible geochemical parameters in a section of the reaction zone are
given in Fig. 1. This section of mixed zone, also known as the ore-forming zone
or "worked" zone according to Pospelov (1973), is subdivided into three parts
corresponding to three types of mineralization (from bottom to top): (1) Metaso-
matic silica-pyrite ores with low copper content; their formation is related to
growth of the ratio Sr/SO and intensified leaching of wall rocks caused, in
general, by a progressive increase in acidity. If the process proceeds according to
the "recycling" model, the shape and thickness of the metasomatite zone will
vary (Fig. 4). (2) Weakly or non-differentiated massive copper and essentially
pyrite ores. These ores are formed in a zone characterized by maximum acidity
and concentration of Sr. (3) Differentiated ores with high copper, zinc and lead
content. They are formed in the upper section of the worked zone where, it is
suggested, there exists a deficit of Sr. The high content Cu, Pb and Zn in these
ores is explained by variations in the bond strength of metal and sulphide ions.
The bond strength of the non-ferrous metal ions is more than that of the ferrous
354 N. S. Skripchenko et at.

Fe,Cu,Zn
DB, Ba.,Mn

Fig. 4. Genetic relations and zoning of massive sulphide deposits of the Kuroko type (1 ascending
flow; 2 recycling flow), and zoning of pyrite deposits of the Zhairem type (3). Explanations: 1 poly-
metallic; 2 copper; 3 pyrite; 4 silicate-pyrite ores; 5 baryte; 6 baryte-galena; 7 baryre-galena-
sphalerite; 8 galena-sphalerite; 9 sphalerite; 10 pyrite ores; 11 seawater; 12 hydrothermal brine;
13 flow of endogene ore-bearing solutions; 14 direction of flow of exogenic oxygen-saturated
solutions; 15 zones of inhibited flow of endogenic solutions

ion. Solutions containing only minor amounts of the sulphide ion must precipi-
tate Pb - Zn - eu sulphides only.
The morphological and thickness parameters of ore bodies are defined by a
number of factors; among these are local seafloor relief and the speed and form
of the hydrothermal solution flow. The major control on ore body morphology is
the structure of metal-bearing solution channels. Two structural varieties of
simple ore bodies (1 and 2) are given in Fig. 4. Interpretation of the second
variety assumes the possibility of lowering of mixing front to considerable
depths. This is caused by the variation in the morphology of the hydrothermal
ore-solution flow. Mixings, of the recycling nature, promote an increase in the
thickness of the pyritization metasomatite zone. The latter changes its form from
a gently sloping, conformable lenticle (in the first variety) to vertical and funnel-
shaped (in the second variety). One can observe morphological variations of the
pyritization metasomatite zone in various ore bodies within the limits of a single
ore field and even within the limits of one multimodular ore body.

Lateral Zoning

One can observe lateral zoning best in Zhairem type deposits occurring in black
siliceous-clay-lime flyschoid strata series. In this case vertical sections of the ore
bodies are homogeneous. Ore zones replace each other in concentric fashion out-
ward from the centre as follows: siliceous-baryte, baryte-galena, galena-
sphalerite, sphalerite, pyrite (Fig. 2). Ore bodies and ores are characterized as
follows: (1) (non-metallic minerals): (ore minerals) ratio gradually increases from
Zoning of Massive Sulphide Deposits and Their Origin 355

the centre to the periphery. The ore-hosted section of flyschoid is thinly bedded.
As a rule, ore layers are characterized by high metal concentrations. (2) Absence
or rare occurrence of footwall alteration zone. (3) Competing sulphide-oxide as-
sociation among iron mineral assemblages. One can observe an increase of sul-
phides in ores from the centre to the periphery of ore bodies. The central zone is
the site of baryte concentrations, and in some cases anhydrite and goethite. Cop-
per, lead and zinc are represented only by sulphides. (4) Lateral zoning is the
facies type of zoning. This conclusion is emphasized by the absence of the mutual
crossing of outer ore zones by the inner one and also the same disposition of ore
zones from centre to periphery in the different ore bodies (Skripchenko et al.
1971, Skripchenko 1980).
Though the general nature of Zhairem type deposits is of hydrothermal-sedi-
mentary type, the ore minerals (textural) relations are controlled by diagenetic
transformations. These transformations include crystallization and local regen-
eration of the primary ore silt deposits. Framboidal pyrite applies to the earliest
form. Other ore minerals (sphalerite, galena, baryte) are the results of late
crystallization in the order: (1) baryte; (2) sphalerite; (3) galena. Because the sedi-
mentation of iron took place in oxide form and non-ferrous metal as sulphides,
the diagenetic pyrite and Cu - Pb - Zn-sulphide aggregates are observed as
different layers.
Primary ore sediments formed under conditions of sulphide-ion deficit,
hence, only the sulphides of non-ferrous metals were deposited. The prevailing
form of iron deposition was as an oxide. Pyritization of sediments occurred in
thin layers during the diagenetic stage. Goethite was converted to Fe-disulphide,
but thick iron-oxide beds were not pyritized.
Normal interbedding of carbonic-clay-silica-carbonate sediments, sulphide
and oxide ores limits the number of genetic models for Zhairem deposits to that
of precipitation from a hydrothermal brine in basin deeps on the seafloor. The
main pyrite form in the ores is framboidal pyrite. As investigations showed
(Berberian 1981, Skripchenko 1980), this early, diagenetic form is related to
biogenic sulphate reduction.
The textural-mineralogical and genetic features outlined above have been
interpreted in a genetic model suggested by Skripchenko (1983) for ore sediments
of the Atlantis II Deap in the Red Sea. In accordance with this model (Fig. 5), the
generation of non-ferrous metal sulphides occurs in the contact zone between the
hot hydrothermal brine and normal seawater saturated by oxygen. The contact
zone is known as the zone of biogenic sulphate reduction (Bischoff 1974) and
functions as a filter preventing the dispersal of non-ferrous metals because of
their great affinity for sulphide ion. The iron precipitate is oxygen; that of
baryte, sulphate. Other elements, especially manganese, disperse into the sea.
The lateral distribution of non-ferrous metals is determined by differences in
their affinity for S2- and, according to this criterium, should be deposited in the
order: copper, lead and zinc outward from the centre of the periphery. The zone
of zinc accumulation is shifted towards the periphery relative to the lead and
baryte zones of the central part of the Zhairem type ore deposits (Fig. 2).
Two factors, (1) rhythmically-interrupted accumulation of ores and carbonic
flyschoid and (2) relatively high concentration of organic mass, are favourable
 356 N. S. Skripchenko et al.

I holfMjonlalls Itllt'
I
I
l-s6.s·c

D lD 26 3[J]4m5
I
sa tim IS'
J(9'1_
I
I

Fig. S. The sedimentation model for mineralization in the Atlantis II deep (Skripchenko 1983).
1, 2 Brine zones with maximum metal content: 1 iron (over 87 ppm); 2 manganese (over 77 ppm). 3
Upper limit for zinc over 7 ppm. 4 Zone of biogenic sulphate reduction. 5 Brine in flux in axial part of
depression. 1- 4 (in circles) distribution of ore types on deep bottom: 1 iron·copper-zinc; 2 rich cop-
per-zinc; 3 poor copper-zinc; 4iron-manganese. In parantheses: the minor mineral occurrences (py-
rite and anhydrite)

conditions for biogenic iron disulphide formation. Deposition of pyrite in outer

ore zones is explained by the possibility of biogenic sulphate reduction in sediments
only in the outer belt of the ore sedimentation deep. The high-temperature and
chemically corrosive brine in hydrothermal deeps does not promote the possibility
of intensive sulphate reduction in the sediment in the inner ore zones. The litho-
genetic component of ores of the inner zone in Zhairem deposits is generally silica.

Conclusion

Two genetic classes of massive sulphide deposits, vertical and horizontal zoning
types accordingly, are characterized by identical mineral and elemental composi-
tions. The hydrothermal-sedimentary phase of their formation is described by
two models: acidic oxygen and carbonic reduction. The vertically zoned group is
related to middle and final stages of basaltic volcanic activity; they are also
developed in thick carbonic argillaceous formations of shale according to Tval-
chrelidze (1982). High (ore material): (non-metallic mineral) ratios in the ore
bodies and their association with rocks formed under rapid sedimentation rates
suggests that this deposit type is the result of deposition from highly concentrated
ore-bearing fluids (Skripchenko 1966, Henley and Thornley 1979).
The laterally-zoned group of deposits is formed under conditions of low sedi-
mentation and ore deposition rates. As a result of these conditions rhythmic
layers of ore and lithogenic deposits accumulate. In addition, the non-ferrous
metal content of these deposits is large.
The sulphur sources for the two types of massive sulphide deposits are dif-
ferent. In the first group of deposits, sulphur is introduced by the ore solutions
themselves; in the second group, it is taken from seawater.
Several possible sources of metals necessary for ore formation and different
models of their delivery to the site of ore deposition are indicated in Fig. 4. In
Zoning of Massive Sulphide Deposits and Their Origin 357

some cases, activation of ore source occurs during the process of active volca-
nism; in other cases, it is connected with the development of tectonic fractures in
the Earth's crust. Consequently, one may assume a magmatic sulphur source in
massive sulphide deposits of the first group. Movement of concentrated ore-
bearing fluids is also intensified by active volcanism. Where the source rock is cut
by fractures, ore-bearing fluids pass through the rock stratum less intensively
and, as a result, ore accumulation will be defined by the dynamics of sedimentary
processes.

References

Barsukov VL, Grigoryan SV, Ovchinikov LN (1981) Geochemical methods of ore deposits search.
Nedra, Moscow, 318 pp (in Russian)
Berberian TK (1981) Icosahedron - a geometrical model of ordered pyrite framboids. Pap Acad Sci
USSR 258, 2:456 - 458 (in Russian)
Bischoff JL (1974) The sediments of Red sea thermal brine (mineralogy, chemical composition and
genesis) - In: Modern hydrothermal ore deposition. Mir, Moscow, pp 157 -193 (in Russian)
Bogysh lA, Trufanov VN (1983) Temperature regime and zoning of ore formation of polygenetic
massive sulphide lodes (Northern Caucasus). Geol Ore Dep 4:71 -79 (in Russian)
Borodaevskaya MB, Gorzhevsky DI, Krivtsov AI, Ruchkin GV, Skripchenko NS, Tva1chrelidze GA,
Yakovlev GF (1979) Massive sulphide deposits of World. Nedra, Moscow, p 284 (in Russian)
Henley RW, Thornley P (1979) Some geothermal aspects of polymetallic massive sulphide formation.
Econ Geol 74:1600-1612
Holland HD, Malinin SD (1982) Solubility and occurrences of non-metallic minerals. In: Geo-
chemistry of hydrothermal ore deposits. Mir, Moscow, pp 370 - 404 (in Russian)
Large RR (1977) Chemical evolution and zoning of massive sulphide deposit in volcanic terrains.
Econ Geol 72:549 - 572
Large RR, Both RA (1980) The volcanogenic sulphide ores at Mount Chalmers, eastern Queensland.
Econ Geol 75:992-1009
Pospelov GL (1973) Paradoxes geologic physical essence and metasomatoz mechanism. Nauka,
Novosibirsk, 355 pp (in Russian)
Skripchenko NS (1966) Volcanogenic-sedimentary ore deposition. Nedra, Moscow, 292 pp (in
Russian)
Skripchenko NS (1972) Hydrothermal sedimentary sulphide ores of basaltoid formations. Nedra,
Moscow, 241 pp (in Russian)
Skripchenko NS (1980) Hydrothermal-sedimentary polymetallic ores of limeshale formations. Nedra,
Moscow, 215 pp (in Russian)
Skripchenko NS (1983) Deposition and differentiation of ore silts at the Atlantis II deep of Red sea.
Geol Ore Dep 1:74 - 86 (in Russian)
Skripchenko NS, Rozhnov AA, Lytkin VA (1971) Zoning of ore deposits of Zhairem group poly-
metallic deposits (Central Kazakhstan). Geol Ore Dep 5:3 -11 (in Russian)
Smirnov VI (1960) Convergence of the massive sulphide deposits. Vestnic Moscow State Univ Series 4
GeoI2:19-26 (in Russian)
Smirnov VI (1968) Massive sulphide deposits. In: Genesis of endogenic ore deposits. Nedra, Moscow,
pp 586 - 647 (in Russian)
Smirnov VI (1982) Geology of mineral deposits. Nedra, Moscow, 688 pp (in Russian)
Smirnov VI, Borodayev VS, Starostin VI (1968) Massive sulphide ores and deposits of Japan. Geol
Ore Dep 1:17-31 (in Russian)
Stolyarov YuM (1980) Hypogenic sulphide-sulphate zoning of porphyry-copper deposits. GeolOre
Dep 3:47 - 57 (in Russian)
Tvalchrelidze GA (1982) Metallogeny of two main types of the Earth's crust development. GeolOre
Dep 2:3 -13 (in Russian)
Tvalchrelidze AG, Yaroshevich VZ (1981) Genesis of stratiform massive sulphide deposits. Nedra,
Moscow, 131 pp (in Russian)
Physico-Chemical Conditions of Base Metal Sulphide Ore
Formation
A. G. TvALCHRELIDZE 1

Abstract

Primary base metal massive sulphide deposits are formed as a rule in seafloor
depressions. Ore formation is preceded by stratification of seawaters into oxiding
and reducing zones. Hydrothermal sedimentary pyritic bodies are formed in two
stages. During sedimentogenesis, monosulphide iron and non-ferrous thin
sediments are accumulated under thermodynamic disequilibrium conditions.
During diagenesis, sea-floor sulphide sediments are crystallized as a result of
their interaction with pore solutions. The latter, according the thermodynamic
data, are characterized by the buffer ratio of sulphur in oxidized and reduced
forms. The process of diagenesis results in standard base metal zonation. If
pyrite-bearing fluids are characterized by increased total activity of sulphur, then
such a zonation is complicated by an additional zonal distribution of sulphides
and sulphates. The disappearance of anhydrite at the barite-polymetallic level is
due to the spontaneous evolution of the external physical and chemical
parameters against a background of temperature decrease and displacement of
the invariant barite-calcite-anhydrite equilibrium towards to barite crystalliza-
tion field.

Introduction

During the past 25 yrs. a large volume of scientific literature has confirmed the
hydrothermal-sedimentary origin of base metal massive sulphide deposits as pro-
posed by Oftedahl (1958), Smirnov (1960) and Anderson (1969). The geological
conditions and environment necessary for the hydrothermal-sedimentary forma-
tion of pyritic massive sulphide deposits were discussed by Smirnov (1968), and
various aspects of this genetic theory have also been discussed in numerous
publications by Watanabe (1965), Borodajevskaya et al. (1976), Dzotsenidze
(1969), Gilmour (1971), Hutchinson (1973, 1980) and others.
The basic geochemical differences between hydrothermal-sedimentary and
metasomatic sulphide ore formation were first discussed by Skripchenko (1969),

1 Caucasian Institute of Mineral Resources, 85 Paliashvili St., 380062 Tbilisi, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 359

who has shown that synsedimentary sulphide ore deposition has a long and
complex history, involving sedimentation and diagenetic processes. These geo-
chemical differences are not adequately taken into account by some workers. For
instance, Plimer and Finlow-Bates (1978) suggested that primary iron sulphide
minerals (pyrite and pyrrhotite) are directly deposited from seawater pyrite in an
oxidizing environment, pyrrhotite in a reducing environment. Current research
into genetic models for stratiform massive sulphide deposits tends to adopt one
of two approaches:
1. Investigations of the physical aspects of hydrothermal discharge, including
the hydrodynamics of buoyant plumes (Turner and Gustafson 1978, Solomon
and Walshe 1979a, b). Generally, these investigations consider the miner-
alogical zoning typical of these deposits to be the result of the dynamics of
progressive mixing of hydrothermal solutions with seawater.
2. Investigations of the chemical consequences of such a mixing. Generally,
these investigations consider the mineralogical zoning to be due to chemical
gradients established by the mixing of hydrothermal solutions with seawater
(Ridge 1974, Large 1977) or the interaction between ore solutions of possible
magmatogenic affiliation with convecting, entrained shallow pore fluids of
marine origin (Spooner 1980).
In this paper we try to formulate a third geochemical approach.

Hydrodynamics of Interaction Between Bottom Brine and

Hydrothermal Solution

Numerous paleotectonic reconstructions of pyrite deposits in volcanic and

sedimentary rocks (Yakovlev 1984) have shown that primary synsedimentary
pyrite formation proceeds in bottom depressions. Even in paleovolcanic areas
where base metal sulphides are related to rhyolite domes (for instance, in the
Urals-type provinces, Tva1chrelidze 1984), massive ore bodies lie directly in
volcanic craters and other small negative structures in the central parts of the
paleotopographic highs. These structures have sharp contacts with country rocks
(Sato 1977, Tvalchrelidze et al. 1982) and are characterized by specific conditions
of chemical-sedimentary lithogenesis (Skripchenko 1972). Recent investigations
by Sato (1972), Rickard (1973) and Tvalchrelidze (1981) have shown that
syngenetic sulphide ore formation occurs under reducing conditions.
Commonly, the level of the oxidizing-reducing margin of seawaters is fixed
by the appearance of banded iron and manganese formations. Such a margin
may be situated in the upper part of negative structures. If so iron and/or
manganese oxide minerals are deposited directly at the roof of sulphide ore
bodies. In other cases the oxide sediments are distal to, but always younger than,
sulphide ores.
As shown by Sato (1972), Turner and Gustafson (1978) and Solomon and
Walshe (1979a, b), the hydrothermal solutions supplied into the described
negative structures do form anaerobic brine pools due to the different densities
 360 A. G. Tvalchrelidze

SpeCi(te gravity Fig. 1. Density vs temperature re-

08 09 10 f. f lationship of seawater and model
, . . - _........._---'-_---"'--_..1...-_--'-_--'-_ _.1....-----' solutions of different salinity
Seowoter (after Sato 1972, Solomon and
Walshe 1979a)
50

100

150

"
0
h,-
200
1m NoCe
250

0300
density vs temperoture
mixing curve

Fig. 2. Reconstruction of the hydrothermal flow as a

function of the buoyancy main factor (after Solomon
and Walshe 1979a, simplified)

of seawater and hydrothermal fluids. We have recently suggested (Tvalchrelidze

et al. 1982) that bottom brines have the same chemical composition as a
hydrothermal fluid except that their lower temperatures (about 55° - 60°C)
result in oversaturation with respect to ore minerals. Furthermore, any
subsequent hydrothermal discharge into a brine pool that has achieved a state of
stable stratification would result in the development of buoyant plumes due to
the difference in densities between the high-temperature hydrothermal solutions
and the cooled brines of the pool (see Fig. 1).
Investigators of hydrodynamic phenomena (Solomon and Walshe 1979a, b)
postulate a progressive decrease in temperature vertically upwards along the axis
of a buoyant plume formed in this manner. Such a model is illustrated in Fig. 2.
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 361

The marginal stream lines are expressed by a buoyancy main factor ("buoyancy
parameter") LI:

(1)

where: go = density of plume fluid, gl = density of seawater, g = gravitational

constant.
The buoyant plume is predicted to rise a height of 500 m. Such hydrodynamic
hypotheses deserve careful consideration because they are suggested to be a
fundamental cause of the characteristic zoning of pyritic deposits (Solomon et al.
1980). However, any error in the vertical temperature gradient may cause signif-
icant errors in the genetic model. We have tried to solve this problem from heat
balance considerations.
The heat amount (Q) which is supplied in an ore-forming medium from a
hydrothermal flow in hydrodynamically stable conditions can be determined by
the classical equation:

(2)

where: Cp = thermal heat capacity; LI T= temperature difference between a

stream and its medium, which averages tOO°C, according to our (Tvalchrelidze
1981) investigations; and m = vt, where v is mass transfer during the time of
stream stabilization (t).
On the other hand, as the stream has the form of an inverted cone:

m = gO+gl Sh, (3)

6

where: h = the height of the cone axis, S = the area of the cone base. Then:

6Q
h= (4)

where a is the inclination angle of the cone edge.

Using in Eqs. (2), (3) and (4) maximal values v = 5.10 5 g s-t, t = 100 s,
a = tOo (Turner and Gustafson 1978, Solomon and Walshe 1979a), go = 1.1099 g
cm -3, gl = 1.0495 g cm -3 (Tvalchrelidze et al. 1982) and tabular data of
Cp = 0.950 kcal . 1- 1 it could be shown that the maximum height to which a
buoyant hydrothermal stream would rise would not exceed 50 m. Because the
real values of the given parameters should be much lower than those used in cal-
culation, it could be concluded that commonly the real height would average less
than 10 m. So the real buoyant hydrothermal streams are thought to be unable to
provide mineralogical zoning of pyrite deposits.
362 A. G. Tvalchrelidze

Ore Deposition

Based on experimental (Tvalchrelidze and Yaroshevich 1981) and empirical data

(Shanks and Bischoff 1977, Mossman and Heffernan 1978), it was suggested
(Tvalchrelidze 1984) that polymetallic sediments before their crystallization con-
sist of mono sulphide species of iron and non-ferrous metals. So, ore bodies of
the pyritic type must be formed as a result of diagenesis. It is well known that de-
crease in temperature is the main cause of ore precipitation. Because the tempera-
ture of the brine is much lower than that of the hydrothermal fluids, this brine
would be entirely oversaturated with respect to ore minerals (see Fig. 3), and
consequently suspensions of sulphide species would be formed over a range of
heights above the floor of the brine pool. The kinetics of sedimentation by such a
mechanism are governed by Stokes' law, which in the present case can be ex-
pressed as

(5)

where: V = sedimentation rate, r = radius of sulphide species in the suspension,

{h = density of dispersive phase, 17 = brine viscosity.
Using the recently published values (Tvalchrelidze et al. 1982) of these param-
eters, it can be shown that a 4-m-thick ore body could have formed within
tOOO yrs. In other words the rates of hydrothermal-sedimentary sulphide accu-
mulation are many times greater than the rates of normal marine clastic sedimen-
tation, and result in sulphide deposits that contain little detrital matter. As a
result of continued sulphide accumulation in a brine pool, the sediment-brine
interface would tend to come closer to the oxidizing-reducing boundary of the
brine pool surface. So new portions of hydrothermal fluids would discharge into
the oxidizing zone, giving rise to the formation of oxide manganese and iron
sediments.
Another feature of sulphide ore formation is as follows. Accumulating
sulphide sediments contain pore solutions, which of course must be over-

Fig. 3. Temperature dependence of saturation of a

hydrothermal fluid (after Tvalchrelidze and Yaro-
shevich 1981). lAP Total ion activity of a given solu-
tion (log Us = - 3.9; for other thermodynamic data
see Table 1); K equilibrium ion activity. 1 Pyrite; 2
chalcopyrite; 3 sphalerite; 4 galena; 5 equilibrium sat-
uration line
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 363

saturated with respect to ore components. However, deposition of ore species

takes place only over the sediment-brine interface. The fundamental cause of this
event, as has been suggested recently (Tvalchrelidze and Yaroshevich 1981,
Tvalchrelidze et al. 1982), is the protionization of water molecules in the micro-
pore environment and their extremely high compression. Such a water with
distorted properties (water dissipative state, according to Blokh 1984) forms
super-high concentrated solutions which can exist only in a micropore environ-
ment because their saturation does not correspond to the external thermo-
dynamic conditions.
These phenomena have three important consequences:
1. Since the super-high concentrated solutions, as mentioned, can exist only in
pores among sulphide species, sulphide sediments and their pore waters are
relatively independent and may be theoretically considered as a half-isolated
system.
2. The metastable state of pore solutions preserves sulphide sediments from
recrystallization. So, for some time hydrothermal activity would cause con-
vective flow of pore solutions. This migration of pore fluids redistributes ore
components, which would tend to homogenize the chemical composition of
the sulphide deposit. This is why all the conceptions that consider the geo-
chemical zonation of stratiform sulphide deposits to be due to the chemical
evolution of the ore-forming solution and to be a primary depositional pattern
(Large 1977) are open to criticism.
3. Crystallization of ores and consequent hypogene zonation are diagenetic pro-
cesses, which are controlled by interaction between sulphide sediments and
pore solutions. It must be noted that the diagenetic nature of hypogene zona-
tion of base metal sulphide ores was suggested by Govett and Whitehead
(1974), but geochemical aspects of this process have not received proper atten-
tion. It is this aspect of the problem that is discussed below.

Crystallization of Sulphide Sediments and Formation

of Monoascending Zonation

We have recently shown (Buadze and Tvalchrelidze 1980) that during diagenesis
the crystallization of sulphide minerals in sediments takes place under specific
conditions. Briefly, since the sulphur of monosulphides such as FeS or H§ have
a 2 - oxidation state (Rickard 1975), the diagenetic crystallization of disulphide
minerals, in which sulphur has a 1 - oxidation state, requires partial oxidation of
the bulk sulphur content. It has been suggested that, in the diagenesis of sedi-
ments, the oxidizing agent is sulphur (Berner 1972) or polysulphide (Rickard
1975) which has been formed as the result of bacterial reduction of sulphate. It is
clear that such conditions are not applicable to hydrothermal-sedimentary ores,
because, as indicated by detailed isotopic studies (Tvalchrelidze 1984), there is
only minimal involvement of biogenic sulphur. Consequently, most of the pyrite
and other disulphide minerals must have been formed without any significant in-
fluence of bacterial metabolism.
364 A. G. Tvalchrelidze

Wikjord et al. (1976) suggested oxidizing-reducing reactions resulting in the

liberation of hydrogen and consequent pyrite crystallization. Fabriol and Maurel
(1975) believed that under certain conditions pyrite could form as a result of the
reaction of FeS with sulphate ion:

(6)

Simple thermodynamic conclusions, however, indicate that the concentration of

ferrous ion produced by the reaction would not allow the reaction to proceed to
the right.
Berner (1970) considered the reaction with sulphate ion and hydrogen
sulphide:

FeS+3H§+2H+ +SO~- -+FeS2+3So+4Hp . (7)

Under the conditions of elevated temperature of diagenesis (Tvalchrelidze 1981),

however, the elementary sulphur is unstable:

(8)

Summing up reactions (7) and (8):

(9)

Thus, pyrite formation from iron monosulphide needs a buffer ratio of

sulphur in oxidized and reduced forms. Figure 4 indicates that the pyrite crystal-
lization field lies on the univariant equilibrium line between hydrogen sulphide
and sulphide ion. If such a diagram were supplemented with mineral equilibria of
non-ferrous metals with the solution (see Fig. 5), then the bulk thermodynamic
picture of base metal ore formation would be manifested. Activities of the main
components of the model solution are given in Table 1. Table 2 contains some
additional thermodynamic data. The method of calculation was published
recently in detail (Tvalchrelidze et al. 1982, Tvalchrelidze and Yaroshevich 1981).
- ~O ~~TT------------------'
h'sq . \
.--...t...
nm

Fig. 4. Pyrite crystallization field and relation of

sulphur dissolved species under T= 150°C;
2 log as = - 5.5. Equilibria of minerals with
model solution are calculated using Table 1
(after Tvalchrelidze 1981). hm = Hematite;
mt = magnetite; po = pyrrhotite; py = pyrite.
-60 +--~~"'T"'~~~~::;:::::;:::::::;::::::::t Univariant equilibrium lines: 1 between miner-
2 ¥ t5 8 10 72 als; 2 between minerals and a solution; 3 be-
pH tween sulphur dissolved species
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 365

-"o.---------,
I l!

- - gr

10 6 8 to
pH
Fig. 5. Fragment of total thermodynamic diagram illustrating the consequence of crystallization of
sulphide minerals from a model solution under T= 150°C; logaFe2+ = -3.7; logllcu2+ = -17.0;
log azn2+ = - 9.4; log apb2+ = -11.0; I log as = - 3.9; II log as = - 5.5 (after Tvalchrelidze et al.
1982). cz = Chalcocite; cp = chalcopyrite; gl = galena; sph = sphalerite. 1 Pyrite crystallization field;
2 univariant equilibria between minerals and the model solution; 3 univariant equilibria between min-
erals. Thermodynamic properties of the model solution are given in Table 1. Examples of some addi-
tional simple reactions are given in Table 2. For the full thermodynamic picture, see Tvalchrelidze et
al. (1982)

Table 1. Activities of some components of the model solution (after Tvalchrelidze and Yaroshevich
1981)

Component Temperature, °c

25 50 100 150 200 250

- loga;

H2C03 2.4 2.6 2.7

HCOi 3.3 5,0
CaHCOt 3.5 3.9 5.7
MgHCOt 4.4 3.1 3.1
NaS04" 2.1 2.2 2.6
S~- 5.1 3.9
FeOH+ 3.8 4.5 5.3
HSO- 6.3 5.6 5.3
Znd- 4.8 5.3 5.8
ZnCli 5.2 5.7 6.2
CuCl2 6.2 6.7 7.2
CuCI 7.0 7.3 7.8
PbC~- 6.5 6.7 7.2
PbCI3- 6.7 7.2 7.7
p~+ 11.0 11.4 11.0 12.5 13.2
Zn2+ 6.3 7.8 9.4 11.6 13.9
Fe2+ 3.3 3.5 3.7 3.9 4.1
Cu2+ 16.4 17.5 17.0 18.5 20.7
H+ 5.3 4.8 4.3 3.9 4.0
H2S a 4.3/5.9 3.9/5.5 3.9/5.5 3.8/5.4 4.4/4.6
HS- a 5.8/7.4 6.017.6 7.1/8.7 8.3/9.9 10.0/11.0
S2- a 13.6/15.2 13.3/14.9 14.2115.8 15.8/16.9 16.6/18.2

a Note: in nominator - the component activity below - log ams = 3.9; in denominator - the
component activity below - log aUfs = 5.5.
366 A. G. Tvalchrelidze

Table 2. Equilibrium constants of some reactions (after Tvalchrelidze and Yaroshevich 1981)

Reactions Temperature, DC

50 100 150 200 250 300 350

-logK

Pbs-+pif+ +g2- 26.77 24.05 22.00 20.43 19.21 13.35 17.58

ZnS-+Zn2+ +S2- 24.60 22.68 21.17 19.97 19.02 18.30 17.79
FeS-+Fe2+ +S2- 17.19 15.21 14.88
CuFe~-+Cu2+ +Fe2+ +2S2- 62.74 57.30 53.01 49.71 47.00 44.95 43.34
Fe~+H20-+Fe2+ +2S2- +2H+ +t~ 80.81 71.97 64.64 61.19 58.07 56.12
~S-+2H++g2- 1~~ 19.05 18.90 19.34 20.24 21.68 23.41
H2S-+H+ +HS- 6.81 6.95 7.51 8.36 9.59

The diagrams need some clarification. We have mentioned that the pore
water dissipative state results in the crystallization of pyrite due to the interaction
between pore fluids and monosulphides. This process does not change the buffer
ratio of the sulphur dissolved forms, but would decrease the total activity (and
concentration) of sulphur. As simple calculations have shown, after the crystal-
lization of pyrite and other disulphide minerals log as would decrease from - 3.9
to - 5.5. As long as the new hydrothermal flow preserved the initial concentra-
tion (activity) of sulphur in the lower part of the sulphide stratum, a gradual
decrease of sulphur activity would be observed in the ascending section of the
crystallizing sediments. So the first diagram (see Fig. 5, I) represents the thermo-
dynamics of the bottom part of a sulphide body, while the second one (see
Fig. 5, II) refers to its top.
It is well known that non-ferrous sulphides commonly occur only in the upper
part of base metal sulphide ore bodies, where they are crystallized simultaneously
with pyrite. As the pyrite crystallization field here is narrow (see Fig. 5, II) and
log p02 could not range widely, the zonal deposition of chalcopyrite, sphalerite
and galena would be due to a neutralization of pore solutions.
The reason for such a neutralization could be explained (Tvalchrelidze 1981)
by the results of experiment (see Table 3). Analysis of Table 3 indicates:
1. Pyrite does not crystallize from FeS without oxidation of H~.
2. The neutralization of solutions proceeds only after pyrite crystallization
(compare pH values in Fig. 4 and Table 3 !) and only in solutions containing
(NH4)~' So, the neutralization of solutions must be due to partial hydrolysis
resulting in NH 40H formation.
Theoretically, as Na, the main component of any hydrothermal solution
(Skinner and Barton 1973), is a stronger base than ammonium, temperature de-
crease must immediately cause neutralization of fluids. Practically, neutraliza-
tion is prevented both by the buffer ratio of the dissolved forms of sulphur and
by a stable distribution of metals in chloride complexes (see Table 1). The latest
experimental and thermodynamic investigations of hydrolysis under hydrother-
mal conditions (Govorov 1977, Massalovich 1975, etc.) have shown that the hy-
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 367

Table 3. Experimental study of hydrothermal crystallization of pyrite under T = 150°C, P = 500 bar,
experiment duration = 35 days

Experi- Autoclave load pH Stable phase Metastable phases

ment
number Primary Final

T-41 14.68 rnl (NH4hS + 4 ml 7 10 Monoclinic Mackinawite

0.9 m FeS04 pyrrhotite
T-42 200 mg FeS+ 15 mg S 5.5 1.5 Pyrite Monoclinic pyrrhotite
+ 16.68 ml H2S + sulphur + marcasite
T-43 250mg Fe~~ 5.5 4 Monoclinic
+ 18.68 ml H2S pyrrhotite
T-44 200mg FeS 5.5 2 Monoclinic X-ray amorphus FeS
+19.62ml H2S pyrrhotite
T-45 200 mg FeS + 15 mg S 7 Pyrite Sulphur + marcasite
+18.68 ml H2O
T-46 14.68 ml (NH4)2S + 4 ml 7 12 Pyrite Marcasite + monoclinic
0.9 m FeS04 + 15 mg S pyrrhotite

Note: Solutions, the concentrations of which are not shown, are saturated under T = 25°C and
P= 1 bar.

Fig. 6. Phase relations in the system

Ba - Ca - S - 0 (after Gomelauri 1978,
slightly modified)

drolysis would be active only after a decrease of sulphur activity and a displace-
ment of equilibria between the chloride complexes of iron and non-ferrous
metals.
The above considerations indicate that zonal deposition of polymetallic
sulphides, being governed by neutralization of pore solutions, would be active
only in the upper parts of a crystallizing ore body.
Thus, although the primary sedimentation of sulphide deposits takes place
under conditions of general chemical disequilibrium, the final form of the
deposit, particularly with regard to bulk mineralogy and zonal patterns, is deter-
368 A. G. Tvalchrelidze

mined by diagenetic processes and not by the evolution of ore-forming solutions.

This is the main difference between our genetic model and that of Large (1977).
These diagenetic processes take place under chemical conditions that are buf-
fered within narrow ranges of various chemical parameters. Buffering within
narrow ranges explains the chemical and mineralogical uniformity of massive
base metal sulphide deposits throughout the world. A single exception to the
general rule is presented by Kuroko-type deposits, where the usual mineralogical
zonation is complicated by the presence of sulphate zones. In general terms, such
a zonation is expressed by the disappearance of anhydrite at the barite-poly-
metallic sulphide zone boundary. Briefly, this phenomenon may be explained in
two ways: (1) by higher total sulphur activity compared to the model described
above; and (2) by the tendency of the anhydrite phase to dissolve from a barite-
calcite-anhydrite equilibrium assemblage as temperature decreases below 250°C
(see Fig. 6).

Acknowledgment. I would like to express my deepest gratitute to my anonymous opponent who has
read and edited the manuscript of this paper and supplied a number of important and valuable
remarks.

References

Anderson CA (1969) Massive sulfide deposits and volcanism. Econ Geol64:129-146

Berner RA (1970) Sedimentary pyrite formation. Am J Sci 268:1-23
Berner RA (1972) Sulfate reduction, pyrite formation and the oceanic sulphur budjet. In: Dyssen D,
Jagder J, (eds) The changing chemistry of the oceans. Almqvist & Wiksell, Stockholm, pp
347 -361
Blokh AM (1984) Mobilization of matter in superhigh-concentrated solution as a result of the dissipa-
tive state of water in microspaces of the Earth's interior. In: Janelidze TV, Tvalchrelidze AG (eds)
Proc 6th IAGOD Symp, vol I. Schweizerbart, Stuttgart, pp 59 - 67
Borodajevskaya MB, Krivtsov AI, Shiray EP (1976) Peculiarities of early geosyncline volcanism and
ore formation in pyrite-bearing different provinces. Geol Ore Dep 1:20 - 28 (in Russian)
Buadze VI, Tvalchrelidze AG (1980) Geological, geochemical and physico-chemical conditions of
formation of stratiform non-ferrous deposits in sedimentary rocks of the Caucasus. In: Ridge JD
(ed) Proc Fifth Quadrennial IAGOD Symposium, volt. Schweizerbart, Stuttgart, pp 245 - 260
Dzotsenidze GS (1969) Role of volcanism in the formation of sedimentary rocks and ores. Nedra,
Moscow (in Russian)
Fabriol R, Maurel C (1975) Reactions chimiques entre pyrrhotite et solutions d'acide sulfurique:
etude experimentale et thermodynamique. Bull Soc Miner Crystallogr 98:227 - 234
Gilmour P (1971) Strata-bound massive pyritic sulfide deposits - a review. Econ Geo166:1239 -1244
Gomelauri AI (1978) Temperature conditions of barite formation in the Georgian deposits and their
examination in autoclaves. In: Ermakov NP (ed) Thermobarogeochemistry of the Earth's crust
and ore formation. Nauka, Moscow, pp 127 -130 (in Russian)
Govett GJS, Whitehead RES (1974) Origin of mineral zoning in stratiform sulfides; a hypothesis.
Econ Geol 69:551 - 556
Govorov IN (1977) Thermodynamics of ion-mineral equilibria and the mineralogy of hydrothermal
deposits. Nauka, Moscow (in Russian)
Hutchinson RW (1973) Volcanogenic sulfide deposits and their metallogenic significance. Econ Geol
68:1223 -1247
Hutchinson RW (1980) Massive base metal sulphide deposits as guides of tectonic evolution. In:
Strangway DW (ed) The continental crust and its deposits. Geol Assoc Can Spec Pap 20:659 - 684
Physico-Chemical Conditions of Base Metal Sulphide Ore Formation 369

Large RR (1977) Chemical evolution and zonation of massive sulfide deposits in volcanic terrains.
Econ Geol 72:549 - 572
Massalovich AM (1975) Ore transport during pyrite deposition in hydrothermal systems. GeolOre
Dep 2:59 - 63 (in Russian)
Mossman DJ, Heffernan KJ (1978) Simulated low-grade metamorphism in metalliferous mud from
the Read Sea Atlantis II deep. Econ Geol 73: 1150 -1154
Oftedahl C (1958) A theory of exhalative-sedimentary ores. Geol FOren Stockholm FOrh 80:1-19
Plimer IR, Finlow-Bates T (1978) Relationship between primary iron sulphide species, sulphur
source, depth of formation and age of submarine exhalative-sulphide deposits. Miner Deposita
13:399-410
Rickard DT (1973) Limiting conditions for synsedimentary sulfide ore formation. Econ Geol
68:605-617
Rickard DT (1975) Kinetics and mechanism of pyrite formation at low temperatures. Am J Sci
275:636-652
Ridge JD (1974) Notes on boiling of ascending ore fluids and the position of volcanic-exhalative
deposits in the modified Lindgren classification. Geology 6:287 - 288
Sato T (1972) Behaviours of ore-forming solution in seawater. Min GeoI22:31-42
Sato T (1977) Kuroko deposits: their geology, geochemistry and origin. In: Volcanic processes in ore
genesis. Inst Min Metall, London, pp 153 -161
Shanks WC, Bischoff JL (1977) Ore transport and deposition in the Red Sea geothermal system:
a geochemical model. Geochim Cosmochim Acta 41 :1507 -1519
Skinner BJ, Barton PB (1973) Genesis of mineral deposits. Annu Rev Earth Planet Sci 1:183 -211
Skripchenko NS (1969) On the stability of sulphides and oxides during low temperature
mineragenesis. Geochemistry 3:301 - 308 (in Russian)
Skripchenko NS (1972) Hydrothermal-sedimentary sulphide ores in basaltic rocks. Nedra, Moscow
Smirnov VI (1960) Convergence of pyrite deposits. Moscow State Univ Bull GeoI2:19-26 (in Rus-
sian)
Smirnov VI (1968) Pyrite deposits. In: Smirnov VI (ed) Genesis of endogenous ore deposits. Nedra,
Moscow, pp 586-647
Solomon M, Walshe JL (1979a) The behavior of massive-sulphide ore solution entering seawater and
the development of zoned deposits. Bull Miner 102:463 -470
Solomon M, Walshe JL (1979b) The formation of massive sulfide deposits on the sea floor. Econ
Geol 74:797 - 813
Solomon M, Walshe JL, Garcia-Palomero P (1980) Formation of massive sulphide deposits at Rio
Tinto, Spain. Trans Inst Min Metall B89:16-24
Spooner ETC (1980) Cu-pyrite mineralization and seawater convection in oceanic crust - the
ophiolitic ore deposits of Cyprus. In: Stangway DW (ed) The continental crust and its mineral
deposits. Geol Assoc Can Spec Pap 20:685 -704
Turner JS, Gustafson LB (1978) The flow of hot saline solution from vents in the sea floor - some
implications for exhalative massive sulfide and other deposits. Econ Geol 73:1082 -1100
Tvalchrelidze AG (1981) Geochemistry of hydrothermal-sedimentary sulphide ore formation. Geol
Ore Dep 2:1-16 (in Russian)
Tvalchrelidze AG (1984) Ore-Bearing hydrothermal systems of stratiform pyrite deposits. In:
Janelidze TV, Tvalchrelidze AG (eds) Proc 6th IAGOD Symp, vol I. Schweizerbart, Stuttgart, pp
246-257
Tvalchrelidze AG, Yaroshevich VZ (1981) Genesis of stratiform base metal massive sulphide
deposits. Nedra, Moscow (in Russian)
Tvalchrelidze AG, Yaroshevich VZ, Narozauli IG (1982) A genetic model for the Urup copper
massive sulfide deposit. Int Geol Rev 24:11 - 24
Watanabe T (1965) Submarine volcanism and ore deposition. J Geol Soc Jpn 71 :332 - 336
Wikjord AG, Rummery TE, Doern FE (1976) Crystallization of pyrite from deoxygenated aqueous
sulphide solution at elevated temperatures and pressures. Can Miner 4:571 - 573
Yakovlev GF (1984) Volcanogene structures of mineral deposits. Nedra, Moscow (in Russian)
On the Genesis of Barite-Associated with Volcanogenic
Massive Sulfides, Fukazawa Mine, Hokuroku District,
Japan
S. I. KALOGEROPOULOS 1,2 and S. D. SCOTT!

Abstract

The barite bed at Fukazawa mine overlies a massive black ore zone and underlies
a ferruginous chert bed (tetsusekiei). It is characterized by simple mineralogy
which consists predominantly of barite with small but variable amounts of
pyrite, sphalerite, chalcopyrite, galena, hematite, quartz and traces of intimately
intergrown chlorite-sericite. Chemical and textural evidence indicates that barite
formed by the mixing of a circulating barium-rich hydrothermal solution with
cold seawater within an unconsolidated pyritic tuff. Growth of barite displaced
the very fine silicate matrix through a process of fluidization of the tuffaceous
bed during hydrothermal discharge. Variations in the intensity of the hydro-
thermal activity resulted in vertical fluctuations of the redox boundary providing
for replenishment of the barite environment with cold seawater which passed on
to the barite both its sulfur and strontium isotopic characteristics. The barite bed
was affected by later hydrothermal solutions, the activity of which continued
even after the deposition of hanging wall rocks. As a consequence, equilibrium
thermodynamics cannot be applied to describe the chemical behaviour of the
bulk dynamic system.

Introduction

Kuroko-type ore deposits occur in Miocene volcanic and sedimentary rocks of

the Green Tuff Region in Japan. The Hokuroku district located in northern
Honshu (Fig. 1) is the most productive Kuroko-type ore field in the Green Tuff
Region (Sato 1974, Ohmoto and Skinner 1983). Tuffs and tuff breccias in the
Green Tuff Region are also the hosts of relatively small lenticular massive barite
deposits of Kuroko type (Igarashi et al. 1974). Thin (generally less than 1 m),
essentially monomineralic, massive and sandy barite ores also form the upper-
most zone of massive sulfide Kuroko mineralization. The barite bed in the
Tsunokakezawa No.1 ore body of the Fukazawa mine, the subject of this study,

I Department of Geology, University of Toronto, Toronto, Ontario, M5S 1A1, Canada

2 Present address: Institute of Geology and Mineral Exploration, 70 Messoghion Street, Athens 608,
Greece

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 371

Hokko i do

'-...HOKUROKU
DISTRICT

LEGEND
I KUROKO
OREBOO'Y * TOWN

~ MID
MA.RGIN OF
MIOCENE
BASIN
j JAPAN
NATIONAL
RAILWAY H 0 n shu
o :l 10Km.

9L--_,O"-O_-'--_3....00Km .

Fig. 1. Location of Kuroko deposits in relation to the Green Tuff Region (from Hashimoto 1977).
The inset shows the distribution of the Kuroko deposits in the Hokuroku district. The studied barites
are from the Fukazawa mine (underlined)

conformably overlies a massive black ore zone and underlies a bedded cherty tuff
that is variably enriched in iron (tetsusekiei; Fig. 2). Chemical (Shikazono
personal communication 1979) and isotopic (Farrell 1979, Farrell and Holland
1983, Kusakabe and Chiba 1983) studies on barite from the Tsunokakezawa
No.1 ore body have contributed to an understanding of the genesis of the barite
zone. This paper describes a study utilizing mineralogy, fluid inclusions, stable
isotopes, and the relation to the overlying tetsusekiei zone to elucidate that
genesis further.
372 S.1. Kalogeropoulos and S. D. Scott

......' \
I', ('<5:\
I \ \ os'----...J
I
I
o
I
4 10 m
I

" ...... _,11'1.-...I

.>

L
3635 N
l __1 1-
2628E

S7
L-1S0

E - WALL S-FACE W-WALL

LEGEND

L
Iv v v v I BASALT (TUFF. BRECCIA)

t~$= 1 TETSUSEKIEI
3m
1'* =# 1 BARITE

BLACK ORE

I A :y: .6 I SILICEOUS ORE

o 3m

Fig. 2. A part of Tsunokakezawa No.1 ore body on level - 150 at stop 7 (-150'7). The massive
barite zone lies immediately above the massive black ore and below tetsusekiei (ferruginous cherty
tuff). Numbers represent samples referred to in the text. (Courtesy of Dowa Mining Co. Ltd. , Japan)

Mineralogy and Geochemistry

The barite zone at Fukazawa is characterized by a simple mineralogy consisting

predominantly of barite with small but variable amounts of pyrite, chalcopyrite,
sphalerite, galena, hematite, quartz, and traces of intimately intergrown chlorite
- sericite. The sphalerite contains minute inclusions of "chalcopyrite disease"
which is thought by Barton (1978) to be produced by euS-rich solutions reacting
with FeS in sphalerite.

Pyrite
Two textural types of pyrite have been observed in the barite ore zone: (1)
euhedral disseminated clusters; and (2) euhedral to anhedral pseudomorphs after
barite.
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 373

Fig. 3A - E. Textures exhibited by pyrite in the barite zone of Tsunokakezawa No.1 ore body. All
photomicrographs are in plane polarized light. A Euhedral pyrites (black) are exclusively associated
with the tuff component, and both are entrapped by barite. Transmitted light. Band C Intermediate
stages of tuff displacement and entrapment of pyrite by barite. The tuff is dark grey and occupies the
irregular interstices between large aggregates of barite. Transmitted light. D The clay matrix of the
tuffaceous component is completely displaced, and pyrite is concentrated and recrystallized within
barite. Note that regions originally occupied by tuff can still be traced by the contrast in relief of bar-
ite and by the distribution of pyrite. Transmitted light. E Euhedral pyrites within tuff both of which
are enclosed by barite. Reflected light. F Euhedral pyrite inclusion within barite. Note the similarity
in internal texture between this type of pyrite and that of E. Reflected light

Euhedral pyrite grains are usually disseminated in clusters whithin the barite
zone (Fig. 3 D). However, whenever tuffaceous (silicate) material is enclosed by
the massive barite, the pyrites are exclusively associated with the tuff (Fig. 3 A).
Figures 3 A - D depict pyrites whose host is progressively changing from a tuf-
faceous material to barite via displacement of the former. All these pyrites have
identical textures (Fig. 3 E, F). These features suggest that the clusters of euhedral
374 S.1. Kalogeropoulos and S. D. Scott

Fig. 4A - F. Textures exhibited by pyrite in the barite zone of Fukazawa (Tsunokakezawa No.1 ore
body) and Furutobe mines. A Pyrite replacing recrystallized barite along its cleavages. Reflected
light. 8 Fanlike barite blades, partially replaced by pyrite. Plane transmitted light. C Pyrite pseudo-
morphs after fanlike barite. Plane reflected light. D Hematite inclusions within recrystallized barite.
Hematite has cubic and triangular outlines. Plane transmitted light. E Hematite pseudomorphs after
pyrite in tetsusekiei overlying barite zone. Partially plane transmitted and reflected light. F Hematite
pseudomorph after pyrite and quartz-hematite both interstitial to barite. Plane transmitted light

pyrites were enclosed in the barite as the unconsolidated tuffaceous matrix was
progressively displaced.
Pyrite replaced barite along its cleavages (Fig.4A) and pseudomorphed
fanlike barite blades (Fig. 4B, C) . A probable reaction is

The released barium may have been adsorbed by clays and/or precipitated as
barite in the overlying zone.
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 375

Fig. 5. Alternating hema-

tite-rich bands in barite

Hematite

Scattered inclusions of hematite, some apparently pseudomorphous after pyrite,

occur within recrystallized barites (Fig. 4D). At Fukazawa, there is evidence that
hematite has pseudomorphically replaced pyrite in the overlying pyritic
tetsusekiei as well. Therefore hematite inclusions are probably incorporated into
barite by the same process as was described for the "displacement" type of
pyrite.
Hematite also occurs interstitially to barite crystals (Fig.4F) in certain
hematite-rich bands (Fig. 4E). There are two possible explanations for this phe-
nomenon:
1. Pyrite grains within the barite zone may have been oxidized without pseudo-
morphic textures being preserved.
2. The Fe 2 + IFe 3 + ratio in the ascending hydrothermal fluid may have been so
far removed from the stability field of any sulfide that at some level within the
barite zone the solutions yield interstitial hematite. In case (2) any pyrite
initially present within the appropriate parts of the barite zone would also be
expected to be oxidized. The presence of hematite pseudomorphs after pyrite
together with quartz-hematite intergrowths, both interstitial to barite
(Fig.4F) indicates that the second case is the more likely. Moreover, the
observation of alternating hematite-rich and sulfide-bearing barite layers
(Fig. 5) suggests that the redox boundary was fluctuating apparently due to
fluctuations in the intensity of the hydrothermal activity.

Barite

Fanlike blades and sheafs of barite, euhedral to subhedral tubular grains, and oc-
casionally, single barite grains combining these textural features, are observed in
the barite zone. Veins of barite crosscut the barite beds.
376 S. I. Kaiogeropouios and S. D. Scott

Experimental work has shown that (Ba, Sr)S04 is unreactive in terms of

Ba - Sr exchange between aqueous and solid phases. This means that during the
course of precipitation the solid does not readily re-equilibrate in response to
changes in the composition of the aqueous phase and that chemical equilibrium is
established only between the liquid and the last increment of barite precipitated
(Doerner and Hoskins 1925, Gordon et al. 1954, Hanor 1968). This type of
equilibrium has been referred to as "surficial equilibrium" by Barton et al.
(1963).
Recognition of texturally and chemically distinct stages of precipitation with
respect to Sr content can provide information concerning the mechanism(s) of
barite precipitation and the variation of the msiJ.+ / mBa2+ ratio in solution. This is
true only if the initial precipitation follows a prescribed path. For instance, if a
(Ba, Sr)-rich solution suddenly encounters an excess SO~- presumably barite will
be "dumped" and the msiJ.+ /mBa2+ ratio will match the initial ratio in the
solution. Actually, sulfur isotopes of Kuroko barites suggest that the major
precipitating mechanism of these barites is the mixing of a barium-bearing
hydrothermal fluid and cold seawater (Bryndzia et al. 1983, Kusakabe and Chiba
1983).
The internal texture of single barite grains, regardless of whether they are fine
( <0.5 mm) or coarse-grained (0.5 - 2.5 mm) is typically characterized by two
zones with different textural features (Fig. 6).
1. A write clean variety, referred to as zone 1; and
2. A greyish-white, dusty-looking variety of very fine-grained ( <0.2 mm) barite
aggregates, referred to as zone 2. Zone 2 has morphological characteristics
that could be associated with higher degrees of supersaturation than those of
zone 1. The most likely way to achieve this supersaturation would be to mix
(Ba, Sr)-rich hydrothermal fluid with cold SO~- -bearing seawater. This in-
ference can further be supported by the sulfur isotopes of the Kuroko barites
(Kusakabe and Chiba 1983). Single barite grains exclusively representing zone
1 or zone 2, as well as composites of 1 and 2 in zone patterns such as zone 1
(centre)/zone 2 (rim), zone 2 (centre)/zone 1 (rim), and zone 1 (centre)/zone 2
(intermediate)/zone 1 (rim) have been identified in the barite ore. Evidently,
the last zone pattern represents a complete evolution of barite growth.
The observation of (1) transitional textural stages from fan- and sheaflike
barites to tabular barites with identical internal textural features (Fig. 6 B - D),
and (2) the incorporation of a number of textural units which characterize single
barite grains (Le. zones 1 and 2) within individual but larger braik grains
(Fig. 6D, E), suggest that the composite barite grains were most likely produced
by recrystallization of the fan- or sheaflike aggregates. Green et al. (1983) have
shown that the hydrothermal activity in the Fukazawa ore continued even after
the deposition of the hanging wall rocks. Thus, the recrystallization textures ob-
served in the barite zone may have been caused or aided by this later hydro-
thermal solution.
Both textural types of barite (clean zone 1 and dusty-looking zone 2) were
analyzed for their strontium content by energy dispersive electron microprobe. In
addition, qualitative spectral scans were performed on single grains in order to
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 377

Fig. 6A - E. Textural types of barite from Tsunokakezawa No.1 ore body, Fukazawa mine. A Fan-
like barite blades showing a typical dusty-looking fine-grained zone 2 (Z2), and a clean coarser-
grained zone 1 (ZJ). Plane transmitted light. B A polycrystalline barite overgrowth (zone 1) is suc-
ceeded inwards by zone 2 and a central zone 1. Plane transmitted light. CFanlike barite blades of tex-
tural zone 1 with an almost perfect tabular outline. Transmitted light, crossed nicols. D Zoned barites
(Zt and Zz). Grain (0) is composite like those in A and B. The rim of (b) is monocrystalline. Transmit-
ted light, crossed nicols. E Randomly oriented barite grains (as indicated by their zonation) some of
which (0) form a single grain, apparently resulting from recrystallization. Transmitted light, crossed
nicols

characterize the variations in internal composition. Details of the analytical

procedures are in Kalogeropoulos (1982, Appendix I).
Table 1 summarizes the strontium variations along barite blades which form
constituent units of fanlike barite aggregates. These barite blades in turn may con-
sist of two, three or four growth zones which can texturally be classified as zone 1
or 2. These textural zones possess different strontium contents, high values
coinciding with the texturally dusty-looking (fine-grained) barite stage of zone 2.
378 S. I. Kalogeropoulos and S. D. Scott

Table 1. Mean SrO (wtOJo ± 1 if') concentrations of texturally-zoned fanlike barite blades from the
massive barite ore zone of the Tsunokakezawa No.1 ore body (Fukazawa mine)b

Sample Early zone 1 Intermediate zone 2 Late zone 1

F.20.6 0.97 ±0.13(7) 2.14±0.30(9) 0.95 ±0.07(2)

a 1 s = One standard deviation.

b Numbers of analyses are in parentheses.

Table 2. Mean SrO (wt% ± 1 sa) concentrations of texturally-zoned tabular and vein barites from the
massive barite ore zone of the Tsunokakezawa No.1 ore body (Fukazawa mine)b

Sample and Core Intermediate Edge

textural zonation

F.20.6A 0.20±0.17(3) 0.70(2) 0.98 ± 0.13(3)

Zone 1
F.2.13A 1.14 ± 0.27(14) 2.63 ± 0.34(15) 1.02 ± 0.29(24)
Zones 1-2-1
F.20.6B 2.52 ± 0.33(6) 1.26±0.15(5) 1.08 ± 0.23(5)
Zones 2-1
F.20.6C (veinlet) 1.12 ± 0.19(3) 0.77 ±0.06(3)
Zone 1 (early) (late)

a 1 s = One standard deviation.

b Numbers of analyses are in parentheses.

Table 2 summarizes the variations in concentration of strontium (from the

core, through the intermediate zone to the rim) in texturally-zoned tabular and
vein barite. The average strontium content of these barites varies from 0.2 to
2.6 wt% SrO (total range 0.1 - 3.2 wt% SrO). They exhibit three characteristic
zonation trends:
1. An increase in SrO from 0.20 at the core to about 1.0 wt% at the edge (Table
2, texturally zone 1);
2. An increase of SrO from about 1.1 wt% at the core to 2.6 wt% at an inter-
mediate zone which texturally coincides with zone 2, and a subsequent de-
crease at the edge where SrO concentrations approach those at the core
(1.0 wt% SrO; Table 2; textural zonation 1-2-1); and
3. A decrease from about 2.5 wt% SrO at the core to a value of about 1.1 wt% at
the edge (Table 2; textural zonation 2-1).
A late stage barite veinlet crosscutting the bedding of the massive barite ore
zone shows a decrease in SrO from 1.1 wt% in an early formed zone to 0.77 wt%
in a later zone (Table 2; texturally zone 1). Taken together, these data for tabular
and vein barites show a consistent systematic change in composition which is
illustrated in Fig. 7. Early formed barite contains about 0.2 wt% SrO, followed
by intermediate barite at 2.6 wt% SrO and late stage barite at about 0.8 wt%
SrO.
 On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 379

F. 2 .13A F. 20 .6

~~ 2
ZONE 2
---A-
3

-
- I z~f ~5mm=Sf I
IANALYTICAl
c
IERROR Q)
c ZONE 2
0
Sr0 2 ZONE 1
u
.....
(Wt.%) (j)

-
Q)
>
( 3)
m
(2 ) Q)
a:
(3 )
o~~--------------------~
TIME ( LATE)
•
(EARLY)

Fig. 7
Fig. 7. Evolution of SrO (wtOJo) content in tabularlike barite grains. Note the similarity in trend and
concentration level with those in Table 1. The bars incorporate one standard deviation of the mean
and the analytical error

Fig. 8. Spectral scans for Sr(Ka) of two barite grains from Fukazawa (F.2.13A and F.20.6) and one
grain from Furutobe (FUR-86). Note the sharp increase in strontium coinciding with textural zone 2.
Vertical bars represent a relative SrO content of 1 wtOJo

Comparing Table 1 and Fig. 7 one can clearly see not only similar zonation
trends, but also identical SrO concentrations for both the barite blades and single
tabular barite grains. These data together with textural evidence suggest that the
coarse-grained tabular barites most likely evolved from the fanlike or tabular
sheafs by a process of isochemical recrystallization.
Spectral scans for Sr across single grains of barite (Fig. 8) reveals that the
transition from zones of low to high Sr content is generally abrupt and that the
strontium-rich zones represent a small fraction of the total barite volume. Conse-
quently, the chemical and the concomitant textural changes during barite precipi-
tation, whatever their cause, are characterized by an ability to change abruptly
and by a short residence time in the system.

Sulfur Isotopes

Three pyrite samples from the barite ore zone have ~3~ values of + 3.8 and
+ 3.9%0. These lie within the range of ~3~ ( - 0.6 to + 6.05%0; Kalogeropoulos
and Scott 1983) for pyrites from the overlying tetsusekiei bed at Tsunokakezawa
No.1, suggesting a similarity in their environments of formation.
If pyrites are assumed to have attained sulfur isotopic equilibrium with the
coexisting barites (~3~ = + 21.5%0; Kusakabe and Chiba 1983), a wide range of
temperatures, all over 330°C, are obtained from the sulfur isotopic geothermo-
meter (Sakai 1968, Nakai 1970, Ohmoto 1972). These temperatures are in dis-
380 S.1. Kalogeropoulos and S. D. Scott

agreement with the tight cluster (150° -210°C) obtained from oxygen isotopic
geothermometry for the pair barite-water (Kusakabe and Chiba 1983). There-
fore, sulfur isotopic geothermometry based on sulfate-sulfide equilibria does not
appear to give a reliable estimate of temperature, probably because pyrite and
barite from the barite ore zone were not precipitated together, as we have shown
above.

Fluid Inclusions

Barites from the bedded barite of Tsunokakezawa No. 1 ore body have two-
phase (vapour + liquid) fluid inclusions of variable size (10- 30 J.l) and shape
(tabular, ovoid, irregular). Vapour to vapour + liquid ratios range from less
than 0.05 to almost 1. Because of the possibility of leakage along cleavage planes
during heating, fluid inclusions were subjected to freezing experiments before
they were homogenized. All measurements were made on the Linkam TH600
stage (Macdonald and Spooner 1981, details of the analyses are in Kalogero-
poulos 1982, Appendix II). The salinity of the 14 inclusions examined is quite
constant [4.6 ± 0.4(1 s) 3 wtO/o eq. NaCI] and falls within the 3 to 6 wt% equivalent
NaCI obtained for barites from the black ore of Kosaka, Furutobe, Ainai and
Shakanai Kuroko deposits (Lu 1970, Watanabe 1970, Tokunaga and Honma
1974).
The coexistence of vapour-rich with liquid-rich inclusions could be a result of
boiling, leakage or necking down. All of the fluid inclusions studied homo-
genized into the liquid phase and at highly variable temperatures (Table 3). Boil-
ing can thus be eliminated as an explanation.
There is no unequivocal evidence that the inclusions have not leaked. Howev-
er, leakage of a fluid from any inclusion that has not exsolved a daughter crystal
will not affect the salinity estimates derived from that inclusion (although it will
affect homogenization temperatures).
The morphological and microthermometric properties of fluid inclusions in
barite from beds in the Fukazawa mine and from the active geothermal field of
Milos Island (Greece) which shows similar characteristics to Kuroko (Kalogero-

Table 3. Range of homogenization temperatures, mean (± 1 sa) and range of salinity in wtlllo eq. NaCI
for barite from Milos (Greece) and the bedded barite of Fukazawa mine (Japan)

Deposit Sample no. Salinity

wtlllo eq. NaCI

Milos MA 169.6-304 n.d. b

Fukazawa F.2.4 130.8-302 -2.79±0.23 4.61 ±0.4
(-2.4 to -3.2) (4.02 to 5.25)

a 1 s = One standard deviation.

b n.d. = not determined.

3 1s = One standard deviation.

On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 381

o1...-_ _ _---',
50 iJ.

Fig. 9A, B. Fluid inclusions in barite from hydrothermal deposits. A Milos, Greece. Note the process
of necking-down. B Bedded barites, Fukazawa mine. Note the slightly camouflaged necks

poulos and Mitropoulos 1983) are compared in Fig. 9A, B and Table 3. Both sets
of inclusions share characteristics of variable size, vapour Iliquid ratio and
homogenization temperature. The Milos inclusions, however, show indisputable
evidence of necking down (Fig. 9A). The evidence is not so obvious in the
Fukazawa inclusions, but these also can be seen, in places, to have necked down
(Fig. 9 B).
In conclusion, fluid inclusions in barite from the Fukazawa mine have cer-
tainly undergone necking down and possibly some leakage. Neither process will
affect the validity of the salinity estimates, but either would render homogeniza-
382 S.l. Kalogeropoulos and S. D. Scott

tion temperatures useless for the purpose of geochemical modelling. The latter
point is well illustrated by the consistently large ranges of homogenization
temperatures obtained from inclusions in barite from several Kuroko deposits
(Tokunaga and Honma 1974, this chapter, Table 3).

Parameters Affecting the Distribution of Strontium in Barite and Estimation

of Strontium Content in Solution

Textural and chemical zonation on the scale of single barite grains precludes the
use of equilibrium thermodynamics to describe a chemically dynamic system as a
whole. However, certain individual subsystems may be treated using equilibrium
theory. An example of such a subsystem is an individual, texturally and chemi-
cally homogeneous zone of a single barite grain. Changes in the strontium con-
tent of each zone can be used to trace variations in the strontium composition of
the coexisting fluid from which it was precipitated.
The partitioning of strontium between (Ba, Sr)S04 and a coexisting fluid is
described by:
(BaS04)bar + (Sr 2 +)Soln ~ (SrS04) bar + (Ba 2 +)soln
bar soln X bar soln bar soln
a SrS04 a Ba2 + SrS04 m Ba~+ Y SrS04 YBa~-+
K (p, T) = bar soln bar soln bar soln (1)
a BaS04 a sr2+ X BaS04 m sr2+ Y BaS04 Y sr2+
K~~r = (XsrS04 ) bar -;- (mSr2+ )SOln
(1 a)
X BaS04 mBa2+

where: K(p,T) = equilibrium constant, K~~ = distribution coefficient of Sr in bar-

ite, a = activity, X = mole fraction, m = molality (mol kg -1 HP), and Y = activ-
ity coefficient.
The partitioning of strontium in barite is a function of temperature, pressure,
(msr2+/mBa2+) in solution and the mixing properties of the BaS04- SrS04 solid
solution. Moreover, experimental work (Doerner and Hoskins 1925, Gordon et
al. 1954) has shown that the rate of crystal growth may affect the partitioning be-
haviour of strontium between solid and liquid phases. All of these vary with the
degree of mixing between hydrothermal solution and cold seawater at the time of
barite precipitation.
If the fluid was relatively stationary in a closed system and barite precipitated
slowly, the mSr2+/mBa2+ ratio in the fluid phase is expected to increase gradually
with barite precipitation (Gordon et al. 1954). This would result in a progressive
strontium enrichment from early to late barites as they were precipitated out of
this solution. However, such a mechanism can be ruled out for the Fukazawa
barites by the observations that: (1) the relative increase of Sr in the barites stud-
ied in sharp rather than gradual (Fig. 8); and (2) the trend of zonation in barites
(Figs. 7 and 8) is that of an initial increase and subsequent decrease. Therefore,
the mineralizing solution must have been circulating. During the course of such
circulation the hydrothermal solution has been mixing with cold seawater. This
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 383

process is further strengthened by evidence from the sulfur isotopic data of the
Kuroko barites and their morphological characteristics, both of which are men-
tioned earlier in this text.
Barites of the textural zone 1 from the barite bed contain less strontium than
those of the textural zone 2 (Table 1). This observation may represent either (1)
precipitation of the textural zone 1 barites from a hydrothermal solution contain-
ing a low concentration of strontium or (2) a composition closer to "equilibrium"
when thermodynamics began to rule over kinetics. Since no study of kinetics of
barite precipitation has to our knowledge, been made to date we tend to incline
towards accepting case (1). This speculation is further supported by the observa-
tion of Shikazono (personal communication 1979) that barites from stratigraphi-
cally lower zones of the Fukazawa deposit contain less strontium than those from
the barite bed.
Equation (1a) shows that the strontium content of barite is a function of the
distribution coefficient K~~r and of the msr2+lmBa2+ ratio in solution. K~~ de-
creases with decreasing temperature (0.4 at 300°C and 0.16 at 200°C; Starke
1962). Thus, if a low-strontium ascending hot mineralizing solution were to mix
with cold seawater the effect of temperature and that of changing strontium con-
tent would be opposite. Any barite precipitated will have a strontium content de-
termined by the balance of these two parameters.
The evolution in concentration of strontium in solution from which the Fuka-
zawa barites were precipitated can be calculated from the following reorganiza-
tion of Eq. (la):

( msr2+ ) soln 1 ( mBa2+ )soln • (XSrS04IXBaS0 ) bar •

=~ (2)
4
KSr

Even though the X SrSO/X BaS04 ratio in the individual stages of barite precipita-
tion is known accurately from their strontium content, both the (mBa2+ ) in solu-
tion and lIK~~r, which have opposite effects on the (msr2+) in solution upon mix-
ing, are indeterminate because temperature and salinity data for each particular
zone of single barite grains are lacking. Consequently, certain assumptions were
made in order to estimate the strontium content in solution:
1. Barites from the Tsunokakezawa No. lore body of Fukazawa mine
formed from a hydrothermal solution with an average temperature of 250°C
(Tokunaga and Honma 1974, Pisutha-Arnond and Ohmoto 1983) and salinity of
0.8 m NaCI (see proper section); and
2. mBa2+ = msOi- during the texturally different stages of barite precipitation.
Figure lOA summarizes the means and ranges of estimated concentrations of
strontium in solution from which bedded barites and barite veins were formed
(Kalogeropoulos 1982). Figure lOB shows strontium isotopic data for the above
types of barites and for Miocene volcanic rocks (Farrell 1979). Both the isotopic
and compositional data indicate a greater involvement of seawater during precip-
itation of the bedded barites than for the veins. The ranges in Fig. 10 indicate
that the barite veins precipitated from a solution with rather uniform strontium
concentration and isotopic composition. On the other hand, the bedded barites
formed from a solution with variable strontium isotopic values and strontium
384 S.l. Kalogeropoulos and S. D. Scott

Fig. 10. A Means and ranges of calculated

SEA WATER D
strontium in solution (ppm) for a barite veinlet
(x) and for bedded barite zones 1 (.) and 2
1----o>-------.
... --11 BE ODED BA RITE (A). The value of seawater (0) is from Kraus-
kopf (1967). B Means and ranges of 87Sr/86Sr,
I*-i BARITE VEINLET ratios for Miocene volcanics, Fukazawa
barites (veins and bedded) (Farrell 1979) and
12345678 contemporaneous seawater (Dasch and
Calculated Sr in solution, ppm Biscaye 1971)

®
SEAWATER D

1 - - -........--11 BEDDED
BARITE
t-+--I BARITE VEINS

1 - - _ - - - 1 1 MIOCENE VOLCANICS

0.706 0.707 0.708 0.709

87Sr /86 Sr

content, apparently resulting from variable degrees of mixing between ascending

hydrothermal solution and seawater. The addition of strontium from seawater
during precipitation of the fine-grained zones (zone 2) must have resulted in a
concomitant enrichment of these zones in 87Sr. Therefore, the strontium isotopic
values in each barite sample will be dependent upon the relative abundance of
such growth zones.

Conclusions: A Model for Formation of the Barite Zones

Since the solubility of barite in hydrothermal solutions is quite low (Blount

1977), there are two plausible ways of transporting and depositing barite in the
Kuroko deposits: (1) oxidation of a hydrothermal fluid containing barium and
reduced sulfur; and (2) mixing of a barium-rich hydrothermal fluid with seawa-
ter. Sulfur isotopes of Kuroko barites regardless from which part of a deposit
they come (Bryndzia et al. 1983, Kusakabe and Chiba 1983), are consistent with a
derivation of the barite-sulfur directly from seawater sulfate. Consequently, mix-
ing of an ascending barium-rich hydrothermal fluid with seawater (case 2) is the
most probable mechanism for the precipitation of the Kuroko barites. A subordi-
nate amount of barite may have precipitated by process (1). The active participa-
tion of seawater during the course of barite formation is further evidenced by the
oxidation of pyrite to hematite, and by the fine-grained (dusty-looking) barites
characterized by higher strontium content and 87Sr/81Sr ratios (Farrell 1979).
The strontium content of barites precipitating out of a stationary solution in a
closed system is expected to increase progressively with time. However, the ob-
served sharp rather than gradual increase in strontium content within single bar-
ite grains (Fig. 8), and the initial increase followed by a decrease of strontium in
zones barites (Fig. 7) require that the mineralizing solution was circulating.
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 385

Textural evidence indicates that bedded barites at Fukazawa have formed

within a pyritic tuff (Fig. 3) by displacing the silicate matrix. This displacement
seems ·to have been caused by the circulating mineralizing fluid and probably
through a process of fluidization of the unconsolidated tuffaceous bed. An un-
consolidated bed fluidizes when the rate of fluid flow passing through it reaches a
point at which the bed expands slightly, and the particles move apart and are sup-
ported in the upward flowing hydrothermal solution. As the tuffaceous bed ex-
pands during fluidization the redox boundary is displaced upwards (Fig. 11,
stage A). Seawater sulfate persisting probably metastably within the reducing
zone (Malinin and Khitarov 1981) is consumed by barite formation as the front

SEDIMENTATION ( ?)

UPWARD-MOVING
( FORMING) BARITE
2+ 2-
Bo
SO.. )
+METASTA8lE (

ZONE 1 ~~~~5~~t
., DENSE BARITE BED sw
HYDROTHERMAL SOLUTION
STAGE ®
ZONE 2-+~~

STAGE ©

ZONE 1

Fig. llA - C. Stepwise development of the barite ore zone of Kuroko-type ore deposits (not to scale).
High sedimentation rates or relatively massive barite at the base of the pyritic tuff result in channel-
ling of the hydrothermal fluid to the seafloor and consequently much compositional variation in the
overlying tuffaceous unit. During stage A, ascending barium-rich hydrothermal solutions mix with
seawater sulfate. They form barite of the textural type represented by zone 1, within a fluidized pyrit-
ic tuff (see text for details). The fluid flow rates during this stage are expected to be lower than those
of massive sulfide formation, but high enough to initiate and sustain fluidization of the tuffaceous
bed. Stage B, represents the collapse of the fluidized bed as a result of decline of fluid flow rates. Dur-
ing this stage pyrite oxidizes to hematite and fine-grained barite (i.e. textural zone 2) forms. These
barites have higher strontium contents than those measured in textural zone 2, and strontium isotopic
values approaching that of seawater. Stage C is characterized by channelling of the ascending hydro-
thermal solutions due to the already deposited barite and tetsusekiei beds. These solutions result in
the sulfidation of hematite, replacement of barite mainly by pyrite and overall recrystallization
386 S.1. Kalogeropoulos and S. D. Scott

of the barium-rich hydrothermal solution advances upwards. A certain amount

of barium will also be taken up by the clay fraction of the system. These initially
formed barites may be suspended or transported upwards depending on their
mass. As they progressively grow to a size that cannot be further supported by
the drag force of the upwelling hydrothermal solution, these barites begin to set-
tle, forming a dense barite bed (textural zone 1) at the base of the fluidized pyritic
tuff (Fig. 11, stage A).
There are at least three lines of evidence which support a sudden, short, but
nevertheless major influence of seawater on the barite ore environment: (1) the
sharp increase of strontium concentration in single barite grains (Fig. 8); (2) the
smaller volume of the strontium-rich (fine-grained barite) relative to the stronti-
um-poor (coarser-grained barite) zones (Fig. 6); and (3) the sulfur isotopic ratios
of all analyzed barites (Kusakabe and Chiba 1983) which approach that of con-
temporaneous seawater ( + 22%0). This significant participation of seawater may
have occurred during a stage of waning intensity of the hydrothermal activity
when incursion of seawater into the environment of barite growth is possible
(Fig. 11, stage B). During this stage the tuffaceous bed is no longer fluidized and
becomes static; the redox reaction involving pyrite and hematite (Fig. 4D - F)
merely provides a chemical indication of the incursion of oxygenated seawater
and this seawater also provides sulfate to precipitate fine-grained barite (Le. tex-
tural zone 2) from sulfate-deficient ascending hydrothermal solutions. These ver-
tical fluctuations of the redox boundary apparently due to fluctuations in the in-
tensity of the hydrothermal activity are also demonstrated at Fukazawa by the in-
tervening hematite-rich bands in sulfide-bearing barite (Fig. 5). This mechanism
provides for replenishment of the barite environment with seawater sulfate that is
initially stable (downward flow of seawater) and subsequently metastable (up-
ward flow of hydrothermal fluid) as indicated by the replacement of barite by
sulfides. The seawater passes on to the barite not only its sulfur (Kusakabe and
Chiba 1983), but also its strontium isotopic characteristics (Farrell 1979), thus ex-
plaining the inconsistency in stratigraphic relationships between strontium iso-
topes and strontium contents of barites discussed previously. When the hydro-
thermal system re-establishes itself, barites continue to form (equivalent to zone
1) and re-crystallize (Fig. 6). This is followed by a stage of pyrite (py II) deposi-
tion as pseudomorphs after barite (Fig. 4A, C and Fig. 11, stage C).
The mechanism of barite precipitation within an unconsolidated pyritic tuff
which is a precursor of the tetsusekiei bed has contributed significantly to the na-
ture of the overlying tetsusekiei, which acquires its final chemical and textural
characteristics during the conclusion of the ore-forming process (Kalogeropoulos
and Scott 1983). This may coincide with a stage of copper-mineralization (or Cu-
remobilization) evidenced by the "chalcopyrite disease" exhibited by sphalerites
from the barite ore zone. Because the hydrothermal fluid-seawater system has large
ranges of temperature and chemistry existing simultaneously in different parts of
the system and because the system as a whole evolved chemically and thermally in a
complex fashion punctuated by repetitive alternating incursions of sea- and hydro-
thermal waters, equilibrium thermodynamics cannot treat this system as a whole.
Our future research to elucidate the genesis of these barites further should be
focussed on the following two major aspects. The first, involves an oxygen and
On the Genesis of Barite-Associated with Volcanogenic Massive Sulfides 387

sulfur isotopic study of the two different textural types of barites, as they were
identified in this work (Le. zone 1 and zone 2); the second, and most important,
involves a mathematical modelling of the depositional processes.

Acknowledgements. Our studies on barite were conducted as part of the U.S.-Japan-Canada Co-op-
erative Research Project on the Genesis of Volcanogenic Massive Sulfide Deposits (Ohmoto and
Skinner 1983). We acknowledge, in addition to our colleagues on the project, the assistance of and
valuable discussions with the following people on specific aspects of this investigation: L. T. Brynd-
zia, C. Cermigniani, M. P. Gorton, E. T. C. Spooner, University of Toronto; and N. Shikazono of
Tokyo University. Special thanks are extended to the Dowa Mining Co. and particularly to K. Hashi-
moto and J. Date. Moreover, we wish the express our sincere thanks to Drs. P. B. Barton Jr. and
D. J. Vaughan for their constructive criticism which improved the text of this manuscript.
Funding was provided to S. D. S. by the Natural Sciences and Engineering Research Council of
Canada operating grant A7069; the Ontario Geoscience Research Fund grant number 3-706-111-30;
the Natural Science and Engineering Research Council of Canada; and a Japanese Society for the
Promotion of Science Exchange Fellowship; and to S. I. K. by University of Toronto Fellowships and
Ontario Graduate Scholarships.

References

Barton PB Jr (1978) Some ore textures involving sphalerite from the Furutobe mine, Akita Prefec-
ture, Japan. Min Geol 28:293 - 300
Barton PB Jr, Bethke M, Toulmin PIlI (1963) Equilibrium in ore deposits. Min Soc Am Spec Pap
1:171-185
Blount CW (1977) Barite solubilities and thermodynamic quantities up to 300°C and 1400 bars. Am
Miner 62:942 - 957
Bryndzia LT, Scott SD, Farr JE (1983) Mineralogy, geochemistry and mineral chemistry of siliceous
ore and altered footwall rocks in the Uwamuki No.2 and No.4 deposits, Kosaka mine, Hokuroku
District, Japan. Econ Geol Monogr 5:507 - 522
Dasch EJ, Biscaye PE (1971) Isotopic composition of Sr in Cretaceous-to-Recent pelagic forminifera.
Earth Planet Sci Lett 11 :201 - 204
Doerner HA, Hoskins WM (1925) Co-precipitation of radium and barium sulfates. J Am Chern Soc
47:662-675
Farrell CW (1979) Strontium isotopes of Kuroko deposits. PhD Thesis, Harvard Univ, Cambridge,
Mass
Farrell CW, Holland HD (1983) Strontium isotope geochemistry of the Kuroko deposits. Econ Geol
Monogr 5:320-328
Gordon L, Reimer CC, Burit PB (1954) Distribution of strontium within barium sulfate precipitation
from homogeneous solution. An Chern 26:842 - 846
Green GR, Ohmoto H, Date J, Takahashi T (1983) Whole-rock oxygen isotope distribution in the Fu-
kazawa-Kosaka area, Hokuroku District, Japan. Econ Geol Monogr 5:395 - 411
Hanor JS (1968) Frequence distribution of compositions in the barite-celestite series. Am Miner
53:1215 -1222
Hashimoto K (1977) The Kuroko deposits of Japan - geology and exploration strategies. Assos Ing
Min Metal Geol Mex Mem Tech 13:25 - 88
Igarashi T, Okabe K, Yajima J (1974) Massive barite deposits in West Hokkaido. In: Ishihara S (ed)
Geology of the Kuroko Deposits. Min Geol Spec Iss 6:39 - 44
Kalogeropoulos SI (1982) Chemical sediments in the hanging wall of volcanogenic massive sulfide de-
posits. PhD Thesis, Univ Toronto, Can
Kalogeropoulos SI, Mitropoulos P (1983) Geochemistry of barites from Milon island (Aegean Sea),
Greece. N Yearb Miner Mh 1:13-21
Kalogeropoulos SI, Scott SD (1983) Mineralogy and geochemistry of tuffaceous exhalites (tetsusekiei)
of the Fukazawa mine, Hokuroku District, Japan. Econ Geol Monogr 5:412-432
Krauskopf BK (1967) Introduction to geochemistry. McGraw-Hill, New York, p 721
388 S.1. Kalogeropoulos and S. D. Scott: On the Genesis of Barite-Associated

Kusakabe M, Chiba H (1983) Oxygen and sulfur isotope composition of barite and anhydrite from
the Fukazawa deposit, Japan. Econ Geol Monogr 5:292-301
Lu LE (1970) In: Aoki K, Sato T, Tatsumi T (eds) Kuroko deposits and Towada volcanoes. IMA-
IAGOD Meet 1970, Tokyo-Kyoto, Guideb 3, 53 pp
Macdonald AJ, Spooner ETC (1981) Calibration of a Linkam TH 600 programmable heating-cooling
stage for microthermometric examination of fluid inclusions. Econ Geol 76:1248 -1258
Malinin SD, Khitarov NI (1981) Fluid inclusion study of stockwork siliceous orebodies of Kuroko de-
posits at the Kosaka mine, Akita, Japan. Min GeoI28:349-360
Nakai N (1970) Isotopic and chemical equilibrium of C~ - C~ and pyrite-anhydrite in geothermal
areas in Japan. Int Symp Hydroch Biochem, Tokyo (Abstr), p 22
Ohmoto H (1972) Systematics of sulfur and carbon isotopes in hydrothermal ore deposits. Econ Geol
67:551-578
Ohmoto H, Skinner BJ (eds) (1983) The Kuroko and related volcanogenic massive sulfide deposits.
Econ Geol Monogr 5:604
Pitsutha-Arnond V, Ohmoto H (1983) Thermal history, and chemical and isotopic compositions of
the ore-forming fluids responsible for the Kuroko massive sulfide deposits in the Hokuroku district
of Japan. Econ Geol Monogr 5:523 - 558
Sakai H (1968) Isotopic properties of sulfur compounds in hydrothermal processes. Geochem J
2:29-49
Sato T (1974) Distribution and geological setting of the Kuroko deposits. In: Ishihara S (ed) Geology
of the Kuroko Deposits. Min Geol Spec Iss 6:1 - 9
Starke R (1962) Die Strontiumgehalte der Baryte. Freiberg Forschungsh C:150
Tokunaga M, Honma H (1974) Fluid inclusions in the minerals from some Kuroko deposits. In: Ishi-
hara S (ed) Geology of the Kuroko Deposits. Min Geol Spec Iss 6:385 - 389
Watanabe M (1970) Kuroko deposits and Towada and Hakkoda Volcanoes. IMA-IAGOD Meet, To-
kyo-Kyoto, Guideb 3,53 pp
Part IV
Sediment-Hosted Deposits
Diagenetic Features at White Pine (Michigan), Redstone
(N. W. Territories, Canada) and Kamoto (Zaire) *.
Sequence of Mineralization in Sediment-Hosted Copper
Deposits (Part 1)
A. C. BROWN and F. M. CHARTRAND 1

Abstract

Disseminated copper and copper-iron sulfide mineralization at the White Pine

(Michigan), Redstone (N. W. Territories, Canada) and Kamoto (Shaban Pro-
vince, Zaire) deposits are interpreted to have been emplaced during early diagen-
esis by an influx of dissolved copper into initially pyritic basal units of grey-bed
host rocks. This concept is supported by the position and configuration of the
mineralized zones, by numerous textural features, by the zoning of Cu- and Fe-
bearing sulfides as well as affiliated metals (e.g. Pb, Zn, Co and Mo), and by
comparisons with modern analogs of strata hosting stratiform copper deposits.
Most evidence supports an early diagenetic introduction of copper from underly-
ing coarse-grained red beds. Plausible sources of copper include the red beds
themselves, or pene-exhalative solutions which "exhale" into the red beds. Pre-
liminary quantitative data now available suggest that the red beds may have pro-
vided sufficient copper to mineralize basal units of the overlying grey beds; the
tectonic setting and timing of mineralization also encourage further research into
the pene-exhalative model.

Introduction

The years 1960 -1962 were pivotal years in our evolving understanding of the
genesis of stratiform copper deposits. With few exceptions, it was quite normal
until that time to interpret the fine-grained sedimentary-like distribution of cu-
priferous sulfide minerals along bedding as evidence of a sedimentary origin, or
else in another school of thought, to attribute these sedimentary-like features to
magmatic hydrothermal replacements, i.e. to selective telethermal mineralization
a la Lindgren and Gratton.
In 1960, W. S. White of the U.S. Geological Survey instigated several impor-
tant studies of the well-preserved, unmetamorphosed White Pine (Michigan) de-
posit when he reported details on the distribution of base-metal sulfide minerals

* Contribution of the Mineral Exploration Research Institute, Montreal, Canada.

1 Ecole Polytechnique, Dept. of Mineral Engineering, P.O. Box 6079, Sta. "A", Montreal, Quebec,
H3C 3A7, Canada

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Diagenetic Features at White Pine, Redstone and Kamoto 391

in the excellent exposures at the White Pine mine (White 1960). He described a
vertical zoning of Cu,Fe-bearing sulfides, with the ore-zone chalcocite ( + native
copper and silver) grading abruptly into the overlying pyritic zone through an or-
derly array of chalcocite-+bornite-+chalcopyrite-+pyrite; this narrow transition
zone reportedly formed a single undulating blanket-like surface cross-cutting
bedding at gentle angles. Furthermore, White noted the occurrence of minor
amounts of CdS immediately above this bedding-transgressive Cu-Fe transition.
The idea of cupriferous fluids moving upward through the host sediments during
early diagenesis captured the attention of Wm. C. Kelly at the University of
Michigan, and studies designed to test this concept were set in motion.
In 1962, Bartholome commented on the Cu-Fe zoning found in the Kamoto
deposits (Shaban Province, Zaire), and described a mineralogy and zoning of
Cu,Fe-bearing sulfides which parallels the array studied independently at White
Pine. He illustrated in careful detail the fine-scale replacement of pyrite by pro-
gressively more cupriferous sulfides, and noted the gently cross-cutting nature of
the mineral zones grading from a basal chalcocite-bornite zone to an overlying
pyritic zone in the Kamoto district. As a result, an upward influx of cupriferous
solutions was proposed for Kamoto as well.
From those years onward, research into the genesis of stratiform copper de-
posits has been predominantly oriented toward testing and refining this model
which calls for an early diagenetic influx of copper into formerly iron sulfide-rich
host sediments. The present paper is not intended to be a compilation of
observations and interpretations supporting this hypothesis - summaries of that
sort can be found in Rentzsch et al. (1976), Brown (1978, 1980a,b, 1981a),
Gustafson and Williams (1981) and others. It is interesting to note, however, that
the early diagenetic model has survived tests of its accuracy during more than two
decades of detailed investigations of many different stratiform copper deposits
by many researchers.
Secondly, it is pertinent to point out that much progress has been made since
the early 1960s. Until that time, the study of an ore deposit was typically aimed at
descriptions and interpretations of the ore minerals, commonly without great at-
tention to the nature and origins of enclosing host rocks. The latter was due in
good part to the all too pervasive pre-1960 perception that the origins of the host
rocks belonged to pre-ore history and that the prime prerequisite of the host rock
was that firstly, it be structurally deformed to allow entry of magmatic ore solu-
tions and that secondly, it be chemically reactive at the local and microscopic
scale so as to precipitate metals from invading hydrothermal fluids. Today strati-
form copper mineralization is seen to be intimately related in time and space to
the origins of the enclosing host rocks. In fact, mineralization may be at least in-
directly interactive with environments prevailing in the contemporaneous hydro-
sphere, biosphere and atmosphere during deposition of the host sediment (Brown
and Chartrand 1983).
Current evidence for the diagenetic 2 emplacement of copper may be "con-
vincingly for" or at least "permissive", and it is significant that the observations
made thus far belong to these categories and do not conflict with the diagenetic

2 Footnote 2 see page 392.

392 A. C. Brown and F. M. Chartrand

hypothesis. For example, the replacement of pyrite (or a more primitive iron
sulfide), the zoning of sulfides and the gently cross-cutting nature of the sulfide
zones form strong arguments, alone and together, for a post-sedimentary influx
of copper from a source outside of the host sediments.
Other studies, such as those on the derivation of copper from adjacent red
beds (e.g. Lustwerk and Rose 1983, Rose et al. 1984) or pene-exhalative sources
(Brown 1981 b, 1984), have described "permissive" evidences for the diagenetic
influx model - permissive in the sense that the evidence does not conflict with
the diagenetic concept.
In some cases, one may qualify such evidence as being "complementary" or
even "essential". For example, the diagenetic influx model requires adequate
transport mechanisms to introduce copper to the fine-grained portions of the
host rocks, and to bring the copper to the host rocks from plausible distant exter-
nal sources. Calculated solubilities of metals in interstitial brines provide needed
assurance that the metal transport can be adequate. At the same time, it is reas-
suring to find that trace amounts of copper within large volumes of red beds
could release copper to these brines in significant quantities. These observations
on sources and transport of copper are complementary to the diagenetic mineral-
ization model, and in fact the existence of such evidence (or some adequate alter-
native such as the pene-exhalative model) is essential to form a complete source-
transport-deposition model for the diagenetic emplacement of copper in the case
of stratiform copper deposits.

Types of Evidence for the Early Diagenetic Origin of Stratiform Copper Deposits

The precise evidence for the diagenetic influx of copper can only be properly ap-
preciated by direct reference to the extensive literature on stratiform copper de-
posits. The following brief summary is limited to the identification of the types
of evidence currently available on the White Pine, Redstone and Kamoto depos-
its:

1. Position and Configuration of the Mineralized Zone. The cupriferous zone at

White Pine, Redstone and Kamoto is located at the base of a sequence of organ-
ic-rich sediments. The beds overlying the cupriferous zone are pyritic, whereas
the underlying beds are red, hematitic sediments. The location of the cupriferous

2 The usage of the term "diagenetic" with regard to the post-sedimentary emplacement of copper in
stratiform copper deposits may be accepted, on the one hand, and disputed, on the other. Gustaf-
son and Williams (1981) properly limit usage of the term to cases where the copper is redistributed
within the sediments during diagenesis. That concept clearly conflicts with the current model for
stratiform copper deposits in which copper is interpreted to have been introduced from sediments
other than those currently mineralized. However, if one expands one perception of the sediments
involved to include adjacent strata (e.g. underlying red beds), then it may be stated that the copper
was redistributed diagenetically within this single sedimentary sequence.
The term diagenetic is also acceptable if it clearly applies to a period of time characterized by
diagenetic processes; in such cases, time may be denoted by modifiers such as "early" or "late".
Diagenetic Features at White Pine, Redstone and Kamoto 393

zone in the basal reduced beds of an originally iron sulfide-rich stratigraphic unit
lying directly above oxidized sediments would be consistent with a concept
whereby copper was chemically screened from upward moving ore solutions.
The upper limit of the cupriferous zone also transgresses bedding as if it re-
sulted from an upward advancing mineralization front. In the case of White
Pine, the mineralization front is generally abrupt, whereas in the case of Red-
stone and Kamoto, the transgressive nature of the mineralization front is shown
by a general trend from copper-rich to copper-poor sulfides extending over broad
zones, each of which nevertheless transgresses bedding in a gentle manner. More-
over, at White Pine, there is an inverse correlation between the amount of copper
in the lowermost beds of the mine section and the height of the top of the cuprif-
erous zone in the mine section. This correlation strongly suggests that the top of
the cupriferous zone is stratigraphically low where most copper was screened out
of an upward migrating ore solution which added approximately equal quantities
of copper across the base of the host sediments.
2. Zoning oj Metal Sulfides. The stratiform copper deposits of White Pine, Red-
stone, Kamoto and other areas are typically zoned, both with respect to sulfide
minerals of the eu-Fe system, and with respect to other associated minor-to-
trace metals. Among eu-Fe sulfides, the least soluble sulfides such as chalcocite
and bornite occur close to the site of copper influx into the reduced, sulfide-bear-
ing host rocks, whereas the more soluble chalcopyrite and the still more soluble
iron sulfides are located further downstream in the path of the upward circulat-
ing ore solution. The replacement of the more soluble iron-rich sulfides by the
less soluble copper-rich sulfides has been repeatedly confirmed in reports on vari-
ous textural features at White Pine, Redstone, Kamoto and elsewhere.
Other affiliated metals show similar distributions according to their solubili-
ties relative to the eu-Fe sulfides. For example, cadmium, lead and zinc are
clearly located downstream from copper at White Pine, as would be predicted
from their solubilities relative to copper and iron in sulfide-rich host rocks. Re-
cent examinations of lead and zinc at Redstone reveal that these minor metals are
also distributed downstream from copper-bearing sulfides, whereas trace
amounts of molybdenum sulfide at Redstone are located at the base of the miner-
alized zone (i.e. at the site of ore solution influx), again predictable from the low-
er solubility of molybdenite relative to copper sulfides.
The essential point from these observations on metal sulfide zoning is that the
sequence of metal zones is consistent with an influx model, generally pointing to
an upward circulation of ore fluids from adjacent red beds during post-sedimen-
tary time. However, one may ask when, within reasonable post-sedimentary
time, did the fluids enter and mineralize the reduced, sulfide-rich host rocks?
3. Timing oj Copper Mineralization Relative to Early Diagenesis oj the Host
Rocks. Sedimentological studies have provided more precision on the timing of
mineralization relative to the normal diagenesis of the enclosing host rocks. The
data are derived from detailed petrographic analyses and from comparisons with
diagenetic features observed in modern sediments resembling these host rocks.
The Redstone deposit has been studied in this manner, and a close analogy is
drawn here with the sedimentation and early diagenesis of intratidal and supra-
394 A. C. Brown and F. M. Chartrand

tidal (sabkha) sediments deposited within a series of marginal-marine basins

which may be of intracontinental-rift origin.
Copper mineralization at Redstone can be identified as a post-sedimentary
event within the normal early diagenesis of microbial-laminated carbonate beds
of peritidal origin. The sequence of events through sedimentation and early dia-
genesis of the carbonate beds may be summarized as follows (from Chartrand
and Brown 1984):

Stage I. Syndiagenetic Stage: Dolomitization of initial sabkha carbonates; for-

mation of blade-like gypsum; transformation of gypsum to anhydrite; formation
of disseminated pyrite under anoxic conditions in decaying microbial mats.
Stage II. Very Early Diagenesis: Disseminated pyrite euhedra of stage I are
rimmed by epitaxial calcite; gypsum-anhydrite blades are replaced by quartz, cal-
cite and feldspars. Quartz overgrowths consist of more than one generation of si-
lica.
Stage III. Ore-Stage Diagenesis: Copper-bearing sulfides (chalcopyrite, bornite,
digenite, chalcocite) replace quartz, pyrite and remnants of anhydrite; cuprifer-
ous sulfides also form pseudomorphs after blade-like gypsum-anhydrite.
Stage IV. Post-Ore Diagenesis: Neomorphic calcite closes the pores of the miner-
alized carbonate, possibly arresting the process of mineralization by impeding
further influx of ore solution.

The importance of this paragenetic sequence among the early diagenetic min-
erals at Redstone lies in the observation that mineralization was intimately relat-
ed to early diagenesis, but is nevertheless a post-sedimentary event superimposed
on the host-rock carbonates. These microbial-laminated carbonates formed un-
der normal sabkha conditions and proceeded through normal sabkha diagenesis
including the transformation of sulfate phases and the formation of iron sulfide
in the presence of abundant organic matter. However, the normal diagenetic evo-
lution of the sabkha beds was modified by the introduction of copper to the car-
bonates during stage III diagenesis. As mentioned above, the zoning of Cu, Pb,
Zn and Mo sulfides indicates that the direction of copper influx was from the un-
derlying red beds.
Weare currently examining the Zairian mineralization in the same manner,
and we find many parallels with the Redstone mineralization because the initial
sedimentary environment was also of the sabkha type, characterized by microbi-
al-laminated carbonate beds. The proportions of lithotypes, their dimensions
and the concentrations of metals at Kamoto are quite different from those at
Redstone, but the essential lithologies and environmental conditions are similar.
Some important lithologic differences relating to sedimentation and early dia-
genesis deserve attention, however, because these features probably formed close
to the period of early diagenetic mineralization and should limit conditions with-
in the host rocks during mineralization. For example, the Zairian carbonates are
highly magnesian while those of Redstone are predominantly calcitic. In addi-
tion, an enormous volume of authigenic silica has been added to the Zairian sedi-
Diagenetic Features at White Pine, Redstone and Kamoto 395

Stratiform capper

··.........
........ .. mineralization

•••••••••••••••••••••• 00 000
• • • • • • • • • •• 0000 0 0 °0° ...... x ":'. 0 6 0 0°0°00
• •••••••••• • 00000 __oooooooooogogog Red ... beds (hemotlticlogogogogogogo
• •• • • • • • • • • • • • • • • •• • • •• ogogoooo 0 - -- -_-_____ v v 'u-u-o"'oooooooc;ooooooooooooooo
• • • • • • •• • • • • • • • • • • • • • •• ogo~t"\o,..,o,..ot"\o. . . o,.. .... ) 0 0 - --:~==:= Qr~y-beds rp (ilj ) :-_-_-_-_-_-_-_-_-..:
• • • • • • • • • • • 000cCu reservoir )000000000000 0 0 - - _ y c -------------------
••••••••••••••••••••• • ogooo OOOV~~06060g0g0g0g0g0g0g0g0g0g0g0g0g000o0o0000- ~ ~ -0- ~-~-~...:o
•••••••••••••••••••••• 0 cOego 00 ogogogogogogogogogo~o~o~n9nOnonono~~noogogogogogogo
• •
Possible
• • •
felsic ~ II~ oogogogogogogogogogo~ooooooooooooooooooogogogogogogog
•

•••••••••••••••• "
• 0ooooooooooooooooooocRed-beds (hemohhcl0oooooooooooo
• • • •

•
0

°0°0°0°0°0°0°0°0°0°0°0°000°0°0°00
° ° °0°0°0°0°0

. . . . . . ....... ...'I,......
dome • • • • • ••
• I •• ' I • • • • • • • • • • • • • • • • • • • • • • • • • • • •
• • • • • • • • • • • • .. • r , . • • • • • •
• • • • • • • • • •• ·· 1 •••••••••••••••••••••••••••••
. ..................
• •••••••••••••••••
.
Possible alteration • •• •
. . . . . . .....
.·.·.·.·.·.· ·······.·.·.·······
•......•..•..•..••.. .. . .....
................•.....
. . . . . . . . .. 'I ..... ··· .... . . . . . . . .. . ..... .
~
pipes •••••••••••••••••••••• " ••••••••••••••• •••••••••••••••••••••••••••••••••••
• • • • • • • • • • • • •••• •••••••••••••••••••••••
.....•..........................................•I.~

•..•...••.••.•••....• . ........ :.:.: ••..••.•..••..•.••...•..•....•••....•

..••....••••..•.••...• ........ •••.••....••••..••.....•..• ..••.....••..
.... . . . . . . . . . . . . . ..
•.•..•...•.•.•••.
..........•.........
.. ........... .......................
................
. ..................................
•...••.•....
.••.....•• .
................... ....•......... ..•••.....••.•..••• .......... .
....... . ..................... ... .
.. :.:.:.:.:.:.:.:. ~·~:·:·:·:·:·:·:··:·:·:·:·:·:·:·:·~·' I··:·:·:·:··
................ .•.•..••....•.••..•.•.....••. ...... .
. . . . ..
.............. ...•.
............. .••..••......••.....•....••• .. .
. . ..
-.:.:.:.:.: ...•..••..••...•••.•..•••...•.•...••
............. ..
Fig. 1. The fundamental features of a pene-exhalative model which could generate a reservoir of cu-
priferous solutions in permeable red-bed strata underlying initially iron sulfide-bearing and/ or sul-
fate-bearing grey beds (from Brown 1984)

ments, suggesting anormal diagenesis in the early history of these copperbelt host
rocks (Bartholome et al. 1972). Also the Shaban copperbelt ores are cobaltifer-
ous as well as cupriferous, whereas the Redstone mineralization is not cobaltifer-
ous.

Speculations on Ore-Forming Models

Although stratiform copper deposits are commonly considered monometallic

(with, in some cases, accessory economic silver or cobalt), a complete description
of the total added post-sedimentary mineralization commonly includes mention
of significant concentrations of other metals such as Pb, Zn, Mo, etc. The poten-
tial sources of these associated metals, as well as the ore metals, may be summar-
ized in a very limited number of hypotheses consistent with an emplacement of
metals during early diagenesis of the host rocks and with an influx from
underlying red beds.
The derivation of metals from the red beds and associated lithologies, is a
very attractive option. Some very interesting data on this hypothesis are
presented in Lustwerk and Rose (1983) and Rose et al. (1984), and no further
discussion is required in the present paper.
396 A. C. Brown and F. M. Chartrand

A more speculative model which has certain merits is the pene-exhalative

model (Brown 1981b, 1984). This concept is supported in part by general simi-
larities in the tectonic setting of many stratiform copper deposits (including
White Pine and Redstone) and the tectonic model presently adopted by many
geologists in the case of sediment-hosted Pb-Zn deposits of possible submarine
exhalative origin [see, for example, Large (1980, 1983), Finlow-Bates (1980),
Russell et al. (1981) and Russell (1983)]. The tectonic setting is that of intra-cra-
tonic rifting and rapid filling of the downfaulted blocks by red beds and variable
amounts of mafic flows. In brief, it may be possible to generate low-temperature
cupriferous brines which "exhale" into the red beds, and then mineralize overly-
ing basal portions of favourably reducing, sulfide-rich grey-beds (see Fig. 1).

Closing Comment

As suggested initially in this paper, we are not yet sufficiently advanced in our
understanding of stratiform copper deposits to make definitive decisions on their
complete genesis. However, we do have certain guidelines such as those described
here on the timing of mineralization relative to diagenesis, and on the directions
of ore-solution flow. As for long-range transport and ultimate sources of metals,
we are beginning to obtain significant quantitative data, some of which may
eliminate the need for a complex pene-exhalative model. However, in principle,
the discovery of other adequate models does not exclude the possibility of
another potentially adequate model, and at this time the so-called pene-exhala-
tive concept should receive additional thought and research in parallel with other
attractive models.

References

Bartholome P (1962) Les minerais cupro-cobaltiferes de Kamoto (Katanga Ouest). I petrographie.

Stud Univ Louvanium, Fac Sci Kinshasa 14, p 40
Bartholome P, Evrard P, Katekesha F, Lopez-Ruiz JM, Ngongo M (1972) Diagenetic ore-forming
processes at Kamoto, Katanga, Republic of the Congo. In: Amstutz GC, Bernard AJ (eds) Ores in
sediments. Springer, Berlin Heidelberg New York, pp 21-41
Brown AC (1978) Stratiform copper deposits - evidence for their post-sedimentary origin. Miner Sci
Eng 10:145 -163
Brown AC (1980a) The diagenetic origin of stratiform copper deposits. In: Ridge JD (ed) Proc 5th Q
IAGOD Symp, pp 81 - 89
Brown AC (1980b) Stratiform copper deposits and pene-exhalative environments. Geol Soc Am, Ab-
stracts with Programs 13:418
Brown AC (1981a) The timing of mineralization in stratiform copper deposits. In: Wolf KH (ed)
Handbook of stratabound and stratiform ore deposits, vol IX. Elsevier/North Holland Biomedical
Press, Amsterdam New York, pp 1-23
Brown AC (1981 b) Stratiform copper deposits and pene-exhalative environments. Geol Soc Am,
Abstracts with Programs 13 no. 7:418
Brown AC (1984) Alternative sources of metals for stratiform copper deposits. Precambr Res
25:61-74
Diagenetic Features at White Pine, Redstone and Kamoto 397

Brown AC, Chartrand FM (1983) Stratiform copper deposits and interactions with co-existing atmo-
spheres, hydrospheres, biospheres and lithospheres. Precambr Res 20:533 - 542
Chartrand FM, Brown AC (1984) The diagenetic origin of stratiform copper mineralization, Coates
Lake, Redstone copperbelt, NWT, Canada. Econ Geol 80:325 - 343
Finlow-Bates T (1980) The chemical and physical controls on the genesis of submarine exhalative
orebodies and their implications for formulating exploration concepts. A review. Geol Jahrb
040:131 -168
Gustafson LB, Williams N (1981) Sediment-hosted stratiform deposits of copper, lead and zinc. In:
Skinner BJ (ed) Economic Geology. 75th Anniversary Vol, pp 139 -178
Large DE (1980) Geological parameters associated with sediment-hosted, submarine exhalative
Pb-Zn deposits: an empirical model for mineral exploration. Geol Jahrb 040:59-129
Large DE (1983) Sediment-hosted massive sulphide lead-zinc deposits: an empirical model. In:
Sangster OF (ed) Sediment-hosted stratiform lead-zinc deposits. Miner Assoc Can, Short Course
Handb 8:1 - 29
Lustwerk RL, Rose AW (1983) Source and segregation of transition metals during diagenetic forma-
tion of the Redstone stratiform copper deposit, Mackenzie Mts., NWT, Canada. Geol Soc Am,
Abstracts with Programs 15, 6:632
Rentzsch J, Schirmer B, Rolling G, Tischendorf G (1976) On the metal source of non-ferrous miner-
alizations in the Zechstein basement (Kupferschiefer type). In: The current metallogenic problems
of Central Europe. Geol Inst, Warsaw, pp 171 -188
Rose AW, Ohmoto H, Lustwerk RL, Smith AT, Hoy L (1984) Sequence of mineralization in sedi-
ment-hosted copper deposits: Part 3 - Geochemical aspects of the Catskills Formation (Pennsyl-
vania) and Redstone (Canada). Int Geol Congr, Moscow (Abstracts)
Russell MJ (1983) Major sediment-hosted exhalative zinc+lead deposits: formation from hydro-
thermal convection cells that deepen during crustal extension. In: Sangster OF (ed) Sediment-
hosted stratiform lead-zinc deposits. Miner Assoc Can, Short Course Handb 8:251 - 282
Russell MJ, Solomon M, Walshe JL (1981) The genesis of sediment-hosted zinc + lead deposits.
Miner Dep 16:113-127
White WS (1960) The White Pine copper deposit (discussion). Econ GeoI55:402-410
Diagenetic Sulphide Mineralization Within the Stratiform
Copper-Cobalt Deposit of West Kambove (Shaba-Zaire).
Sequence of Mineralization in Sediment-Hosted Copper
Deposits (Part 2)
J. CAILTEUX 1

Abstract

The stratiform copper-cobalt mineralization in the Shaban Mines Group took

place in intertidal sediments enclosing the stromatolitic masses of the R.2.1.3
unit and around similar masses corresponding to the R.2.3.1 unit.
The cristallization of primary sulphides displays a vertical zoning with pyrite
at the base and at the top of the ore bodies. They form micro- and macrocycles
suggesting a sequence of copper and cobalt mineralization. This mineralization
results from a diagenetic process involving a mixing of stagnant interstitial water
with brines which were enriched in metals and which migrated through still
porous sediments. Secondary reactions with the brines enriched previously
formed sulfides with additional copper and/or cobalt.

Introduction

The metallic stratiform mineralization of the Roan Supergroup in Shaba is com-

posed mainly of copper and cobalt sulphides distributed principally along two
mineralized units of the Mines Group (R.2) 2 (Table 1). These units have been
studied in detail at Kamoto (Kolwezi) by Bartholome and his colleagues (1962,
1963, 1969, 1972, 1974).
Prospecting by drilling on the West-Kambove deposit in the heart of the
Shaban copperbelt (Fig. 1) has identified copper-cobalt mineralization similar to
that of the classic copperbelt ore bodies (Cailteux 1983) in the lower R.2.3 unit
(the highest formation of the Mines Group). The presence of a number of oxi-
dized deposits associated with this stratigraphic horizon has been known for
some time, as in the case at Etoile, Kakanda, Kamatanda, Kamfundwa, Luisha,
Shamitumba, etc. The mineralization is concentrated in an alteration pockets be-
tween residual dolomite mounds, but its' origin has never been carefully studied.
Oosterbosch (1962) and Francois (1973, 1974) noted only rare and irregular
occurrences of pyrite as well as copper-cobalt anomalies in the lower R.2.3, and
it has been supposed that the metallic concentrations were produced by a

1 GECAMINES, P.R.D., Bureau d'Etudes Geologiques, Likasi, Shaba, Zaire

2 For terminology see Table 1.

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Diagenetic Sulphide Mineralization Within the Stratiform Copper-Cobalt Deposit 399

Table 1.

Kundelungu K.s Upper Kundelungu Group

Supergroup K.i Lower Kundelungu Group

RA Mwashya Group
R.3 Dipeta Group
Roan R.2.3
Supergroup R.2. Mines Group R.2.2
R.2.1
R.l R.A.T Group
Unconformity _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Kibarian Basement

KASENGA

ECHELLE

o 20 40 60 80 l00Km

LEGEHOE
- .- Llmit.s d. l'Arc
Cuprif.r. shabi.n
- .. - lImltes du Copperbelt
zambien
• Giuments m~tallih!ru
2 •

Fig. 1. General distribution of metalliferous deposits in the Central African Copperbelt

supergene enrichment of minor amounts of background sulphides, as in the case

of the R.4 unit in Shituru (Lefebvre 1974).
Our objective is to examine the sulphide mineralization of West-Kambove in
the regions beyond recent supergene alterations in order to compare the mineral-
ization of the R.2.3 unit with that of usual mineralized bodies of the copperbelt
and to draw genetic conclusions from these observations.
400 J. Cailteux

Sediments of tbe Kambove Mines Group and tbe Distribution of Copper-Cobalt

Mineralization

At Kambove, the Roan Supergroup lies on the Upper Kundelungu Group

(K.S.2.) which crops out further north (Figs. 2 and 3). As is usual in Shaba, the
Roan occurs as "ecaiIles" ("scales" or mega-breccia fragments) of Mines Group
(R.2) strata floating between the crushed R.A.T. Group unit (R.1) and Dipeta
Group (R.3) 3.
It is possible to recognize three very distinct entities within the sediments of
the Mines Group:
1. A dolomitic unit (R.2.1) including stratified dolomites and laminites, with a
massive stromatolitic dolomite at the top and a chloritic microsandstone with
pyroclastic characteristics at the base;
2. A shaly dolomitic unit (R.2.2);
3. A talc-dolomite unit (R.2.3) including stromatolitic massive dolomites
laminites, talc dolomites and chloritic dolomitic shale bands in its upper part.

K ...

EJ
..... ..
superieur
I(undtlungi.n i ~j~~i·~~·~·
K. ,".

rn K.i. 122
.-L Fai>!e
~ R.'
[::::::'·;::1 : 1~~ ~~~ 3
o 5000m

Fig. 2. General geology in the central portion of the Shaban Copperbelt (after Fran~ois)

3 For more complete geographic and stratigraphic information, the reader should consult Cahen
(1954, 1974), Demesmaeker et al. (1962), Bartholome et al. (1972) and Fran~ojs (1973,1974).
 1.5005 5~ON
S200E N200W (1)
E
c
® @ ~
1.400 n'
1.400 (I)

K.i.II/, ... ... ......... ~KUnd~Ung i .n E.

/'
/ 'go
" " sup. Q;
'*' .. .. .. .. Ks . 2 (1)
R.4.2 1 // /.
BRECHE HETEROGENE ~
MWASH I EH SUP. /' ~-"
" ( R .] .t ~s. 2) '" ~.
/'~.4.1 ./ ~2y'
./'MWASH IE H / ' " .~ lit .br0Y",.' " e.
I HF.~ ~ ~.
/! ., lit .. .. ... o·
c
./ j ..... . . tOOO ~
./'/ / J.... " I It .." F II
i'l- '" 0;
t;-1r..
-Co""'> ,- -,..
-4 -'0/111 ~ I 8r.0:,.,-
~" ...
. ,.'!:t-.-'" 4 So.
.. .. S·
tOOO J! ~"'-'!....-.--:::::':'-=___-:-::-:-:-:-::-...L..__---, go
(1)
. 1 ", • '" LEGENOE
~
Ie; G GI R.2 .1.3 !':.
:::;
.~
. .0
.... - - . . . . . . _tod,.JJ-~ ~ -
+ _;t.-.-
- mll R.2.1.1 .t R.2.1. 2
- . .. .. ....'"
1.5005 ~ Kundelun\lien
,. - 3
K$ , 2 o 100 200 m
~ l!::i!I Brec;h. R.3.t hehirogene Q
:g
~
()
Fig. 3. Vertical profile across the Roan of Kambove along the line A-B o
<J
~
o
(1)
'0
o
~.

~
402 J. Cailteux

The R.2.1 and R.2.3 beds are similar in composition and constitute units
characteristic of an intertidal environment at the edge of salt-saturated lagoons
(Cailteux 1977a, 1977b, 1983). The intervening R.2.2. sediments were formed in
a more marine or open lagoonal environment than the R.2.1 or R.2.3 beds.
The West-Kambove deposit is located asymmetrically on the northern flank
of a syncline. The mineralization is situated mainly below and above a stromato-
litic unit (R.2.1.3) in what is called the lower ore body (in R.2.1.1, in R.2.1.2 and
at the base of R.2.1.3) and the upper ore body (at the top of R.2.1.3, in R.2.2.1
and at the base of R.2.2.2).
However, the upper portions of R.2.2 and the R.2.3 also contain sulphides,
commonly in considerable quantities. In R.2.2, bituminous horizons (2.2 and
3.2) may contain chalcopyrite and more rarely carrollite, except where they are
pyritic. At the base of R.2.3, the sulphides are principally cupriferous; theyoc-
cupy the dolomitic shales, and the laminites enclose the stromatolitic masses
(R.2.3.1.1 and R.2.3.1.2). Within the laminites of the R.2.3.1.2 unit, as well as in
the talc-dolomite unit of R.2.3.1.3, both copper and cobalt sulphides form
lenticular ore bodies rich in carrollite, chalcopyrite and bornite. Finally, sul-
phides can also be found in R.2.3.2 in some dolomitic shales.

Mineralography of the Sulphide Assemblages

Sulphide Zoning

The primary sulphides 4 of the classic copperbelt ore bodies generally display a
vertical zoning which can be readily observed in horizontal planes on the north-
ern flank of the West-Kambove syncline, as well as in adjacent poorly mineral-
ized deposits 5 (Fig. 4). From the stratigraphic bottom to the top, within the lower
mineralized body, a pyrite-dominant zone lies below the chalcopyrite-bornite
zone which is itself overlain by the carrollite-rich zone. The three zones overlap
to a considerable extent. Locally, chalcopyrite and bornite are not developed and
the carrollite zone overlaies directly the pyritic zone.
Within the upper mineralized body, the zoning is similar but inverted. Carrol-
lite is situated near the base and is overlain by bornite-chalcopyrite and then by
pyrite. In poorly mineralized deposits, however, pyrite appears at the base of the
ore body as well, and is overlain by co-existing chalcopyrite, bornite and car-
rolite.
In R.2.3, the laminite unit R.2.3.1.2 is especially characterized by a zonal dis-
tribution of sulphide minerals grading from richer to poorer cupriferous sedi-
ments:
pyrite --+ chalcopyrite --+ bornite --+ chalcopyrite --+ pyrite
These zones form cycles with ticknesses which vary from a few centimetres to
one or more metres. Figure 5 describes a complete cycle with a tickness of about
17 cm, including two partial cycles.

4 "Primary" sulphides are those which precipitated before or during diagenesis.

5 "Poorly mineralized" deposits are those containing less than 20/0 copper.
 KAmbov.·O" .. t .'lanc ~ord
t:l
j;;'
KYA .l0
Pro iii id'Alj •• '{!i
KW . 2368 ::;l
~
KW. 23iA
;:;.
KW. 239 Vl
C
-g:
RZ22 ~
~
~.
R
' p\ 2
~ ~
a ~ ~.
:\ ') I
• I •
I I o·
::;l
I'
I! Alterat ion
R. Z21~~~.-.= .=: sup.rgtn. ~
R { S-
1\ S)-'-'-'-'7 5'
~ '1
(,
S-
I
1\ RZJJIG
I '.
I :
'\
G
';'., />
/
I"!
/ . .
'"
...;t:
e:
I ' Alltratlon
\ \ ~• t
: I suporg'no ;;
\ . Carrolli It
\ \ . ...............
Chalcopyr; It RZI.JIG 3
\\ Pyritt (")

\ AII.ration suporgent 13 1
'> .g
..... "0
13
1
........ '"";'
6 S~al .. doloml!iqutl ........
(")
I' 6
G ' ....
g.
1I Dotomie mQ,~ive
R
-------
6 16 1 ci stromatolites ~
J ~
G I. t:l
l Qminites Alleration Z "0
G ~ suptrgolnt '"
1 ............ -'- ~.
Dolomios stralilith I
1\ I:!I 2
1 R.lLZ 1131 \. loIicrogris cMorlttu.
j I ........ 1:1 Q carc,,'.r. pyrocla5tiqut
v v > R.2.IJ
v .---'-- ChorbonnfUX R
R2111 ~ R.V.! I~~ .. V v
2 v
Un ptU charbonntu. ~II
'J I ·'.sul tu rts
[f I
I V

8
Fig. 4. Profiles of primary sulphide minerals in lower Roan strata at the West Kambove deposit (North limb) and in nearby poorly mineralized deposits
404 J. Cailteux

CHALCOPYRITE DIGEN ITE

KWF. 1166 PYRITE BORNITE PHASES
PRINCIPALES
Ech .427 --- --- --------------- ---- --+------

• PYRITE

--!~~------~----------r---------
•
•
•
• • CHALCOPYRITE

CHAlCOPYRI TE
BORNITE

F:"
- :- .

•
• •
CHALCOPYRITE

•
.•
•
~
BORN I T E

--~------~~----r--------
o
,/
/
• • CHA LCOPYRI TE

/1
,/

•
2
PYRITE

3Cm.

Fig. 5. A complete micro-cycle of zoned primary sulphides in the laminite bed of unit R.2.3.1.2
Diagenetic Sulphide Mineralization Within the Stratiform Copper-Cobalt Deposit 405

Primary Paragenesis

Pyrite forms small grains, or large patches of poecilitic pyrite encircling gangue
minerals. In association with copper- and cobalt-bearing minerals, pyrite is en-
closed by these economic sulphides (Fig. 6b, c and e), but in some cases it con-
tains them (Fig. 6a). The growth of pyrite took place especially around bornite,
porphyroblastic carrollite, etc. (Fig. 6b).
Chalcopyrite and bornite comprise the primary cupriferous mineralization.
Bornite is not consistently present throughout the mineralized layers, but it is
consistently associated with chalcopyrite, in some cases as exsolutions in chalco-
pyrite.
Digenite and chalcocite (or djurleite) are also primary where they are in ex-
solution with bornite (Fig. 6a). They are found in the bornite-rich beds and ac-
count for the strongest copper mineralization at West-Kambove.
Cobalt associated with copper forms carrollite. This mineral is most com-
monly present in isolated poecilitic grains encircling chalcopyrite and bornite, or
in association with chalcopyrite alone. In the R.2.3 unit, it is found mainly in the
the lower talc-rich layers.

A Secondary Diagenetic Sulphide Assemblage

During diagenesis, previously formed sulphides (or those formed during still
earlier diagenetic processes) reacted with the surrounding solutions and were at
least partially transformed to new mineral assemblages.
Bornite has been transformed locally into digenite on the periphery of grains
(an enrichmenUn copper with an accompanying loss in iron), and digenite has
been altered in turn to chalcocite. These associations are not supergene in origin
because they are found as inclusions within later primary sulphides.
In more cobaltiferous strata, the borders of chalcopyrite and bornite grains
are replaced by carrollite. In this reaction, digenite is produced as a result of the
liberation of copper from chalcopyrite or bornite in the presence of cobalt; the
digenite produced rims on and veins within the chalcopyrite and bornite. As for
the iron from chalcopyrite and bornite, it has been removed in solution, or more
rarely, it has been reprecipitated as pyrite within carrollite (Fig. 6b, d, e).
The reactions produced can be outline as follows:
1. For chalcopyrite:
2 CuFeS2+2 Co -+ CuS· CO~3+CU"" +2 Fe""
chalcopyrite carrollite
5 CuFeS2+ 13 Cu -+ 2 CuS· 4 CU2S+5 Fe ....
chalcopyrite digenite
2. For bornite:
3 CUsFeS4 + 6 Co -+ 3 CuS· CO~3+ 12 Cu+3 Fe ....
bornite carrollite
406 J . CaiIteux

Fig. 6. a Pyrite (PY) enclosing the primary assemblage bornite (BN) - chalcocite (CS). R.2.3 .1.2,
Drill hole KWf. 1035, Sample 1161. b The association digenite-chalcocite (DO. 1) enclosing grains of
pyrite (PY. 1); carrollite (CR) formed from a reaction between copper sulfides and introduced
cobalt. A later pyrite (PY. 2) surrounds this early assemblage and is in turn partially enclosed by late
copper sulphides (digenite-chalcocite, DO.2). R.2.2.1.1, Drill hole KWf. 1138, Sample 1511. c De-
position of chalcopyrite (CP) as a rim around a grain of pyrite (PY). R.2.1.1, Drill hole KWf. 1150,
Sample 1311. d Remnant bornite (BN) from a grain initially transformed to carrollite (CR) and
digenite (DO). R.2.2.1.1, Drill hole KWf. 1166, Sample 840. e Chalcopyrite (CP) partially trans-
formed to carrollite (CR) and digenite (DO) . Remnants of digenite remain in the carrollite and the
liberated iron is precipitated as pyrite (PY. 3). R.2.1.1, Drill hole KWf. 1150, Sample 1291
Diagenetic Sulphide Mineralization Within the Stratiform Copper-Cobalt Deposit 407

5CusFeS4+11Cu ~ 4CuS·4Cu§+5Fe""
bornite digenite
In some cases, digenite is transformed into chalcocite, possibly with the aid of
cobalt, and as a result the assemblage appears to be enriched in copper:
CuS· 4 Cu§+2 Co -+ CuS· C0 2S3 +S"" +8 Cu""
digenite carrollite
CuS· 4 Cu§+Cu ~ 5Cu§
digenite chalcocite

The carrollitization of copper sulphides is especially intense in the lower

mineralized body which is the richest in cobalt. In the talc-rich layers of R.2.3,
this process is clearly linked to the crystallization of carrollite porphyroblasts
because certain strata without porphyroblasts show the chalcopyrite to be intact.
In reference to digenite, Dimanche (1974) makes a distinction between
natural digenite, the stability of which is assured only by an iron content of less
than 1070, and a digenite solid-solution metastable at ordinary low temperatures.
Dimanche showed by microprobe analyses that the digenite-type minerals which
he identified at Kipushi and Kamoto were too poor in iron (less than 0.1 wtOJo) to
belong to the group of natural digenites, and that these grains are probably com-
posed of anilite. In contrast, the same mineral, associated with carrollite at
Kamoto, contains a minor amount of cobalt (0.3 wt%).
Carrollite porphyroblasts enclose gangue minerals (chlorite, dolomite and
quartz), primary sulphides (pyrite, chalcopyrite and bornite) and secondary para-
genetic sulphides. It is therefore possible to observe bornite inclusions which
have reacted further toward digenite-chalcocite closer to the borders of por-
phyroblastic grains than in the core regions. In the same way, the outer edge of
chalcopyrite inclusions has been transformed into digenite and carrollite (com-
monly confused with porphyroblastic carrollite).
Locally (especially in the upper mineralized body), chalcopyrite and bornite
porphyroblasts have developed in a similar fashion. For example, chalcopyrite
porphyroblasts may have bornite inclusions intact in the core and bornite inclu-
sions partially or totally transformed into digenite-chalcocite at the periphery.
These chalcopyrite porphyroblasts may also have undergone subsequent
carrollitization. In the same way, bornite porphyroblasts contain grains of
digenite-chalcocite, in some cases ringed with carrollite and still containing
remnants of the initial bornite.
So, there is no doubt that the crystallization of the porphyroblasts started in
the primary stage of mineralization and then continued during the secondary
diagenetic reactions involving primary and early secondary copper sulphides.
A late diagenetic copper mineralization (especially bornite, but in some cases
chalcopyrite) formed overgrowths on carrollitic porphyroblasts and filled the
micro fractures in the carrollite.
408 J. Cailteux

Discussion

Numerous observations on the Kamoto and Kambove deposits sufficiently dem-

onstrate that from the top of the R.1 up to and including the R.2.3 unit, the
sediment-hosted deposit is of a tidal type (Bartholome et al. 1972; Cailteux
1977a, b, 1983). The evidence of evaporite minerals and textures (gypsum, an-
hydrite, evaporite swellings, collapsed salt structures, etc.) in the Mines Group,
of high-salinity fluid inclusions (Pirmolin 1970), and of neoformed minerals
(dolomite, magnesite, etc.) strongly support arguments in favour of the presence
of brines and, therefore, tend to confirm a diagenetic model for metal enrich-
ment. The above-described observations permit a definition of the mecha-
nism.
Firstly, there is no doubt that the primary mineralization of R.2.3 has an
origin similar to that of the classic mineralized bodies of the Shaban copperbelt.
The sulphide paragenesis is identical and the sedimentary environment is similar.
The early emplacement of iron sulphides is well established from observations
on the bituminous layers of R.2.2, at the base of R.2.3, as well as in poorly
mineralized classic lodes where pyrite generally forms grains of the framboidal
type accompanied to a minor extent by inclusions of chalcopyrite.
In the second stage of mineralization, a major chemical reaction occurred be-
tween stagnant interstitial waters present within the sediment and a metal-rich
brine (copper, cobalt) of higher pH and higher Eh circulating within more porous
layers. The reaction transformed early pyrite into copper sulphides and copper-
cobalt sulphides.
At West-Kambove, there is no systematic replacement or destruction of
pyrite; commonly only a simple coating of pyrite by cupriferous sulphides is
observed.
A non-framboidal pyrite identified in the diagenetic succession is closely asso-
ciated with the copper-cobalt mineralization. In some cases, it includes primary
sulphides (including secondary diagenetic products) or constitutes growth rings
around the primary sulphides.
The primary paragenesis indicates that the emplacement of the sulphides
required a certain length of time during which it underwent a secondary reaction
with the brines. Several mineral phases formed in succession and sulphide zones
were developed. The sequence from one phase to the following phase caused the
progressive replacement of previously formed sulphides (or of those in the pro-
cess of being formed). This process led to the "carrollitization" of cupriferous
sulfides as well as the transformation of bornite into digenite where high-grade
copper occurs.
Electron microprobe analyses show that grains of pyrite are zoned with slight-
ly cobaltiferous outer edges (Bartholome et al. 1971, Cailteux 1974). Probably
the origin of cobalt in the pyrite rims is related to the enrichment of cobalt in
copper sulfides.
The transformations are well known for copper. For cobalt, the emplacement
hypothesis is supported by laboratory experiments on copper-cobalt sulphide
ores. Among many types of metallurgical leaching studies (see, for example, Di-
manche et al. 1979), it is especially interesting to note the results obtained from
Diagenetic Sulphide Mineralization Within the Stratiform Copper-Cobalt Deposit 409

experiments using K and Na chlorides in concentrated solutions (F. Dimanche, J.

Frenay, personal communication 1984). The authors note that chalcopyrite and
bornite transform into digenite, which is then transformed rapidly into covellite.
Carrollite is changed slowly into a cobaltiferous digenite, which then gives rise to
a cobaltiferous covellite.
The vertical zoning of sulphides within classical mineralized bodies, (i.e.
Pyrite -+ Sulphides of (Fe - Cu), (Cu - Co), (Fe - Cu) -+ Pyrite,
exhibiting a symmetrical distribution of zones with regard to the R.2.1.3 unit),
disagrees with the concept of an influx of a brine from an underlying conglomer-
ate (e.g. a channel conglomerate) situated at the base of the Mines Group. Such
conglomerates do not exist in the Kambove region. Furthermore, one should find
a zoning similar to that at Kamoto, with chalcocite at the base, followed by
bornite, chalcopyrite and then pyrite at successively higher stratigraphic levels.
However at Kamoto, can one be sure that a supergene process has not altered the
primary profile of the sulphides? For the R.2.3 unit, the lenticular configuration
of ore bodies, the distribution of the ore lenses and the apparent delicacy of the
sulphides within homogeneous sediments dozens of metres thick (laminites) sug-
gest that such a placement mechanism would be improbable. As well, it is not
reasonable to contemplate the downward circulation of a brine through the
R.2.1.3 unit and into the underlying beds. The occurrence of pyrite in the poor of
the mineralized bodies contradicts that concept. On the other hand, a
comparable relationship does not exist between the sulphide mineralization and
stromatolitic beds within the R.2.3 unit.
In conclusion, rather than propose a late inflow of ore solutions along frac-
tures in the sediment, it is more probable, as suggested by Brown (1978), that the
brine migrated through the sediment itself and formed ore bodies during early
diagenesis when the sediment was still sufficiently porous and permeable. Figure
7 illustrates this concept for the metal emplacement in Shaba.
In addition, it is highly probable that changes in pH and Eh conditions at the
interface between the metalliferous brine and the interstitial pore solution, com-
bined with chemical (e.g. salinity) - biological variations and changes in the
metal content of the brine, caused a variable precipitation of sulphides within the
sediment as well as secondary reactions. Annels et al. (1983) confirmed this hy-
pothesis following studies of the Zambian Copperbelt: they concluded that the
increases in cobalt in the sediment and the Cu/Co ratio in carrollite are related to
increases in the quantity of carbonate in the host rock, and also result from mod-
ifications in the activity of sulphur due to variations in the production of biologi-
cally-reduced sulphur.
As to the origin of the Mines Group metals in Shaba, a possible hypothesis
would be that of leaching under oxidizing conditions of a pyroclastic level such as
that found at the top of the R.A.T. Group which underlies the mineralized
bodies. The abnormally high background level of metal in this type of sediment
could have been taken into solution during diagenesis and accumulated in the
basin waters (Cailteux 1983). It is then possible that this water, already enriched
in ionic metals, could have been trapped in the sabkha-zone and formed metallif-
erous brines.
 SABKHA e
o

1";\ 1";\
f.'\ R.2 .I.J 1";\ ~
r.\ 1";\ 1";\%ZI:rr.\ 1";\

POUR LA
COUCHE MINERALISEE SUPERIEURE

SABKHA CONTINENT
Zona d" prtllirc'ratt"on tillS mali.r." o"9Qnlf~.4
el J. rCt/uclitM h"cl(J~r".nnc

POUR LA
COUCHE MINERALISEE INFERIEURE

Silumures rleh.~ en mill"x Limi/~ du uux in/.rslillOlI.. charScu en H2S,CHI/,C02.S()I/.~k.

x,..X)C"IC
~ M,sr.lion cI~ uumu,.es ,i Irlvers 1• .$ s.~,Jlmenls x)tx)(,t Incru$I... I /Dns cJ.'."lpora.s ~
(")

~
Fig. 7. Genetic model for the emplacement of stratiform metal deposits in Shaba ~
Diagenetic Sulphide Mineralization Within the Stratiform Copper-Cobalt Deposit 411

Acknowledgements. The management of Gecamines has authorized the publication of this work, and
I wish to express my appreciation for this permission. I would also like to acknowledge Prof. F.
Dimanche of the Universite de Liege for the many useful discussions that we have had on this topic.

References

Annels AE, Vaughan DJ, Craig JR (1983) Conditions of ore mineral formation in certain Zambian
Copperbelt deposits with special reference to the role of cobalt. Min Dep 18:71 - 88
Bartholome P (1962) Les minerais cupro-cobaltiferes de Kamoto, Katanga-Ouest (Shaba). I. Petro-
graphie - II. Paragenese. St Univ Lovanium, Fac Sci 14, 16. Leopoldville (Kinshasa)
Bartholome P (1963) Sur la zonalite dans les gisements du Copperbelt de l'Afrique Centrale. Sym-
posium. Problems of postmagmatic ore deposition. V. 1. Prague, 317 - 321
Bartholome P (1969) Kamoto (Katanga) et White Pine (Michigan), deux gisements stratiformes de
cuivre. Acad Roy Sc d'Outre-Mer Bull Seances (Bruxelles) 397 - 410
Bartholome P, Katekesha F, Lopez-Ruiz J (1971) Cobalt zoning in microscopic pyrite from Kamoto,
Republic of the Congo. Min Dep 6:167 -176
Bartholome P, Evrard P, Katekesha F, Lopez-Ruiz 1, Ngongo M (1972) Diagenetic ore forming pro-
cesses at Kamoto, Katanga, Rep. of the Congo. In: Amstutz GC, Bernard AJ (eds) Ores in sedi-
ments. Springer, Berlin Heidelberg New York, pp 21-41
Bartholome P (1974) On the diagenetic formation of ores in sedimentary beds, with special reference
to Kamoto, Shaba, Zaire. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes.
Liege, pp 203 - 213
Brown AC (1978) Stratiform copper deposits - evidence for their postsedimentary origin. Min Sci
Eng Vol 10; 3:172-181
Cahen L (1954) Geologie du Congo Beige. Vaillant-Carmanne, Liege
Cahen L (1974) Geological background strata to the copper-bearing strata of Southern Shaba (Zaire).
In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Liege, pp 57 -77
Cailteux J (1974) Les sulfures du gisement cuprifere stratiforme de Musoshi, Shaba, Zaire. In:
Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Liege, pp 267 - 276
Cailteux J (1977) Particularites stratigraphiques et petrographiques du faisceau inferieur du Groupe
des Mines au centre de I' Arc cuprifere Shabien. Ann Soc Geol Belg T 100:55 -71
Cailteux J (1977) La succession stratigraphique du C.M.N. (ou R.2.3.) au centre de la sous-province
cuprifere shabienne. Ann Soc Geol Belg T 100:73 - 85
Cailteux J (1983) Le "Roan" shabien dans la region de Kambove (Zaire). Etude sedimentologique et
metalloglmique. Thesis Univ Liege. Belgium
Demesmaeker G, Francois A, Oosterbosch R (1962) La tectonique des gisements cupriferes strati-
formes du Katanga. In: Lombard J, Nicolini P (eds) Gisements stratiformes de cuivre en Afrique.
2eme partie. Lusaka. Assoc Serv Geol Afr Paris 47 -115
Dimanche F (1974) Paragenese des sulfures de cuivre dans les gisements du Shaba (Zaire). I. Kipushi,
II. Kamoto. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Liege, pp
185 -201
Dimanche F, Frenay J, Herman JC (1979) Lixiviation ammoniacale d'un minerai cupro-cobaltifere
sulfure. Ann Soc Geol Belg T 102:199-211
Francois A (1973) L'extremite occidentale de I' Arc Cuprifere shabien. Etude geologique. Gecamines.
Likasi, Shaba, Zaire
Francois A (1974) Stratigraphie, tectonique et mineralisations dans I' Arc cuprifere du Shaba (Rep. du
Zaire). In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Liege, pp 79 -101
Lefebvre 11 (1974) Mineralisations cupro-cobaltiferes associees aux horizons pyroclastiques situes
dans Ie faisceau superieur de la Serie de Roan a Shituru Shaba, Zaire. In: Bartholome P (ed) Gise-
ments stratiformes et provinces cupriferes. Liege, pp 103 -122
Oosterbosch R (1962) Les mineralisations dans Ie systeme de Roan au Katanga. In: Lombard J, Nico-
lini P (eds) Gisements strafiformes de cuivre en Afrique. lere partie. Assoc Serv Geol Afr Paris.
71-136
Pirmolin J (1970) Inclusions fluides dans la dolomite du gisement stratiforme de Kamoto (Katanga
Occidental). Ann Soc Geol Belg T 93:193-202
Geochemical Aspects of Stratiform and Red-Bed Copper
Deposits in the Catskill Formation (Pennsylvania, USA)
and Redstone Area (Canada). Sequence of Mineralization
in Sediment-Hosted Copper Deposits (Part 3)
A. W. ROSE, A. T. SMITH, R. L. LUSTWERK, H. OHMOTO, and L. D. Hoyt

Abstract

Noneconomic red-bed copper occurrences in the Devonian Catskill Formation

are localized by concentrations of reducing plant fragments, commonly in the
basal parts of fining-upward fluvial (or tidal?) cycles. The occurrences are within
and near thick zones of red mudstone to fine sandstone interpreted to have been
deposited in relatively inactive parts of an extensive alluvial plain in an arid envi-
ronment. Regionally, the occurrences are most abundant adjacent to areas of
major sediment input, and in more detail are controlled in part by sandstone beds
which were more mature and probably more permeable than the rest of the
formation. The copper minerals were emplaced after earliest diagenesis, but
before induration and formation of quartz overgrowths. Thin zones of transgres-
sive marine sandstone are present near most occurrences. Copper has been
redistributed within the Catskill, being greatly depleted from nearly all red
(oxidized) sediments, but showing highly variable values in green and gray
(reduced) sediments. Mobilization of copper from the red beds into chloride-
bearing pore fluids of marine or evaporative origin is proposed, followed by flow
of these fluids through the more permeable sandstones until organic reductants
were encountered which precipitated the copper by bacterial sulfate reduction.
The flow was driven by compaction, sea-level changes, and gravity flow down
the alluvial plain. Red beds near the Coates Lake stratiform copper deposit in the
Redstone area show similar depletion of copper.

Introduction

The source of copper and other metals in red-bed copper and stratiform copper
deposits, and the geological-geochemical controls and processes forming these
ores, are topics of continuing discussion (Gustafson and Williams 1981). The
intent of this paper is to summarize evidence on these problems from geochemi-
cal studies of red-bed copper-uranium occurrences in the Devonian of Pennsyl-
vania and the Precambrian Redstone stratiform copper deposit in the Northwest

1 Department of Geosciences, The Pennsylvania State University, University Park, PA 16802, USA

Geology and Metallogeny of Copper Deposits

(ed. by G.H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
 Geochemical Aspects of Stratiform and Red-Bed Copper Deposits in the Catskill Formation 413

Territories, Canada. Though neither of these areas currently produces copper,

the Redstone deposit would be economic in a more accessible area, and both
areas serve to illustrate the processes and relations in producing districts.

Red-Bed Copper-Uranium Occurrences in the Upper Devonian Catskill

Formation, Pennsylvania

The Catskill Formation is a complex clastic wedge of dominantly fluvial red and
gray sediments that prograded to the west during erosion of the Acadian High-
lands along the east coast of the United States (Fig. 1). The formation thins
rapidly westward, from about 2400 m near Lehighton at the eastern limit of the
exposures to less than 800 m in central and north-central Pennsylvania. The
Catskill overlies and to the west thins and interfingers into non-red marine shales
and turbidites of Middle and Upper Devonian age. The Paleozoic sediments were
extensively folded in late Paleozoic and now crop out along narrow belts in much
of the region.

- - - -- - - ----..,
"~*i
A
'4
'

Cotsklll Format ion

P recombrlan basement
x Cu - U in Catsk i II
$ Sandstone- type U
o 10 20 30 40 M il es • Pegmatite or Granite
I I ! I I ' I 'I I
I
.. Uranium w i th magnet i te ore
o 20 40 60 Kilometers
• Uron l um ve i ns

Fig. 1. Distribution of Catskill Formation, red-bed copper-uranium occurrences, sandstone-type

uranium occurrences, other uranium occurrences, and geologic provinces in Pennsylvania and ad-
jacent states
 414 A. W. Rose et al.

The Catskill Formation has been divided into four magnafacies utilizing re-
peating patterns of individual facies (Smith and Rose 1985). These magnafacies
show an overall progradational pattern (Fig. 2). Over most of the area, the lower
part of the Catskill (Magnafacies A) is composed of red shale and fine-grained
sandstone containing plant fragments and root casts, interlayered with gray to
green fine-grained sandstones and shales occasionally containing marine fossils.
This magnafacies (A) is interpreted as interbedded nonmarine and marine sedi-
ments deposited in an intertidal to supratidal environment.
The middle part of the Catskill (Magnafacies B) is typically composed of
fine, red sandstone to mudstone in fining-upward cycles, interpreted as the de-
posits of relatively small meandering streams on an alluvial plain. The basal
zones of some cycles are green and locally contain plant fragments. Caliche in
mudstone horizons near the top of some cycles suggests an arid climate. Oc-
. casional thin tongues of marine sandstone within Magnafacies B are recognized
by green color, bi-directional ripples and cross-beds, flaser bedding, planar
bedding and occasional marine fossils. These tongues apparently reflect trans-
gressions of marine (or at least tidal) conditions across the alluvial plain and ap-
pear to be of considerable importance in forming the deposits, as discussed later.
The upper Catskill contains two magnafacies. Magnafacies C is composed of
extensively cross-bedded, gray-green sandstones and conglomeratic sandstones
deposited by northwestward-flowing braided streams, and is limited to three
areas which are interpreted as loci of sediment input from a source area that was
uplifted contemporaneously (Fig. 3). Magnafacies D overlies Magnafacies B or
C, and is composed of thick, fining-upward cycles with a much higher propor-
tion of sand than Magnafacies B. The siltstone and mudstone in the upper part of
these cycles are red; the lower sandstones are commonly gray or green. This
magnafacies is interpreted as the deposit of relatively large meandering streams.

IZOl Duncannon
~ member
IZ7.1 Sherman Creek
IZ2:..J member
~ Irish Volley
IZ!.:.J member

Fig. 2. Inferred depositional environments and relations of sedimentary units in the Catskill Forma-
tion and underlying rocks. IV is equivalent to Magnafacies A of this text, SC to Magnafacies B, and
D to Magnafacies D (from Smith and Rose 1985, modified after Rahmanian 1979)
Geochemical Aspects of Stratiform and Red-Bed Copper Deposits in the Catskill Formation 415

~HE/~~IVIV
NY
O. 50 100km PA I ACTIVE

.
~----'.
...... COASTAL PLAIN
SCALE • @ • •
• • • Seronlon 0
, I Monlltollo
EXPLANATION '(

~ (0
~c:
it/Cit
--
@ clusler of four or more INACTIVE ".
Cu-U occurrences COASTAL PLAIN

•
Cu-U occurrence
U-rich occurrence ,/'
,,.
II •

Allenlown
.
"
INACTIVE :
COASTAL PLAIN •
x • ,,
,
L-------------~/~'7M~D~-------------rDEL

Hg. 3. Location of red-bed copper-uranium occurrences and sandstone uranium occurrences relative
to sediment input areas (Magnafacies C) in the Catskill Formation (after Smith and Rose 1985)

The red-bed copper-uranium occurrences are lens-shaped to tabular and typi-

cally occupy zones of plant fragments either in the basal sandy to conglomeratic
portions of fining-upward cycles, or in thin gray siltstone units. The mineralized
zones range in size from 0.1 to 1 m thick and extend 3 to 10 m along bedding.
Chalcocite and digenite furnish grades of 0.05 to 1070 copper and are accom-
panied by small amounts of uranium (20 - 1000 ppm), only rarely as recognizable
uraninite, along with enrichments in arsenic, lead, and vanadium. The copper
and uranium are localized by concentrations of plant fragments, which apparent-
ly acted as reducing agents.
Several lines of evidence indicate a low temperature of copper deposition.
Sulfur isotopes of sulfides range from - 34 to -18%0 (CDT), averaging - 27%0,
suggestive of biogenic sulfate reduction at temperatures less than 100°C. The
grain size of copper minerals and the gangue minerals is very fine, typical of sedi-
ments rather than hydrothermal processes. Bornite from the occurrences exsolves
chalcopyrite upon heating above 75°C, indicating formation as a meta-stable
phase at low temperatures (McCauley 1961). The age relations of the sulfides
indicate a diagenetic origin, as discussed later.
The red-bed copper-uranium occurrences are mainly in Magnafacies B, with
some in Magnfacies A and D. They are lacking in Magnafacies C, but tend to be
concentrated adjacent to the input areas marked by the distribution of Magna-
facies C.
An additional local control on the distribution of Cu minerals is provided by
relatively coarse-grained bodies of sandstone which probably had high perme-
ability prior to late Paleozoic deformation. Most sandstones of the Catskill are
"dirty", low-rank graywackes with a high proportion of phyllite grains and fine-
grained matrix, but the sandstones near ore bodies are significantly different.
"Mineralized" sandstones collected within or immediately adjacent to mineral-
ized zones were compared with sandstones of similar grain size collected nearby,
 416 A. W. Rose et al.

but outside the mineralized zones. The mineralized sandstones characteristically

have 70 to 750/0 framework grains (mainly quartz) and only 20 to 30% ductile
grains (rock fragments, micas, matrix) compared to 55 to 65% framework grains
and 30 to 45% ductile grains in the "nonmineralized" sandstones (Smith 1983).
Some of these mineralized sandstones are marine as discussed below. The higher
permeability of the mineralized sandstones prior to deep burial and deformation
would have allowed preferential flow through these zones.
Of 30 localities examined in detail, 23 have exposures of sandstones inter-
preted as "marine bars" or "barrier sandstones" (Smith 1980, Smith and Rose
1985) within 10m above the Cu occurrence. At many of the remaining seven
occurrences, exposures are poor and such sediments could be concealed. Also,
most of these seven occurrences are near other occurrences that do exhibit marine
sandstone facies. This correlation with thin tongues of marine sediment appears
highly significant in view of the effect of Cl-bearing waters on copper solubility.
The solubility of copper as Cu 2 + or complexes of common natural anions
with Cu 2 + is very low at pH values greater than 6, whereas very few stream or
groundwaters have pH values less than 6 (Rose 1976). In addition, Fe oxides are
very effective adsorbents of copper at pH values above about 6 (Benjamin and
Leckie 1981). Thus, in a red-bed depositional environment, one would not expect
much copper mobility.
However, in the presence of chloride under intermediate oxidation condi-
tions, strong complexes of Cu + with CI - are formed, and the solubility of cop-
per is greatly increased (Fig. 4), so that high solubilities extend to pH 8 or higher.

a b
.8 .8

.4
Eh Eh
v. v.

3 5 7 9 11 3 5 7 9 11
pH pH
Fig. 48, b. Solubility of copper at 25°C as a function of pH, Eh, and CI- , showing increase at pH
6- 8 in the presence of CI. 8 No chloride, ES = 10- 4 m; b0.5 m CI, ES = 10- 4 m, cov = covellite;
an = anilite; dj = djurleite (after Rose 1976)
Geochemical Aspects of Stratiform and Red-Bed Copper Deposits in the Catskill Formation 417

The ingress of marine waters during the occasional transgressions provides a

convenient source of Cl-rich solutions. If small amounts of organic matter or
Fe 2 +-bearing clay or chlorite were present in the initial sediment, O 2 would be
consumed and the pore fluid would reach an appropriate oxidation state. Waters
of compaction released from underlying marine sediments, or evaporation of
fresh waters in an arid environment are other possible sources of Cl. Uranium is
also relatively soluble in solutions of intermediate oxidation state, especially if
some carbonate is present in solution.
The abundance of copper in the Catskill and related rocks provides more
direct evidence for the source of metals. So far as is known, the source area for
Catskill detritus was similar to the source for the underlying Middle Devonian
marine sediments, the main difference being the environment of deposition and
the resulting color (Sevon et al. 1978, Willard 1939). Essentially all sediments in
this sequence are low-rank graywackes and equivalent shales and mudstones.
Figure 5 shows copper contents of a set of Catskill and pre-Catskill rocks.
The Catskill samples were collected in a previous study of uranium (Pirc 1979,
Pirc and Rose 1981, Rose and Wright 1980), and were intended to be representa-
tive of the various colors and grain sizes of Catskill sediments. The pre-Catskill
rocks were collected for this study. In the reducing marine pre-Catskill rocks,
copper contents are mostly in the range 15 - 30 ppm and show a possible
correlation with Fe content. Changes with grain size appear to exist, but are
relatively minor. These samples are believed to represent the copper content of
detritus entering the Catskill sedimentary basin. In contrast, in the Catskill, red
(oxidized) sediments are generally much lower in copper, mostly 0 - 5 ppm, and
gray to green (reduced) sediments vary widely, with a few values over 400 ppm.

Pre-Catskill

15 Cu
DRed
• Green or gray

M.Catskili

o
z

U.Catskill
Fig. S. Abundance of total copper (HF-
HCl04 digestion) in red and green-gray
sandstones and shales of the Catskill and
pre-Catskill Formations near Lehighton,
_ " ' - '_ _L - - - '_ _. L - _....... •• , , Pottsville, and Harrisburg, P A
418 A. W. Rose et al.

There is no obivous correlation with grain size or with major elements in the
samples; high values are found in gray-green rocks ranging from mudstone to
medium-grained sandstone, and low values in red rocks ranging from mudstone
to fine sandstone.
These data indicate major redistribution of copper in the Catskill during de-
position and early to middle diagenesis. Much copper appears to have been lost
from the red beds. Uranium behaves somewhat similarly, being enriched in some
reducing horizons, though not to the extent shown by copper. The rocks are now
highly indurated and impermeable so the redistribution must have occurred
before appreciable burial and compaction. In the exposures near Lehighton from
which these samples were collected, no red-bed copper occurrences were previ-
ously recognized, but copper at levels of 100 to 400 ppm is present in numerous
locations, even though not visually evident.
Loss of 20 ppm copper from a 30-m-thick section of sediment (25 ppm initial
to 5 ppm final) could furnish copper for a layer 0.3 m thick with 0.20/0 copper,
which is a common concentration in the red-bed copper occurrences (Fig. 6).
This calculation assumes that all transport is normal to bedding. Lateral migra-
tion from a large volume of sediment into a similar small volume could create
higher grades or thicker mineralized zones. Thus, ample copper is available to
form the red-bed copper occurrences by depletion of copper from nearby
sediment.
The time of copper mobilization is also indicated by the age relations of
minerals in the red-bed copper occurrences. In the earliest di-agenetic stage,
quartz and other detrital grains are coated by hematite and locally by fine clay
that apparently infiltrated into the stream bed (Walker et al. 1978, Loucks et al.
1977). In reducing environments manganiferous calcite cement has formed, and
can be seen to surround the clay coats in a few instances. A middle diagenetic
stage is represented by chlorite as an alteration of rock fragments and biotite and
as radiating and fibrous crystals, apparently formed in relatively large pore
spaces before compaction, but later than calcite because it is not found in calcite-
cemented zones. Copper sulfides accompany some chlorite and also appear to
have filled relatively large pores prior to compaction, or have replaced organic
matter. A late diagenetic stage is characterized by quartz overgrowths which
cover clay coats and are not found with copper minerals. In sediments of the Frio

--f --- ---f --- ----i --

-----------
---------
------
-
--- ------------
--- - -- - - -- -~-
--------------------
------ -- - - - --- - --- Fig. 6. Model for formation of red-
-----j----- --1-=--= -=-- -------~- --- ----
==1.:- _- _- _-_-_
bed copper-uranium occurrence in
- _-_- -_ -=- --=- -_ - =- -= -: _: the Catskill Formation. Arrows in-
dicate flow of copper-bearing pore
fluid out of red shale and along bedd-
---------------------- ing in more permeable sandstone,
D sondslone f3 red shale - - - flow polh and localization of copper in zone of
plant trash with dimensions approx.
~ Co - U mineralizalion associoled wilh planl Irash
0.5 m thick x5 mx5 m
Geochemical Aspects of Stratiform and Red-Bed Copper Deposits in the Catskill Formation 419

Formation on the Gulf Coast of the United States, quartz overgrowths are first
observed at depths of 1 to 2.5 km (Loucks et al. 1977) and are assumed to be
formed at a similar stage in the Catskill. A final stage is hematite along fractures,
perhaps related to recent weathering.
Based on the above age relations, copper was emplaced after earliest diagene-
sis (after hematite, clay coats, and calcite cement), but probably before quartz
overgrowths and before compaction and induration of the sediments.
Flow of pore fluids during diagenesis could be driven by compaction, gravity-
driven flow of groundwater down the alluvial fan, or small changes in sea level as
demonstrated by the thin marine tongues. The flow caused by these processes
would be concentrated in the network of relatively coarse-grained sands de-
posited in channels of the larger streams. These permeable sands may be more
abundant near the sediment-input areas.
In summary, the following factors appear to control the formation of the red-
bed copper occurrences in the Catskill Formation:
1. Plant fragments act as reductants to localize sulfide precipitation.
2. Chloride-bearing solutions tend to mobilize copper; these solutions might be
seawater emplaced during small marine transgressions, pore fluids from
underlying compacting marine sediments or evaporated fresh water in the arid
coastal plain.
3. The process of forming red beds appears to have liberated copper, perhaps by
oxidation of the iron-bearing silicates initially hosting copper, and possibly af-
fected by the behavior of the resulting Fe-oxides.
4. Permeability of sediments; the coarser, better-sorted sands deposited in chan-
nel bottoms of the larger streams on the alluvial plain, and the thin tongues of
sand in the marine transgressions, would be locations of maximum flow.
5. Copper content of sediments; the copper content of the relatively "dirty"
Catskill sediment probably was slightly higher than many other sedimentary
sections. Also, data in the literature suggests that copper is concentrated in the
finer sediments with higher Fe (WedepohI1964).
6. Hydraulic gradients causing flow of pore fluid, as dicussed above.

Redstone Area

The stratiform to lenticular copper occurrences in late Precambrian sediments of

the Redstone area in the MacKenzie Mountains, Northwest Territories, Canada
are described in a companion paper by Brown and Chartrand, this Vol., and by
Ruelle (1982) and Lustwerk and Rose (1983). The hosts of ore are one or more
beds of reducing marine algal carbonate of sabkha and shallow marine origin.
These units are underlain by 1000 m or more of Redstone River Formation,
composed of red siltstone deposited on an evaporative, alluvial plain and/or
continental sabkha, as indicated by common gypsum-anhydrite as cement and
locally as thin beds of gypsum-anhydrite in the siltstone. The pore fluids in these
sediments clearly were saline with elevated chloride content during part of their
diagenetic history.
420 A. W. Rose et al.

Fig. 7. Abundance of total copper in

20 Transition Zone-Coates Lalle
siltstone and sandstone from the
• Red Redstone River and Transition Zone
N
o Gray or green near the Coates Lake stratiform cop-
per deposit, Redstone Belt,
MacKenzie Mts., NWT. The ore oc-
Redstone R. Fm.-Coates Lake curs within the lower Transition
Zone and is underlain by 1000+ m of
Redstone River Formation. The
Thundercloud area, 40 km to the
south, contains a thinner section of
Thundercloud Range the same rocks in exposures along the
margin of the basin
.0
20 40 =60 80' 1!l0>100
Cu(ppm)

Figure 7 shows analyses for copper in numerous samples of siltstone and

sandstone from the Redstone River Formation and the Transition Zone contain-
ing the ore. Over half of all samples and about 750/0 of the red samples contain
<2 ppm copper. Some gray to green samples contain a wide range of higher
values. Most other transition elements (zinc, nickel) show fair to good correla-
tion with iron, and it appears that copper in detritus entering the basin probably
shared this correlation. Although some sandstones with low iron content may
have originally contained only a few ppm copper, the average siltstone with 3%
to 8% iron probably contained 15 to 50 ppm, based on worldwide averages and
estimates of the type of detritus entering the basin. The pattern is thus similar to
that found for the Catskill; namely, that copper is extensively redistributed
during diagenesis of sediments near "sedimentary" copper deposits; most of
these rocks, especially red beds, are greatly depleted in copper, and appear to
have been the source of copper in the ore. This type of copper depletion may be a
guide to favorable areas for sedimentary copper deposits.

Acknowledgments. The research described above has been supported by a grant from the National
Science Foundation and by the Mineral Conservation Section of the Pennsylvania State University.
We are indebted to Prof. A. C. Brown, Mr. Frank Chartrand, and Dr. John Ruelle for discussion
and assistance with the Redstone studies and to Prof. E. G. Williams and Dr. R. L. Slingerland for
ideas and discussion on the Catskill studies.

References

Benjamin MM, Leckie JO (1981) Multiple-site adsorption of Cd, Cu, Zn and Pb on amorphous iron
oxyhydroxide. J Colloid Interf Sci 79:209 - 221
Gustafson LB, Williams N (1981) Sediment-hosted stratiform deposits of copper, lead, and zinc. In:
Skinner BJ (ed) Economic geology 75th anniversary, vol Econ Geol Publ Co, pp 139 -178
Loucks RG, Bebout DG, Galloway WE (1977) Relationships of porosity formation and preservation
to sandstone consolidation history - Gulf Coast Lower Tertiary Frio Formation. Gulf Coast As-
soc Geol Soc 27:109 -120
Lustwerk RL, Rose AW (1983) Source and segregation of transition metals during diagenetic forma-
tion of the Redstone stratiform copper deposit, MacKenzie Mts., NWT, Canada. Geol Soc Am
(Abstracts with Programs) 15:632
Geochemical Aspects of Stratiform and Red-Bed Copper Deposits in the Catskill Formation 421

McCauley JF (1961) Uranium in Pennsylvania. Penn Geol Sur (4th) Bull M43, 71 pp
Pirc S (1979) Uranium and other elements in the Catskill Formation of east-central Pennsylvania.
Ph D Thesis, Penn State Univ 300 pp (unpublished)
Pirc S, Rose A (1981) Uranium anomalies in paleoaquifers near sandstone-type uranium deposits in
the Devonian Catskill Formation of Pennsylvania. J Geochem Explor 15:219-231
Rahmanian VDK (1979) Stratigraphy and sedimentology of the Upper Devonian Catskill and upper-
most Trimmers Rock Formations in central Pennsylvania. Ph D thesis, Penn State Univ, 329 pp
(unpublished)
Rose AW (1976) The effect of cuprous chloride complexes in the origin of red-bed copper and related
deposits. Econ GeoI71:1036-1048
Rose AW, Wright RJ (1980) Geochemical exploration models for sedimentary uranium deposits. J
Geochem Explor 13:153 - 179
Ruelle JC (1982) Depositional environments and genesis of stratiform copper deposits of the Red-
stone Copper Belt, Mackenzie Mountains, NWT. Geol Assoc Can Spec Pap 25:701-737
Sevon WD, Rose AW, Smith RC, Hoff DT (1978) Uranium in Carbon, Lycoming, Sullivan and
Columbia Counties, Pennsylvania. Guidebook, 43rd Field Conf Penn Geol, Harrisburg, PP, pp
7-22
Smith AT (1980) Stratigraphic and sedimentologic controls for copper and uranium in red-beds of the
Upper Devonian Catskill Formation in Pennsylvania. MS Thesis, Penn State Univ, 216 pp (un-
published)
Smith AT (1983) Geology of the red-bed Cu-U occurrences in the Upper Devonian Catskill Forma-
tion, Pennsylvania. Ph D Thesis, Penn State Univ, 240 pp (unpublished)
Smith AT, Rose AW (1985) Relation of red-bed copper-uranium occurrences to the regional
sedimentology of the Catskill Formation in Pennsylvania. In: Woodrow DL, Sevon WD (eds) Geol
Soc Am Spec Pap 201: 183 -197
Walker TR, Waugh B, Crone AJ (1978) Diagenesis in first-cycle desert alluvium of Cenozoic age,
southwestern U.S. and northwestern Mexico. Geol Soc Am Bull 89:19-32
Wedepohl KD (1964) Untersuchungen am Kupferschiefer in Nordwest-Deutschland. Oeochim
Cosmochim Acta 28:305 - 364
Willard B (1939) The Devonian of Pennsylvania. Penns Geol Sur (4th) Rep 019, 4!U pp
Stratabound Copper Deposits in East South-Central
Alaska: Their Characteristics and Origin
M. K. SOOD, R. J. WAGNER, and H. D. MARKAZIl

Abstract

Six stratabound Kennecott-type copper deposits are being studied to evaluate the
characteristics of the ore-host rock system. Three (Mountain Grill, Radovan, and
Clear-Porcupine) are in the Middle to Upper Triassic Nikolai Greenstone - a
thick sequence of subaerial amygdaloidal tholeiite flows - and the other three
(Binocular, Nelson, and Peavine) are in the Upper Triassic Chitistone Limestone,
a platformal limestone and dolomite unit.
X-ray diffraction, petrographic and chemical studies have led to the identifi-
cation of ten ore minerals in the deposits, the most abundant being digenite,
chalcocite, djurleite, covellite, bornite, and chalcopyrite. Minor arsenic and
antimony sulfide phases are also present. Mineral deposits in greenstones have
both Cu and Cu - Fe sulfides in contrast to the essentially Cu-sulfides in lime-
stones.
The mineralization is proposed to have formed in two stages. Hydration-de-
hydration reactions associated with alteration-metamorphism of the Nikolai
Greenstone liberated copper from the metal oxides and the mafic minerals. Cop-
per was carried by the circulating fluids along the existing fracture-fluid pathway
system, reacting and preCipitating ore minerals in them. The weak and widely dis-
tributed mineralization in the belt is probably of this type. Greenstone alteration
was probably coincidental with Cretaceous accretionary orogeny as shown by K-
Ar systematics (Silberman et al. 1981).
Cenozoic deformation induced a steep fracture system that enhanced the
porosity and permeability of the rocks, thus focusing the meteoric water circula-
tory system for effective and efficient solution and transport of metals. Activity
in such a system at this stage was probably driven by a thermal charge from the
Tertiary intrusives. Lack of extensive alteration of the Chitistone Limestone and
of recrystallization at the ore vein-limestone contacts suggests that the ore
solutions were at relatively low temperatures. The mineral associations and
textures suggest that the upper temperature that prevailed in the Triassic ore-host
rock system was about 200 °C.

1 Department of Earth Sciences, Northeastern Illinois University, Chicago, IL 60625, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 423

Introduction

The Triassic age Nikolai Greenstone and Chitistone Limestone belt lying in the
central part of the McCarthy Quadrangle, Alaska, (141 0 -144 0 long and
61 0 - 620 lat), is a host to stratabound copper mineralization. Of particular in-
terest have been the limestone-hosted copper deposits at Kennecott considered to
be of low-temperature hydrothermal origin (Bateman and McLaughlin 1920,
Jensen and Bateman 1981). These are large, high-grade deposits characterized by
simple mineralogy, limited host-rock alteration, and sparsely distributed gangue
minerals.
There are a number of other copper occurrences in this Triassic belt that have
affinities with those at Kennecott, but have not been systematically studied. This
paper describes the geologic and mineralogical characteristics of six Kennecott-
type copper deposits both from greenstone (Mountain Grill, Radovan, and
Clear-Porcupine) and limestone (Binocular, Nelson and Peavine), with special
emphasis on the alteration-metamorphism of the Nikolai Greenstone, the
character of the ores, thermal regimes in the ore-host rock system, and the
structural focus of mineralization. The objective is to develop a model for the
genetic evolution of the widely scattered copper mineralization in the McCarthy
Quadrangle and to study its relation to the accretionary tectonics in east-south-
central Alaska.

Location

The McCarthy Quadrangle is located about 260 miles east of Anchorage and
extends to the Canadian border in east south-central Alaska (Fig. 1a). The
generalized geology of the northwest trending Nikolai Greenstone and Chitistone
Limestone belt and the locations of the copper deposits are shown in Fig. 1b.

Geologic Setting

Lithounits in the McCarthy Quadrangle constitute Upper Paleozoic to Cenozoic

age stratigraphy (Fig. 2).
The Upper Paleozoic andesitic island-arc sequence and argillites and lime-
stones of the Skolai Group are unconformably overlain by Nikolai Greenstone,
Chitistone Limestone, basinal Nizina Limestone, and the Triassic-Jurassic
McCarthy Formation. Jones et al. (1977) termed this cohesive lithostratigraphic
setting the allochthonous Wrangellia terrane that comprises the major part of the
McCarthy Quadrangle. The superjacent Cenozoic units comprise largely silicic
and felsic shallow-level intrusives, and andesitic to rhyolitic rocks of the
Wrangell Lava (Fig. 2).
Southern Alaska - an example of collisional tectonics - is composed of
many disparate terranes generally becoming progressively younger to the south
424 M.K. Soodetal.

S. ::,,____I:gOO~==~ISO ~,,(s
O-==-"",:,~OC::===~SO

iio::=-=-===?~2_ _.i:OO
SO ~=:::::!.ISO "lO~( nos

Fig. la. Location of the McCarthy Quadrangle in relation to the major structural features in southern
Alaska (modified from MacKevett et al. 1977)

Fig. lb. Scheme of the 15-min-quadrangles, the location of the mineral prospects of the present
study, and the distribution of the Nikolai Greenstone and Chitistone Limestone belt in the McCarthy
Quadrangle (based on MacKevett 1978, and MacKevett et aI. 1977)

F.:l
l2J Tertiary intrusives c. Mineral prospects

• Jurassic Chitina 1. Radovan

_ Valley bathallth 2. Nelson
3. Mountain Grill
~ Chitistane Limes tone L. Binocular
Triassic 5. Pe a vine
fiITj Nikoloi Greenstane 6. Clear-Porcupine
K. Kennecott
... .. .. Fault 81, . 15' quadra ngle
- - Geologic boundary
 62.0~.L ·00' "3'00' en
I DBI l,O' I D61 3D'
D5 1 .
DL }'" . "2'00'
0' 30'
' D21 ""00'
D7 ' .. 1 '. (0" D31 Dll ~2"00'
o/..
ig
::s
~'~ P-
(j
. . \. "'-1'(' I .g
'8
C6 C5 Cl.
i
":" .......... .,.....""
~ L~,
"~:~~~"';il~<~~~SO:.GC2t:==~~ ~.
en
s·
t'!1
«;;~[~~S~, C1 ~
'en"
a
9"
(j
.•....1 ::~:~!t;:~~~::.:.:~.;:;:::..··· '::\.
g
...
83 82 . '." . ~::::81) 30' ~
~
~
;><'
!'?
;i
...g .
Q
~
f1
AB A7 A3 A2 Al ~
:;;.
~.
(')
:::::':::·::'::::'::'W":'.':::::::::. en
·e·····················....... §
P-
Order IT r ...... O
Of)ge s f: ........ ::l.
°Ult ..... u
og.
···0···.1..... .
~"00 ' 1 . • .•... .){ & , I I ~,·oo·
lLL-OO 30' ltC,l.OO· 'l0' 1'2°00' 30' 11.1-00'
~
Fig.lb Scale
426 M. K. Sood et al.

...
c.
Y
...
..
,..
..
'f
«'
oJ
,...J
)

Wro~gell ., "'.., )
r

..,
.. "

~ : lava .1 ('., r.,' } O"t."",

'" .... .l "" ,. < ,. (' .., (' ~ Mainly
Super jacent
to "' .. "'oJ 4. .. ~ "\I /dacite
lithologies Mainly
grano-

Late triassic

Middle and (or)

La te triassic
Wrangellia
terrane

Middle triassic

- Gabbro

Upper paleozoic

Fig. 2. Generalized lithostratigraphic relations in the McCarthy Quadrangle (based on MacKevett

1978, Silberman et al. 1981)

(Fig. 3). Csejtey et al. (1980) consider Wrangellia terrane a part of the accre-
tionary continental margin of western North America. It joined the geologically
and stratigraphically similar Peninsular terrane (Csejtey et al. 1978, Jones and
Siberling 1979), occurring to the west of the Quadrangle, in a tectonic coupling in
the Middle Jurassic. This new coherent terrane, named the Talkeetna superter-
rane by Csejtey et al. (1980), was emplaced in Alaska during the Middle
Cretaceous accretionary orogeny. The Wrangellia part was juxtaposed in the
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 427

o
64-

62~

'--__ 200 KILOMETERS

-'--_-----"

Fig. 3. Tectonostratigraphic terrane map of Southern Alaska. Location of the McCarthy Quadrangle
is shown by the box (modified from Csejtey et al. 1982)

northeast against the Precambrian Yukon-Tanana terrane along a right lateral

strike-slip (or transform) Denali fault system - a tectonostratigraphic suture
(Csejtey 1976, Csejtey et al. 1982).
In the northeastern part of the McCarthy Quadrangle, the Totschunda fault
is a conjugate part of the eastern portion of the Denali fault system (Fig. 3) that
developed in the Cenozoic deformational phase along the reactivated Cretaceous
fault system (Richter and Matson 1971) as a postaccretionary feature (Csejtey et
al. 1982).
The high-angle reverse or a thrust fault in the central part of the McCarthy
Quadrangle, herein referred to as the Nelson thrust fault (Fig. 1b), is a prominent
feature that affects the Triassic stratigraphy. It is probably related to the
Cretaceous accretionary process. Along its trend, the Nelson thrust is disrupted
by steep-dipping, northeasterly and easterly faults. Kennecott-type mineraliza-
tion is largely localized within these younger, northeasterly fracture systems.
In the southwestern part of the Quadrangle, the Cretaceous or Tertiary age
Border Ranges fault (Fig. 3) is a high-angle reverse fault that grades into a thrust
fault to the west of the Quadrangle (Mackevett 1978).

Lithologic Characteristics of the Triassic Units

The Triassic Nikolai Greenstone (dark) and Chitistone Limestone (light gray)
form an impressive and cohesive geologic unit whose contact is marked by a
428 M. K. Sood et al.

NELSON PEAVINE BINOCULAR Fig. 4. Lithostratigraphic relations at the

mineral prospects of the present study
(partly based on MacKevett 1978)
;z:=t TC LI;c:=.::;::

~ITCl~

RAOOVAN MOUNTAIN CLEAR

GRill PORCUPINE

CHITINA CHITISTONE NIKOLAI fAULT CONTACT

INTRUSIVES LIMESTONE GREENSTONE
·L--------v-------~·

JURASSIC TRIASSIC

lithologic color contrast along its entire length in the Quadrangle. The strati-
graphic relations at each prospect of this study are shown in Fig. 4.

Nikolai Greenstone

The Nikolai Greenstone is comprised of a series of i5-cm to 4-m-thick subaerial

basaltic flows which, due to their overlapping relations, appear to be continuous.
Mackevett (1978) estimated the cumulative thickness for the unit to be about
4 km and reported that pillowed structures are present in the basal portions in the
northeastern part of the Quadrangle. However, in the study area the unit is mas-
sively bedded and abundantly amygdaloidal.
At Radovan, the Nikolai unit is intensely fractured and altered along the frac-
tures; at Peavine and Mountain Grill prospects, it is affected by faulting and
overturned folding (Plate I) and at Clear-Procupine, it has been contact meta-
morphosed by the Jurassic Chitina Valley granodiorite.
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 429

Plate I. Faulting and overturned folding in the Triassic Iithounits at the Mountain GriII Prospect can
be seen in the central part of the picture which is taken looking north. Nikolai Greenstone (darker) is
seen in the left central part. In other places, the greenstone/ limestone contact is cohesive and struc-
turally conformable

Petrographically, the Nikolai rocks display the effects of pervasive, low-tem-

perature alteration-metamorphism, but without penetrative deformation.
Primary minerals (augite, labradoritic plagioclase, relict olivine, hypersthene,
magnetite, and ilmenite) and textures (fine-grained volcanic, porphyritic, inter-
granular, and subophitic) are preserved to varying degrees (Plate IIa). Plagio-
clase is a common phenocryst. A variety of minerals fill the amygdules (Plate
lIb).
The representative mineralogy (Plate II a, b, c) of the unit based on the
examination of about 100 samples is given below:

Primary Minerals (Plate II a, b, c)

Augite 40 to 50070
Plagioclase (An 35 to An 65) 35 to 45%
Olivine
Hypersthene
Pigeonite
} Not always present. Up to 5%

Magnetite
Ilmenite
Apatite
} Up to 10%
 ...."""o

~
?'<:
en
o
o
Q.
~
e!.
Plate II. a Nikolai Greenstone in partially altered state, but the volcanic texture is preserved. Chlorite (Ch) and epidote (Ep) are the principal
products of alteration-metamorphism in this sample. b Calcite (Ct) filling an amygdule in Nikolai Greenstone. c Epidote (Ep)-filled fracture in
Nikolai Greenstone. d Sparry calcite (Ct) filling the openings in dolomite (limpid or clear variety)
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 431

Secondary and Alteration Minerals (Plate II a, c)

Kaolinite Sericite May total up to 300/0
Chlorite Prehnite
Pumpellyite Calcite
Quartz Epidote
Hematite Mica
Laumontite
Amygdule Minerals (Plate lIb)
Chlorite Natrolite
Calcite Thomsonite
Quartz Native copper
The average chemical composition of the Nikolai rocks is given in Table 1.
The rocks are largely quartz/hypersthene normative, but do contain normative
olivine/nepheline. Major and minor element chemistry of less altered samples
shows a relatively high degree of compositional uniformity. Their chemical rela-
tion in terms of Ah03 - MgO - FeO show ocean island-continental transitional
tectonic affinities.

Chitistone Limestone

The Chitistone Limestone generally forms thick beds and overlies the Nikolai
Greenstone as a cohesive structurally conformable unit. At Nelson (Fig. 4), the

Table 1. Average chemical compositions of Nikolai Greenstone, McCarthy Quadrangle, Alaska

Oxides Wt%

Radovana Peavinea Mountainb Grill Nelsona MacKevett and Muller et al.

Prospect Prospect Prospect Prospect Richter (1974) (1974)

Si02 51.12 49.05 54.78 50.28 50.30 49.30

Al20 3 13.22 16.10 17.49 14.27 15.23 14.46
F~03 4.54 4.95 2.3 6.49 5.46 4.26
FeO 8.94 7.19 6.73 6.46 6.72 6.76
MgO 6.51 8.34 3.89 7.37 7.25 8.01
CaO 9.27 9.24 12.75 10.09 9.87 10.92
Na20 2.97 2.59 0.04 2.53 3.26 3.64
K20 0.17 0.49 0.22 0.40 0.50 tr
Ti02 2.62 1.65 1.64 1.72 1.47 1.30
P 20 S 0.27 0.16 0.13 0.18 0.17 0.10
MnO 0.22 0.17 0.09 0.15 0.19 0.18
Total 99.85 99.94 100.50 99.94 99.98 99.98
Volatiles 4.99 4.26 17.26 3.60 3.60 3.90

Analysis by K. Ramlal.
tr = trace
a Average of 3 analyses.
b Single analysis of epidotized sample.
432 M. K. Sood et al.

Chitistone is overlain by Nikolai Greenstone along the Nelson thrust contact; at

Peavine and Mountain Grill prospects the overturned folding and reverse fault-
ing has resulted in the limestone wrapping around the Nikolai Greenstone (Plate
I). At the Clear-Porcupine prospect, Chitistone is metamorphosed to skarn and
marble near its contact with the Jurassic intrusives (MacKevett 1978). At
Binocular, Peavine, and Nelson prospects, the limestone is abundantly fractured
with most trending almost north-south and northeasterly. Calcite and dolomite
veins, hairlike to several inches wide, commonly occur criss-crossed, parallel,
and less frequently braided. Breccia and collapse features have been observed at
Nelson.
The Chitistone unit is predominantly a limestone. Dolomite, dolomitic lime-
stone, and chert are better developed in the stratigraphically higher parts. For the
most part, it shows mineralogical and chemical purity within the lithofacies.
The main minerals are calcite [micrite sparry (Platte lId) fibrous, crystalline,
and intracrystalline varieties] and dolomite [limpid or clear (Plate lId), fine-
grained matrix, granular and coarse-grained]. Clays, quartz, pyrite, and chert
constitute the minor minerals.
Comprehensive studies of the stratigraphic section at Nelson Prospect show
that the unit is comprised of many microlithofacies conforming to the classifica-
tion of Folk (1962), Dunham (1962), and as amplified by Wilson (1975). Litho-
facies characteristics for the stratigraphically lower 125 m compare well with
lithofacies characteristics of the Chitistone unit at the Donoho Peak section at
Kennecott (Armstrong and MacKevett 1982) and the section at Green Butte
(Armstrong et al. 1969). Of particular importance is the fact that gray-lime mud-
stone facies in the lower 60 m seems to be the most continuous horizon of the
Chitistone and, therefore, a good marker bed in the Quadrangle.

Table 2. Chemical composition of Chitistone Limestone, McCarthy Quadrangle, Alaska

Oxides WtOJo

Nelsona Binocularb Peavine"

Dolomite Dolomitic Limestone

limestone

Si~ 0.62 2.07 1.16 3.0 nd

Al2~ 0.12 0.50 0.37 <0.10 nd
FeO 0.09 0.13 0.15 1.62 0.16
MgO 18.57 2.04 0.74 17.20 0.50
CaO 33.07 50.77 53.46 24.36 53.69
MnO 0.05 0.47 0.03 nd 0.01
H 20 total tr 0.78 0.44 1.15 0.19
C~ 46.38 42.85 42.86 40.42 42.49
S (ppm) 78 nd tr nd nd

Analysis by K. Ramlal.
tr = trace
nd = not determined
a Average of 3 analyses.
b Single analysis.
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 433

Argillaceous limestones are prominent in the stratigraphically lower parts and

these contain burrows, crinkles, pelletoids, peloids, and other sedimentary struc-
tures. The more dolomitic upper parts of this platform sequence resemble the
"zebra bed" described for the stratigraphic section at Kennecott (Armstrong and
MacKevett 1982). The Chitistone Limestone was deposited in subtidal to supra-
tidal environments with possible sabkha affiliation.
The CaC0 3 /MgC0 3 ratios (1.39 to 71.78) for Chitistone correspond to
dolomites, dolomitic limestones, and limestones.
Chemical composition for the Chitistone unit is given in Table 2. Some
dolomitic samples from the Nelson prospect show sulfur contents of 78 ppm.

Table 3. Geological characteristics of copper mineralization in Nikolai Greenstone, McCarthy Quad-

rangle, Alaska

Characteristics Radovan Prospect Mountain Grill Pro- Clear-Porcupine Pro-

spect spect

Lithologic Intensely fractured; Faulted, overturned Fractured; affected by

calcite, chlorite, and folding; veins of epi- Jurassic grandiorite
laumontite; common dote intrusion
fracture fillings
Alteration (Plate II) Quartz, chlorite, Calcite, chlorite abun- Quartz, calcite sul-
pumpellyite, K-feld- dance appears to have fides near the J uras-
spar, mica, and hema- "bleached" the host sic intrusive; epidote
tite rock
Form and control of Veins up to 45 cm Veins 5 to 10 cm wide; Veins up to 10cm
ore wide; disseminations; native Cu in epidote wide; disseminations;
limited reaction; ores veins; ores localized to native Cu in amyg-
confined to northeast- shear zones dules; ore along
erly fractures northeasterly fracture
system
Ore mineralogy (See Plate IlIa, b)
Major Digenite, chalcocite, Chalcocite, digenite, Bornite, chalcopyrite,
bornite, chalcopyrite bornite digenite
Minor Covellite, enargite, Chalcopyrite, covel- Covellite
stibnite lite
Secondary Malachite Malachite Malachite
Texture Massive; replacement Massive/granular; Massive covellite re-
features of bornite by some sulfide and host places bornite along
other phases rock replacement fracture
Mineral associations Bornite-digenite-chal- Chalcopyrite-digenite- Bornite-chalcopyrite-
copyrite chalcocite digenite
Metal content of
ores
Cu (WtOJo) 45 to 71 4a 55 to 58
Fe (WtOJo) 2 to 8 nd 10 to 15
Ag (ppm) 50 to 200 <3 131 to 300

a Grade of disseminated ore in Nikolai Greenstone.

nd = not determined
434 M. K. Sood et al.

Mineralization

The distribution of known copper occurrences in the McCarthy Quadrangle

defines a northwesterly trend in conformity with the Triassic belt (Fig. 1b).
Geologic characteristics of mineralization in the Nikolai Greenstone at Radovan,
Mountain Grill, and Clear-Porcupine prospects are summarized in Table 3 and
for those occurring in Chitistone Limestone are summarized in Table 4.

Table 4. Geologic characteristics of copper deposits associated with Chitistone Limestone, McCarthy
Quadrangle, Alaska

Characteristics Nelson Prospect Binocular Prospect Peavine Prospect

Lithologic Platformal carbon- Fractured dolomite, Lime mudstone, dolo-

ates, abundant criss- lime mudstone, veined mite facies brecciated
cross and irregular and solution features
calcite, dolomite fill-
ings
Alteration Dolomitization, minor Rusty stains from py- Hematite stains,
silicification, recrys- rite weathering cherty zones; some re-
tallization on fracture crystallization
walls
Form and control of Swell and pinch reach Fracture control of Up to 30 cm wide,
ore 60 cm in width, stock- ore veins, disseminat- steep fracture fillings
work system, only ed, minor bedding
very limited bedding control
control
Ore mineralogy (See Plate IIIc and d)
Major Digenite, chalcocite, Digenite, chalcopyrite Chalcopyrite, digenite,
djurieite, covellite, pyrite
bornite
Minor Enargite Covellite Covellite
Secondary Malachite, azurite, Malachite, azurite Malachite, azurite
cuprite
Texture Massive, fine to Massive, fine-grained; Massive, granular; re-
coarse granular; bor- mutual replacement placement between
nite-chalcopyrite, di- features in ores principal phases; co-
genite-chalcocite-djur- vellite alteration
leite exsolution; bor-
nite replaced by di-
genite
Mineral associations Bornite-chalcopyrite Chalcopyrite-digenite Chalcopyrite-digenite-
digenite-djurleite-chal- covellite
cocite
Metal content of
ores
Cu (Wt%) 49 to 76 10" 15"
Fe (WtOJo) 0.2 to 0.3 nd nd
Ag (ppm) 190 to 400 14 11

" Grade of disseminated ore in Chitistone Limestone.

nd = not determined
 til
::i
I>l
S-
O"
o
c
::s
P-
()
o
"0

"8
o
.g
~.
1n'
S·
tr1
I>l
~
til
o
c
:r
()
(I)

a
...
e.
~
:0<-
f'!
[...
()
0"
~
g
~.

~.
~
P-
...o
€:::s
Plate III. a Bornite (Bn) and chalcopyrite (Cp) showing replacement relations in ores in Nikolai Greenstone at Radovan. Note the embayed
margins with silicate vein material. b Bornite (Bn) and chalcopyrite (Cp) replacement relations in the sulfide ores at the Clear-Porcupine
prospect. c Mineralization in Chitistone Limestone. Note the distinct ore vein-host rock relations. d Digenite (Dg) chalcocite (Cc) and .....
""'"
v.
bornite (Bn) in the ores at the Nelson prospect. Note the exsolution relations
436 M.K. Sood et aI.

Nikolai Fig. 5. Relation of copper and

80 Hosted 10Cl-I00
Sulphides silver in the Kennecott-type ore
deposits (Kennecott data from
~ Bateman and McLaughlin 1920)
Chilistone
,,-Hosted
80 Sulphides

20

o~-----T------~----~~----~------'
100 200 300 400 500
All ppm

Tkeir comparison brings out the following points:

1. In both the units, the structural control of mineralization predominates.
2. The fracture system is relatively continuous and prominent in the Chitistone.
3. Digenite is more abundant than chalcocite in the Kennecott-type ores.
4. Copper sulfides, digenite, and chalcocite are present in both the units, but
bornite and chalcopyrite are relatively abundant in the ores in greenstone, par-
ticularly at Clear-Porcupine. (Plate IIIb)
5. There is no pervasive deformation of the ores.
6. Ore vein-host rock contacts are generally distinct, sharp, or with low degree of
alteration reaction.
7. Secondary alteration minerals, malachite, and/or azurite, occur at every
prospect.
8. Ore textures and structures indicate precipitation from fluids.
9. Silver content parallels copper content (Fig. 5).

Discussion

The ore-host rock system in Nikolai Greenstone is a representative of low-tem-

perature, hydrothermal alteration-metamorphism. Rock-water interaction
through hydrothermal circulation at the ocean ridges is well recognized and es-
tablished for the alteration of ocean crust (Rise Project Group 1980, 1981, Mevel
1981). Nikolai Greenstone has such alteration features in common; it lacks pene-
trative deformation as shown by the preservation of primary textures and
mineralogy in the rocks and by crystallization of secondary hydrous minerals in
veins, amygdules, and as pseudomorphs of magmatic minerals.
Recently, Seyfried (1982) and Reed (1983) have given comprehensive experi-
mental data and calculations, respectively, for the seawater-basalt reaction to
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 437

300 DC. Many minerals found in metabasalts are also produced in seawater-basalt
reactions. Reed (1983) concludes that the seawater-basalt ratio greather than 30
favors the formation of chlorite (hydration in natural systems), whereas a ratio
of less than 30 promotes the crystallization of epidote (dehydration in natural
systems at depth). Therefore, the dehydration reactions in the lower parts of a
lava pile could make significant contributions to the hydration in the upper parts.
Jolly (1974) has correlated the mobilization and enrichment of copper in the
Keweenawan basalts to the hydration and the precipitation of native copper-
epidote to the dehydration of the basalts. Depth-dependent, dehydration-hydra-
tion reactions in volcanic piles, therefore, become important as they affect
chemical and elemental redistributions. The variation in the fluid composition as
it penetrates the rock is probably responsible for the diversity observed in the
mineralogy of altered mafic volcanics.
Chlorite, pumpellyite, quartz, calcite, laumontite, and epidote mineralogy
found in the veins is also found in altered Nikolai Greenstone. Despite their
somewhat irregular distribution, such hydrous mineral assemblages form repre-
sentative paragenesis of low-temperature alteration-metamorphism of mafic
volcanics (MeveI1981). Using the fluid pressure vs temperature data by Nitsch
(1971) and Liou (1979) on the stability relations for the prehnite-pumpellyite
(chlorite + albite) reaction, we conclude that the Nikolai assemblages formed
around 250 DC and pressure of 1 to 2 kb - the pressure-temperature conditions
of prehnite-pumpellyite facies metamorphism (Coombs 1960, Turner 1981).
The presence of primary bornite-chalcopyrite association and the identifica-
tion of a high chalcocite phase in the ores of this study are also in conformity
with temperature estimates of 200 to 300 DC.

Table 5. Selected trace metal contents in epidote-quartz veins in Nikolai Greenstones, McCarthy
Quadrangle, Alaska

(ppm)

Radovan Prospect" Kennecottb

RMS-8 RMS-9 9 18 E3

Width of vein (cm) 10 25 10 <3 30

Ag 3 <3 2 0.5 30
Co 18 47 15 30 15
Cr 74 150 70 150 300
Cu 10830 264 5000 300 >20000
Ni 20 83 30 50 15
Pb nd nd 10 ND 10
Zn 36 76 10 15 10
As <50 nd 20 ND ND

aAnalysis by K. Ramlal.
bFrom Silberman et aI. (1981).
ND = not detected
nd = not determined
438 M. K. Sood et al.

Silberman et al. (1981), on the basis of oxygen isotopic fractionation studies

on the Nikolai (and the epidote-quartz mineral veins in it) estimate the tempera-
ture for hydrothermal alt~ration-metamorphism of Nikolai to be about 200°C.
At Radovan, laumontite-ore intergrowth is an important indicator that
alteration and ore formation, at least in Nikolai Greenstone, was a related pro-
cess. Ore vein-host rock contacts, for the most part, are sharp.
In Table 5, the trace element contents of epidote veins from the Nikolai
Greenstone are compared. There are obvious similarities in the trace metal con-
tents of epidotes both at Kennecott and those of the present study. This would
indicate to us that the Nikolai unit in the belt, or at least in these two places,
experienced similar thermal conditions and chemical changes.
The ore-host rock system in Chitistone Limestone displays several important
characteristics:
1. Ore veins are in intimate and sharp contact with the unaltered limestone and
have matching walls suggesting fracture fillings by relatively cold solutions.
2. Ore vein-host rock contacts are embayed and separated by thin, calcite/
dolomite walls indicating reaction and accompanying precipitation/recrystal-
lization (Plate IIIc).
3. Stockwork arrangement of closely-spaced fracture fillings.
4. Limited or minor disseminations and bedding replacement.
Dolomitization is the principal alteration and is expressed on all scales, but
commonly discordant to bedding. It is probably premineralization and of hydro-
thermal origin.

Model for Mineralization

Undoubtedly, the Nikolai Greenstone due to its intrinsically high copper content
(MacKevett et al. 1981), is the most likely source of the metal for the Kennecott-
type mineralization in the McCarthy Quadrangle. Copper is known to concen-
trate in magnetite (200 to 880 pm), ilmenite (90 to 300 ppm), and mafic silicate
minerals (50 to 320 ppm) (Cornwall 1956, Duncan and Taylor 1968).
The following processes are considered to have been important in the concen-
tration of copper from such minerals into Kennecott-type deposits. Hydration
associated with the alteration-metamorphism of the Nikolai Greenstone rocks
released copper. With progressive hydration, copper ions were mobilized and
enriched in fluids in the fractures as water was fixed in the structure of the
mafics.
Our studies of the mineralogy and chemistry of the veins and the Nikolai sug-
gest that the alteration characteristics of greenstones were analogous to those de-
scribed by Sinclair (1977) at White River in Yukon and by Silberman et al. (1981)
at Kennecott. Sinclair (1977) envisaged that the fluids which caused the alteration
and mineralization in Triassic greenstones were metamorphic waters mixed with
connate waters derived from the Upper Paleozoic sedimentary rocks. Silberman
et al. (1981) favors the modified connate water concept on the basis of their
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 439

oxygen isotopic values of + 1 per mil for vein epidote in the Nikolai Greenstone.
We consider that this thermal-alteration and ore-forming event was of a regional
nature and that the chemistry of the fluids that affected the greenstone of the
present study were similar to those proposed by Sinclair (1977).
Dehydration-hydration reactions in mafic volcanic piles can result in
pervasive mobilization, redistribution, and concentration of elements during
prehnite-pumpellyite facies conditions. Development of chlorite, pumpellyite,
and epidote occurs first on the fracture walls before fluids diffuse into the
homogeneous interiors of the rock. The intensity and nature of fractures and
fluid pressure control the extent of the alteration.
We propose that dehydration-hydration reactions in this early stage may have
been responsible for the widespread, but low-grade disseminated, thin-fracture
filling, and native copper mineralization in the Nikolai. Particularly important in
this respect are the ore-laumontite intergrowths and native copper-epidote asso-
ciations observed in the Radovan area. The relatively iron-rich (see Tables 3 and
4) character of the ore is also an important feature of this stage.
Lack of extensive alteration particularly near the veins in Chitistone Lime-
stone suggest that the ore solution at least when they arrived there were not very
hot or acidic (Sood and Wagner 1981). MacKevett et al. (1981) favor a low-tem-
perature (100°C) meteoric water hydrothermal circulatory system for the ore-
forming process at Kennecott.
Our studies of the Clear-Porcupine prospect indicate that the Kennecott-type
mineralization is younger than Jurassic as it is not affected by the Jurassic
Chitina Valley batholith. On the basis of K-Ar systematics, Silberman et al.
(1981) reported an age of 112± 11 m. y. for metamorphism of the Nikolai Green-
stone, coincident with the Cretaceous accretionary event.
Structural relations are also important for timing of mineralization. The Nel-
son thrust fault occurs close or near to the centers of intense mineralization (Fig.
1b). It is probably related to the late phase of the Creataceous accretionary event or
the later Cenozoic deformation that activated the Denali system in its eastern
portion forming the younger Totschunda fork (Fig. 3). The study area is character-
ized by three principal fault sets, i.e., northerly, northeasterly, and easterly. The
Nelson thrust at Radovan, is displaced by steep-dipping northeasterly faults, and at
the Mountain Grill area by an easterly fault set. Copper mineralization is localized
by such (probably Cenozoic) fracture systems. They are better developed in the
Chitistone Limestone than in the Nikolai Greenstone. The expanded fracture
system at this stage, further enhanced the porosity and permeability of the units,
thus focusing the meteoric water circulatory system for effective and efficient
solution and transport of metals. We conclude that formation of the Kennecott-
type mineralization initiated with the main phase of the Cretaceous accretion, but
occurred mainly in its late phase, perhaps continuing into the Tertiary.
The proximal relations of the mineralization to the Tertiary intrusives, both
at Kennecott and the prospects of this study in the McCarthy B-4 Quadrangle
suggests that the Tertiary igneous activity may have thermally charged and driven
the system. Although data are incomplete in this respect, we have, however,
noticed high potassic alteration zones (see also MacKevett and Radtke, 1966) and
relatively high As and Sb contents in the ores in the limestones.
440 M. K. Sood et al.

The size of the deposits may be more a function of the availability of the frac-
ture system and cavities which are preferentially developed in limestone.

Conclusion

It is concluded that Kennecott-type mineralization in southeastern Alaska,

though of a different character in the Nikolai Greenstone and the Chitistone
Limestone units is the result of a continuous mineralizing epoch. It is, most like-
ly, the last major chemical event affecting the Triassic rocks may, therefore, be
Late Cretaceous to Tertiary age.
Dehydration-hydration reactions of prehnite-pumpellyite facies metamor-
phism in the volcanic pile contributed to the ore-solution chemistry. The upper
temperature prevalent in the ore-host rock system, as indicated by mineral as-
sociations, was about 250 DC. Prominent mineralization in the Triassic belt is a
function of the disposition of the fracture system and the efficiency of solution
and transport by water circulatory systems thermally charged by the Tertiary
igneous activity.
The control of Kennecott-type mineralization by a fundamental process of
ore formation operating on a regional dimension strongly favors the presence of
large mineralization centers. The alteration characteristics of the Nikolai Green-
stone, particularly the MgO and K20 contents, and the fracture analyses in the
Triassic units should be utilized in exploration for additional deposits.

Acknowledgments. This study has been supported by grants (to MKS) by Geneva Pacific Corp. (now
Testing Sciences, Inc.), Glenview, IL, USA. The help of C. Witt and F. Megiris with the drafting of
the diagrams and of M. McGuigan for typing the manuscript is gratefully acknowledged.

References

Armstrong AK, MacKevett EM Jr (1982) Stratigraphy and diagenetic history of the lower part of the
Triassic chitistone limestone, Alaska. US Geol Surv Prof Pap 1212A:A1-A26
Armstrong AK, MacKevett EM Jr, Silberling NJ (1969) The Chitistone and Nizina limestones of part
of the southern Wrangell Mountains, Alaska - a preliminary report stressing carbonate petrogra-
phy and depositional environments. US Geol Surv Prof Pap 650-0:-49-062
Bateman AM (1932) Notes on Kennecott-type copper deposits, Glacier Creek, Alaska. Econ Geol
27:297-306
Bateman AM, McLaughlin OH (1920) Geology of the ore deposits of Kennecott, Alaska. Econ Geol
15:1-80
Coombs OS (1960) Lower grade mineral facies in New Zealand. Rep Int Geol Congr, 21st Session
Norden 13:339 - 351
Cornwall HR (1956) A summary of ideas on the origin of native copper deposits. Econ Geol
54:351-373
Csejtey B Jr (1976) Tectonic implications of Late Paleozoic volcanic arc in the Talkeetna Mountains,
Southcentral Alaska. Geology 4:49 - 52
Csejtey B Jr, Nelson WH, Jones OL, SilberlingNJ, Oean RM, Morris MS, Lanphere MA, Smith JG,
Silberman ML (1978) Reconnaissance geologic map and geochronology, Talkeetna Mountains
Quadrangle, northern part of Anchorage Quadrangle and southwest part of Healy Quadrangle,
Alaska, Scale 1:250,000. US Geol Surv Open File Rep 78-558-A, 62 pp
Stratabound Copper Deposits in East South-Central Alaska: Their Characteristics and Origin 441

Csejtey B Jr, Foster HL, Nokleberg WJ (1980) Cretaceous accretion of the Talkeetna Superterrane
and subsequent development of the Denali fault in Southcentral Alaska. Geol Soc Am (Abstracts
with Programs) 12:409
Csejtey B Jr, Cox DP, Evarts RC (1982) The Cenozoic Denali fault system and the Cretaceous accre-
tionary development of Southern Alaska. J Geophys Res 87:3741- 3754
Duncan AR, Taylor SR (1968) Trace element analyses of magnetites from andesitic and dacitic lavas
from the Bay of Plenty, New Zealand. Contrib Mineral Petrol 20:30 - 33
Dunham RJ (1962) Classification of Carbonate Rocks according to depositional textures. In: Ham
WE (ed) Classification of carbonate rocks. AAPG Mem, pp 108 -122
Folk RL (1962) Spectral division of limestone types Symp. Classification of carbonate rocks. In: Ham
WE (ed) Classification of carbonate rocks. AAPG Mem, pp 62-84
Jensen ML, Bateman Am (1979) Economic mineral deposits. John Wiley & Sons, New York, 593 pp
Jolly WT (1974) Behavior of Cu, Zn, and Ni during prehnite-pumpellyite rank metamorphism of Ke-
weenawan basalts, Northern Michigan. Econ Geol 69: 1118 -1127
Jones DL, Silberling NJ (1979) Mesozoic stratigraphy, the key to tectonic analysis of Southern and
Central Alaska. US Geol Surv Open File Rep 79-1200,41 pp
Jones DL, Silberling NJ, Hillhouse JW (1977) Wrangellia - A displaced terrane in northwestern
North America. Can J Earth Sci 14:2565 - 2577
Liou JG (1979) Zeolite facies metamorphism of basaltic rocks from the Taiwan ophiolite. Am Miner-
al64:1-14
MacKevett EM Jr (1976) Mineral deposits and occurrence in McCarthy Quadrangle, Alaska. US Geol
Surv Misc Field Stud Map M F 773-B Scale 1:250000, 2 sheets
MacKevett EM Jr (1978) Geologic map of McCarthy Quadrangle Alaska. US Geol Surv Misc Invest
Ser Map 1-1032 Scale 1:250000
MacKevett EM Jr, Radtke AS (1966) Hydrothermal alteration near the Kennecott copper mines,
Wrangell Mountains area, Alaska - A preliminary report. US Geol Surv Prof Pap
550-B:B165 - B168
MacKevett EM Jr, Richter DH (1974) The Nikolai Greenstone in the Wrangell Mountains, Alaska,
and nearby areas. Geol Assoc Can, Cordilleran Section, Program and Abstracts 13:14
MacKevett EM Jr, Albert NRD, Barnes DF, Case JE, Robinson K, Singer DA (1977) The Alaskan
mineral resource assessment program: background information. US Geol Surv Cir 739, 23 pp
MacKevett EM Jr, Armstrong AK, Potter RW, Silberman ML (1981) Kennecott-type copper depos-
its, Wrangell mountains, Alaska, an update and summary. In: Silberman ML et al. (eds). Symp on
Mineral Deposits of the Pacific Northwest. US Geol Surv Open File Rep 81-355:51
Mevel C (1981) Occurrence of pumpellyite in hydrothermally altered basalts from the Vema fracture
zone (Mid-Atlantic Ridge). Contrib Mineral Petrol 76:386 - 393
Muller JE, Northcote KE, Carlisle D (1974) Geology and mineral deposits of Alert-Cape Scott map
area, Vancouver Island, British Columbia. Geol Surv Can Pap 74-8:77 pp
Nitsch H (1971) Stabili-tats-beziehungen von prehnit-pumpellyite-haltingen paregenesen. Contrib
Mineral Petrol 30:240 - 260
Reed MH (1983) Seawater-basalt reaction and the origin of Greenstones, and related ore deposits.
Econ Geol 78:466-485
Richter DH, Matson NA (1971) Quaternary faulting in the eastern Alaskan Range. Geol Soc Am Bull
82:1529-1540
Rise Project Group (1980) East Pacific Rise: Hotsprings and geophysical experiments. Science
207:1421-1433
Rise Project Group (1981) Crustal processes of the mid-oceanic ridge. Science 213:31-40
Seyfried WE (1982) Hydrothermal alteration of basalt by seawater under seawater-dominated condi-
tions. Geochim Cosmochim Acta 46:985 -1002
Silberman ML, MacKevett EM Jr, Connor CL, Mathews A (1981) Metallogenic and tectonic signifi-
cance of whole rock potassium - Argon ages of the Nikolai Greenstone, McCarthy Quadrangle,
Alaska. In: Silberman ML et al. (eds). Symp on Mineral deposits of the Pacific northwest. US Geol
Surv Open File Rep 81-355:53 -73
Sinclair AJ (1977) The White River Copper deposit, Southwestern Yukon. Geol Assoc Can (Program
with Abstract) 2:49
442 M. K. Sood et al.: Stratabound Copper Deposits in East South-Central Alaska

Sood MK, Wagner RJ (1981) Copper mineralization at Glacier Creek, McCarthy Quadrangle, Alas-
ka. Geol Soc Am Abstr (Northcentral Section) 13:317
Turner FJ (1981) Metamorphic petrology: mineralogical, field and tectonic aspects, 2nd edn. Mc-
Graw-Hili, New York, 451 pp
Wilson JL (1975) Carbonate facies in geologic history. Springer, Berlin Heidelberg New York,
471 pp
Major Element Geochemistry of the Host Rocks in Some
Sediment-Hosted Copper Deposits
B. MOINE 1,2, L. GUILLOUX 3, and D. AUDEOUD 2,4

Abstract

The chemical compositions of fine-grained sedimentary rocks provide important

genetic indications when their mineralogical significance is brought out. The
chemical approach is essential when the primary features have been obliterated
by metamorphic crystallization. Recent progress in the geochemistry of shales
and marls from evaporite-bearing series is of special interest with regard to
sediment-hosted copper deposits.
Two major chemical characteristics are obvious in the environment of the
Ore-Shale deposits in the Zambian Copperbelt based on comparison of 378
analyses on systematically chosen samples from drill holes and cross-sections in
Konkola, Chingola, Chambishi, Mindola, Rokana, and Luanshya (metamor-
phism from greenschists to lower amphibolite facies) with shales and marls from
common platform series. Firstly, the relatively high Mg (and Li) contents are
characteristic of magnesian clay minerals, in the early members of evaporitic
sequences. This is an ubiquitous feature whereas anhydrite is only of local occur-
rence. Secondly, primary (premetamorphic) fine-grained rocks with very high K-
feldspar and comparatively low quartz contents (around 63"70 feldspar, 10%
quartz and 27% chlorite-rich clays) are inferred from extremely high KP con-
centrations (often ~10%) and K/AI ratios in the Ore-Shale Formation. These
compositions cannot be obtained by sedimentary processes only and a large part
of the feldspar is probably of diagenetic/hydrothermal origin.
In Shaba (Kamoto, Kambove) the Cu and Co concentrations are carbonate-
hosted. However, very uncommon chemical compositions are exhibited by the
Mg-chlorite-quartz-dolomite rocks of the underlying RAT formation (32
analyses): high Mg, Li and low alkali contents. They could partly derive from
felsic-volcanic glass altered by reaction with Ca- and Mg-rich and alkali-poor

I Laboratoire de Mineralogie, UA67 du CNRS, Universite Paul Sabatier, 39 Allees Jules Guesde,
31400 Toulouse, France
2 Laboratoire de Petrographie, UA805 du CNRS, Universite Claude Bernard, 27 - 43 Bd du 11
Novembre, 69622 Villeurbanne Cedex, France
3 Departement Gites Mineraux, Bureau de Recherches Geologiques et Minieres B.P. 6009, 45060
Orleans Cedex, France
4 Centre de Recherches sur la Geologie de I'Uranium B.P. 23, 54501 Vandoeuvre-Nancy Cedex,
France

Geology and Metallogeny of Copper Deposits

(ed. by G.H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
444 B. Moine et aI.

brines. Such brines have been observed in fluid inclusions related to V-mineral-
ization which occurs at the top of the formation.
In contrast, the German Kupferschiefer (Mansfeld district, 22 analyses) does
not show major element compositions different from those of common black
shales.

Introduction

The chemical study (major and selected trace elements) of the rocks hosting
stratiform Pb, Zn and Cu mineralization, in particular of the free-grained sedi-
ments and metasediments such as in the African Copperbelt and Kupferschiefer
types of deposits, provides interesting information regarding metallogeny. Such
quantitative descriptions of argillites, marls and other fine-grained sediments
usefully complement the mineralogical observations with sedimentological inter-
pretation in view. As illustrated by numerous studies in metamorphic domains
(Shaw 1956, de La Roche 1965,1968,1974, Moine 1971), these chemical data are
necessary for the precise identification of the sedimentary origin of metamorphic
rocks in which the original mineralogy has been changed and the primary struc-
tures have been modified.
Regarding metallogeny, one might expect information on the following:
1. The sedimentary maturity of the host rocks.
2. Characteristics indicating a specific depositional environment, especially a
restricted (evaporitic) environment as many authors have pointed out the
frequent association of Pb, Zn and Cu stratiform deposits with evaporitic
series (Renfro 1974). Several recent works (Trauth 1974, Kulke 1976, 1978,
Jarousse et al. 1978 and Moine et al. 1981) have defined the chemical charac-
teristics of clays and marls of the evaporitic series thus contributing to the
identification of these series in the metamorphic domains.
3. The possible contamination by volcanogenic materials.
4. The possible diagenetic/hydrothermal alteration which can be assumed in the
case of compositions unexplained by sedimentary or volcanic processes.
This approach is illustrated here by the results of the chemical study on the
Copperbelt of Central Africa, i.e. the Ore-Shale of Zambia and the bottom of
the Mine series of Shaba. These results are included in the work of Guilloux
(1982) and Audeoud (1982) who carried out geochemical studies in cooperation
with the first author (B.M.).

The Ore-Shale of Zambia

Samples and Chemical Data - Metamorphism

Many papers have been devoted to the geology of the Zambian Copperbelt in
particular the composite studies by Mendelsohn (1961), Garlick and Fleischer
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 445

(1972) and Fleischer et al. (1976). The stratiform Cu mineralization is located in

the Lower Roan Group (around 1000 m.y. old) directly overlying the Kibarian
(around 1300 m.y.) gneissic and granitic basement. In the area to the southwest
of the Kafue anticline (Fig. 1), Annels et al. (1983) summarized the lithological
succession of the Lower Roan as follows:
1. Hangingwall: argillites and quartzites with local dolomite interbeds;
2. Ore-Shale Formation: siltstones, argillites, carbonaceous argillites and impure
dolomites;
3. Footwall: footwall conglomerates, quartzite and argillite, lower conglo-
merate, quartzite (from top to bottom).
The mineralization is located in the Ore-Shale (Ore-Shale Alignment), then,
towards the southwest, in the basal arenites. To the northeast of the Kafue dome,
the Ore-Shale is no longer observed, but its probable arenite equivalent (Binda
and Mulgrew 1974) contains extensive mineralizations (Mufulira).
Figure 1 shows the location of the eight series of samples taken in the drill
holes and cross-sections in the Ore-Shale Alignment. Chingola and Rokana are
located in the carbonaceous facies of the Ore-Shale, whereas the other deposits
are in the littoral facies according to the distinctions made by Annels (1974, 1983;
Fig. 1). Table 1 presents concise information on the distribution of the 378
samples between the various formations and on the respective thickness sampled.
Detailed descriptions of the profiles and rocks are given by Guilloux (1982).
Chemical analyses were performed on each sample for major and trace elements
(including Cu, Pb, Zn and U) and an X-ray diffraction study, for the semi-quan-
titative determination of the mineralogical composition. The main data of the
chemical study of major elements are presented in this paper. The analyses were
carried out by standard techniques in the Societe Nationale Elf-Aquitaine
(SNEA) laboratories in Pau and some additional data were obtained at the
Centre de Recherches Petrographiques et Geochimiques in Nancy-Vandoeuvre
(see footnote, Table 1).
Although terms applied to sedimentary rocks are used here (siltstones, argil-
lites, etc.) the Zambian Copperbelt formations are distinctly metamorphic. In

Fig. 1. The Ore-Shale Alignment,

Zambian Copperbelt. The basemen
is shown as dashed. a, band c: re-
spectively, littoral, transition and
carbonaceous facies of the Ore-
Shale (after Annels et al. 1983).
The sampled districts are shown
with the corresponding index letters
in parentheses
446 B. Moine et al.

Table 1. Distribution of the samples from the Ore-Shale Alignment, Zambia (see also Fig. 1)

HW a ORE FW

Konkola (K) 22 (21.3) 15 (33)

Chingola (D) 24 (257.4)C 22 (18)b 17
(12.4)
Chambishi (J) 15 (15.4) 14 (16) 19
(54.5)
Mindola (M) 7 (36.2) 31 (29.6) 7 (5.5)
Mindola (MD) 3 (16.3) 28 (22.4) 7
(43.7)
Rokana (SB) 8 (36.3) 13 (11.9) 15
(55.6)
Luanshya (H) 10 (14.6) 22 (30.9) 7
(20.6)
Luanshya (SA) 11 (20.2) 25 (53.4) 19
(39.2)
Total 78 177 106

a HW, ORE and FW: Hangingwall, Ore and Footwall Formation, respectively.
b 22(18) means 22 samples along 18m, etc.
C Plus 17 (235) Upper Roan.

particular, the rocks of the Ore-Shale mainly occur as fine-grained micaschists

which may contain carbonates, talc or tremolite. Drysdall et al. (1972) have
shown that the grade of metamorphism increases from the biotite zone in the
northwest to the epidote-actinolite zone in the southeast. According to the vari-
ous mineral associations of mineral phases and some compositions of phengites
in the ubiquitous quartz-phengite-biotite-K. feldspar paragenesis (experimental
work by Massonne 1981 and Massonne and Schreyer 1983a), the approximate
metamorphic conditions can be defined as follows: P = 2 kb, T= 420°C,
X CO2 = 0.5 at Chingola and Konkola; P = 3kb, T = 460°C, X CO2 = 0.2 at
Chambishi and Rokana; P = 6.5 kb, T = 550°C, X C<>2 ~ 0.2 at Luanshya. The
medium to high pressure conditions observed at Luanshya are consistent with the
presence of Roan metasediments with talc, kyanite and phlogopite in the Solwezi
area, to the northwest of Zambia (Vrana and Barr 1972, Audeoud, unpublished
observations, experimental work by Massonne and Schreyer 1983b).

Magnesian Characteristics

Table 2 presents typical chemical compositions mostly of the rocks of the Ore-
Shale Formation (n° 3 to 11).
The Ore Shale Formation is divided, in places, into impure dolomite, ore-
shale and hangingwall argillite, but these units do not seem to correspond to
major geochemical and lithological variations. These divisions cannot be taken
into account in a very composite study. Table 2 shows that the rocks of the Ore-
Table 2. Some typical chemical compositions from the Ore-Shale Alignment"
a::
~.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Q
Wt070 K28 MS11 b K9 D55 D61 119 M08 M38 SA35 MS30 b MD5 7413 7339 FBl SH t!l

Si02 72.0 80.10 59.6 65.3 62.6 57.0 56.43 63.7 55.5 55.46 62.15 53.4 45.02 61.48 58.10 i
AI20 3 11.9 8.05 16.6 15.3 14.5 18.1 14.41 8.5 17.9 17.36 16.94 17.35 18.56 17.60 15.40
Fez03 t 1.4 2.81 3.7 4.3 2.8 4.1 3.7 2.4 6.1 5.0 c 2.48 4.51 6.10 3.65 6.72 d
MgO 2.2 0.36 6.3 1.7 7.5 5.6 2.31 10.2 7.0 4.54 3.23 9.20 8.92 1.40 2.44
CaO tr
l§.
1.13 0.5 tr tr 0.5 4.2 3.1 0.2 1.53 0.01 1.43 1.64 0.47 3.11 ~
Na20 0.2 0.50 0.2 0.1 0.2 1.0 0.72 1.3 0.4 0.34 0.41 0.12 0.66 0.22 1.30 .:a
K20 8.7 5.57 10.2 5.2 6.3 11.5 9.58 4.2 9.9 9.88 10.18 7.13 4.92 12.0 3.24 o
....
Ti02 0.7 0.21 0.9 0.9 0.8 0.9 0.8 0.4 0.9 0.63 1.12 1.05 1.01 0.65 :;-
n>
I.L 12 (1.03) 1.4 5.0 4.0 1.3 6.75 4.7 1.8 (2.89) 1.84 4.96 12.27 (8.68) ::c
Total 98.3 99.76 99.4 97.8 98.7 100.0 98.90 98.5 99.7 97.63 98.36 99.15 99.10 99.64 g
CO2 0.48 1.45 0.07 2.63 ~
~
"en
C 0.02 0.16 1.43 0.91 0.03 1.3 0.01 0.80
0.48 S·
H 2O 0.39 5.00 C/J
S 0.07 0.1 0.04 0.04 0.01 (2.7) 0.2 0.03 1.05 0.25
~
n>
" 1,2: Footwall Formation. 3 and 6 to 11: Ore-Shale Formation. 4,5: Lower Banded Shale. See Table 1 for provenance. 12, 13:: Argillites from evaporitic W
series (Triassic, Aquitaine Basin, France; in Jarousse 1978).14: Potassic shale from the Fucoi"d Beds, Scotland (Bowie et al. 1966).15: Mean composition of e:
the shales, after Clarke (1924). Analyses 1, 3, 4, 5, 6, 8 and 9 by SNEA (Pau). The quantity of Si, AI, Fe, Ca, K and Ti was determined by X-ray ~
fluorescence and of Na and Mg by atomic absorption after calcination at 1 050 °C and measurement of ignition loss. Total S was measured by combustion in ~o
the Leco furnace and total C(C02 + organic C) by combustion and coulometry. Analyses 7, 11, 12 and 13 by CRPG (Nancy) by automatic optical emission en
spectrometry with microwave plasma excitation. 8-
b MS11, MS30: Analyses from Luanshya quoted by Mendelsohn (1961). n
c Fez0 3 = 1.4070; FeO = 3.3%
d Fez0 3 = 4.02%; FeQ = 2.45% ~tl
.g
o
~.
Ci

£
448 B. Moine et al.

Shale are different from the common shales (analysis 15) by their frequently high
magnesium and very high potassium contents.
In sedimentary rocks, high Mg concentrations are related to carbonates
(dolomite when Mg is associated with Ca and more rarely magnesite) or to mag-
nesian clays (Mg-smectites, chlorite, corrensite, sepiolite). Various geochemical
and mineralogical studies (Trauth 1974, Kulke 1976, 1978, Jarousse et al. 1978,
Moine et al. 1981) showed that many evaporite series, in particular the carbonate
platform series such as the Triassic of Western Europe and North Africa contain
argillites and marls rich in magnesian clays and consequently showing very spe-
cific chemical compositions. For the study of metamorphic rocks, these chemical
characteristics are presented in the [AI- (Na + K)] - Mg - Ca diagram of Fig. 2, a
variation of the AI-Mg-Ca diagram of Moine et al. (1981) for the study of
feldspar rocks. Figure 2a shows the distinct difference between the compositions
of common sediments where the predominant clays are illites, montmorillonites
and kaolinite associated with calcite and dolomite (+ quartz + feldspars) and
those of argillites and marls of the evaporite series dominated by the association
of magnesian clays, illites and dolomite with a calcic tendency towards anhydrite.
Figure 2 b shows the distribution of all the samples of the Ore-Shale
Alignment. They are remarkably similar to the evaporitic series. Only the
Chingola samples, mainly those of the Lower-Banded Shale, equivalent here of
the Ore-Shale, correspond to predominant illite sediments (analysis 4, Table 2).
The location of a sample along the (AI- Na - K) - Mg line depends only upon
the relative quantity of the aluminous and magnesian clays and not upon the
total quantity of clays in the initial rock. Consequently, arkoses with low clay
proportion, buth with high chlorite/illite ratio are grouped with the chlorite-rich
argillites (for instance, rocks 1 and 3, Table 2 are projected on rather close
points). The importance of magnesian clay is very clear for the Ore-Shale (Table
2, analyses 12 and 13 of evaporitic argillites in comparison with analyses 3,5,6,8
and 9), and less so for most rocks of arenaceous origin. This evaporitic charac-
teristic is supported by the high Li concentrations (up to 600 ppm) noted in some
samples. At Mufulira, some of the analyses given by Malan (1964) of the argillite
associated with the stromatolite layer of inter BIC deposits reveal similar Mg-
rich compositions.
It has been accepted for many years that the sediments of the Lower Roan
were deposited in an evaporitic environment. Anhydrite was observed in
Mindola, Chambishi, Mufulira, etc. and Annels (1974) reported the formation of
sulphides from anhydrite. Other minerals such as magnesian tourmaline
(XMg = 0.63 and 0.74 in two analyses made by L. Guilloux) and scapolite also
provide evidence of an evaporitic heritage in metamorphic rocks; the high
quantity of albite in certain districts (Mindola, Mulfulira) may have the same sig-
nificance. The identification in the overall Ore-Shale Alignment of the mag-
nesian clay heritage, reveals the very general character of these evaporitic trends.
These aluminous magnesian clays probably were not formed by direct precipita-
tion, but rather by diagenetic transformation of illites in the basin.
 ~
cta ~.
o..,
tI1

\ "~
Cal c ite t Annydrlle
"- \
Ca
~
g.
-;.
""
ua rl z "'1- \ ~.
+
Ir
IAlb 11C " '\ cS> ::;-
'<
IK - F e l ds p a r '\~o \ o
-,
\ \ :;
(It

X
'\ \ g
® ~
\\ o
()

~
5'
CIl
o
.. 3
(It
K., o lln l l(
CIl
90A I- (N .,,·KI
....."
M o nt mor dlOnt l CS
.... ... :.. t
Corr
~ ensite . .-....
'".. . " . .. ., .AT , ~
Talc \"'_~!'_~' . .',,0,· X
M a gne s 1t c ,.' ... - •• 5l
, " t-'\'b
9\j\~.,
.. -' .'.,I ... .. 8-
10 cnloriles I, C, 50 '3C, (j
.g
~
Fig. 2 a, b. (Al- (Na + K» - Mg - Ca (milliatoms) for the identification of the (magnesian) evaporitic trend. a Evaporite argillites, marls and
carbonates (black triangles; data from Jarousse, 1978) compared with common sedimentary rocks (hatched; compilation by CRPG). Clay .g'='
minerals from Weaver and Pollard 1973. II CI , 13 CI and I 9 CI : illite/chlorite (mol) = 111, 3/1 and 9/1, respectively. b Ore-Shale Alignment ~.
of Zambia (density curves), RAT Formation from Shaba and Mansfeld district '"

t
'Cl
450 B. Moine et al.

Potassic Characteristics

Table 2 shows that the KP concentration of the Ore-Shale rocks frequently

exceed 100/0 which is quite unusual for rocks termed "argillites" and "siltstones"
(Pettijohn 1957). However, the Na20 concentrations are low except in some
albite-enriched rocks.
The Si/3 AI- (Na + K)/ Al diagram (Fig. 3) exhibits a distinct grouping of the
samples apart from the common sedimentary compositions. The initial sedimen-
tary rocks were quartz poor and very rich in K-feldspar. Only the Lower-Banded
Shale of Chingola is projected on the shale domain. The arenites (mostly the
Footwall Formation) contain feldspar-rich arkose compositions.
In the K/(AI- Na) - FM/(AI- Na) diagram (Fig. 4), quartz, carbonates
(except magnesite) and albite exert no influence and the rock distribution is
examined in comparison with the relative proportions of the various clays and K-
feldspars. The rocks studied here also show unusual compositions with a dense
grouping near the chlorite-K-feldspar line and only the Chingola rocks exhibit
predominantly illite normal compositions. Some samples are shifted towards
high FM/(AI- Na) values, probably because of magnesite. The arkosic rocks of
the Footwall Formation are grouped near the K-feldspar pole. However, to our
knowledge, the K-feldspar and magnesian chlorite association defined by the
original Ore-Shale rocks has no equivalent either in the common sediments or in
the evaporitic series. The argillites and marls of the evaporitic series often
contain, with magnesian clays, illites very rich in K and small quantities of
authigenic K-feldspar (Moine et al. 1981), but the K/(AI- Na) values seldom
exceed 0.5. The origin of the very particular composition of the Ore-Shale poses
a difficult problem requiring a more thorough discussion.

Sil3 AI
3

Fig. 3. Si13AI vs (Na+ K)/ Al (milliatoms) exhibiting the low quartz content and the high K-feldspar
content of the precursors of the Ore-Shale compared with common sedimentary rocks (cf. legend
Fig.2b)
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 451

K/(AI - Na)

....
K-Feldspar ' .
,,-

.,- -,
I ..

Talc

o. uar ..
+ ~arbonates
Alb i t e

,
/' '"
.
, ... -.... ".,. -9'" ".... FM/(AI-NA)
Montmorilionites __ ~_ ''1 ~~.,.:: - "-
.' Unit ,,- -Alra~ulgite~'\. . , -..... "" -~ Chloritcs

Fig. 4. K/(Al- Na) vs FM/(Al- Na) (milliatoms) showing the unusual K-feldspar-chlorite associa-
tion in the precursors of the Ore-Shale. FM: Fe - 0.5 S + Mg + Ca - C so that FM = 0 for pyrite,
calcite and dolomite (cf. legend Fig. 2b)

Origin of the Shale

The pre-metamorphic mineralogy of the Ore-Shale may be determined by distrib-

uting the elements between the following mineral phases: quartz, albite, K-feld-
spar, illite, chlorite and dolomite. The average composition given by Weaver and
Pollard (1973) was taken for the illite and the (Mg, Fe)sAI~i301O(OH)g formula
for chlorite. Table 3 shows the calculated results for some rocks. Very high K-
feldspar contents, low quartz and clay contents with high chlorite/illite ratios are
noted. Only the Chingola rock (D 55) has a mineralogical composition similar to
that of the average shales (SH)
The relative proportions of quartz, feldspars and clays in the Konkola,
Chambishi and Chingola Lower-Banded Shales samples are presented in Fig. 5.
The position of samples (numbers) of Table 2, the arkose domain and the posi-
tion of the greywackes and the shales are also plotted in this diagram. The rocks
of Chambishi and Konkola Ore-Shale fall in a specific domain away from the
common sediments. A high concentration of the samples may be noted around
an average point corresponding to 10070 quartz, 63% feldspar and 27% clay.
Although these rocks are not argillites, the term is used because of the presence
of fine-grained metamorphic rocks richer in phyllites (micas) and quartz than the
initial rocks.
No intense modifications of the chemical compositions took place during
regional metamorphism. With certain exceptions, the metamorphic recrystalliza-
452 B. Moine et al.

Table 3. Calculated mineralogical composition of the precursors of some Ore-Shale micaschists'

D55 H29 MD5 K9 SH

Illite 47.6 31.4 17.3 6.7 45.0

Chlorite 4.5 20.4 10.3 21.8 11.7
Albite 0.9 0.9 3.8 1.9 13.8
K-feldspar 12.8 30.5 55.6 59.2 2.0
Quartz 34.2 16.8 13.0 10.4 27.5
Clay min. 52.1 51.8 27.6 28.5 56.7
Feldspar 13.7 31.4 59.4 61.1 15.8

• These samples are quoted in Table 2 except H29, a sample from Luanshya. Illite to quartz = 100;
quartz to feldspar = 100.

(to· FELDSPAR

_ KONKOLA
cF.W . • ORE . H.W.A.
_CHAMBISHI
6 F.W • ORE . H.W.F.
_CH INGOLA
+ CARBONATES
o lower banded shale

.00

o o
c
OUARTZ so CLAY

Fig. 5. Calculated proportions of quartz, clay minerals (illite + chlorite) and feldspar in the pre-
cursors of the environment of the mineralization in some Ore-Shale districts, Zambia. Abbreviations:
FW: Footwall Formation; ORE: Ore Formation; HWA: Hangingwall Argillite; HWF: Hangingwall
Formation. Ow, SH: Mean composition of graywackes and shales, respectively. M: chemical
analyses from Mendelsohn 1961. 1 to 14: analyses from Table 2. A, B, C: see text

tions are mainly isochemical owing to the low porosity limiting the fluid quantity
and preventing their circulation. The characteristics described here are by no
means related to the metamorphic grade.
The Footwall Formation arenites are arkoses and conglomerates unusually
rich in K-feldspar (Fig. 5), and one may suggest that the K-feldspar of the Ore-
Shale is mainly detrital. But a sedimentary process has to be envisaged producing
a considerable K-feldspar enrichment in the fine-grained fraction with a feld-
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 453

spar/quartz ratio increasing from 1 to 6. This would be a trend contrary to that

generally occurring and difficult to explain. An argument invalidating this hypo-
thesis is given in the Ti0 2 - AIP3 diagram (Fig. 6). In the argillaceous and marly
series a correlation is observed between Ti0 2 which is mainly contained in the
clay minerals and AIP3 which measures the quantity of these minerals mixed
with varied quantities of quartz and carbonates. The trend determined by many
samples of the French Pyrenees PaleozoIc is given as an example (study in pro-
gress by B. Moine and F. Saupe, CRPG, Nancy). The rocks of the Ore-Shale are
located on the argillaceous trend and, for these elements, no considerable dif-
ference is observed between the Chingola rocks of known argillaceous origin and
the Ore-Shale of Konkola and Chambishi (and other districts) whose origin is yet
to be determined. The rocks of the Footwall Formation show a lower
Ti0 2 / AIP3 ratio, and the presence of 15070 clay and 20% feldspar instead of
35% quartz could not shift them in the Ore-Shale domain.
The above observations also do not corroborate the hypothesis of a conta-
mination by acidic volcanic material (cinerites) whose composition may be com-
pared to that of the Tuff Marker Beds, very rich in K-feldspar, describe by Crox-
ford (1964) in the Mount Isa deposit (Fig. 6). Moreover, no evidence of volcanic
activity was noted in the Zambian Lower Roan.
If the Ore-Shale is of argillaceous origin, its high K-feldspar content has
therefore been acquired after the sedimentation by reaction with solutions in
equilibrium with this mineral. During this transformation, Ti and Al may have
remained inert which is commonly the rule in metasomatic processes (Moine
et al. 1982). Illite, the dominant mineral of the initial shales of Chingola may
have reacted according to the following equilibrium:
KAI§i~IOlO(OHh+ 6Si0 2 + 2K + = 3 KAISi 30 s + 2H +
(muscovite for illite) (quartz) (K-feldspar)

Ao Tt~
0
•
K KONKOLA
10 0 CHINGOLA
J CHAMBISHI
Py PYRENEES

,-'\>'1 ....

05
"
,-
FOOTWALL ....

.J

.
~- T .M .B ...... j
.r '\>'1 "" I ~ - .. - - ---

-- --
"" , %AIPJ
l
~-

5 10 15

Fig. 6. Ti02 VS AI20 3 (wtOJo) for the Ore-Shale districts shown in Fig. 5. TMB: Tuff Marker Beds
from Mount Isa (Croxford 1964) (cf. legend of Fig. 5)
454 B. Moine et al.

This reaction consumes quartz and makes the mineralogical composition of

the argillaceous rocks evolve according to the B trend, parallel to C in Fig. 5.
Located in space between sedimentation and metamorphism, this transformation
would be diagenetic in the broad sense of the word. It may occur during early or
during burial diagenesis. The temperature is not determined, but hydrothermal
conditions are more consistent with the intensity of the transformations under-
gone than the low temperature conditions. The formation of the magnesian phyl-
lites discussed above is also diagenetic. As this phenomenon is widely developed
in the evaporitic series it is likely of early diagenesis and could not be contem-
poraneous with feldspathization.
Intense feldspathization of argillaceous rocks has seldom been described.
Moine (1974) has used the example of the Fucoi'd Beds of Scotland to explain the
origin of the feldspar-rich micaschists observed in the carbonate platform series
of the Upper Proterozoic of Central Madagascar. In the Fucoi'd Beds (Cambrian)
adularia-rich shales and potassic marls (analysis 14, Table 2) are intercalated
between quartzites and dolostones. Discussions on potassic characteristics have
permitted to compare the hypotheses of a volcanic or diagenetic origin. A sound
argument in favour of the diagenetic hypothesis was furnished by the determina-
tion of the crystallization age of the adularia (KI Ar datation) at around 100m. y.
after that of the sedimentation (Bowie et al. 1966, 1967). It is noteworthy that the
two above examples are situated like the Zambian Ore-Shale, in restricted
environments.
Finally, the chemical study of the Ore-Shale shows that these rocks cannot be
considered like ordinary, fine-grained, detrital sediments affected by regional
metamorphism. We agree on this point with Darnley's conclusions (1960) to
which specialists have not taken much notice. Without entering upon the metal-
logenic problems, our conclusion is in agreement with Annels' work (1974)
emphasizing the role of diagenesis in the formation of some sulphides.
In contrast, Guilloux (1982) has shown the Mansfeld Kupferschiefer to be
shales and calcareous-dolomitic marls with a mainly illitic argillaceous fraction.
Its relationship with the evaporitic conditions (some samples contain anhydrite)
does not particularly affect the geochemistry of the silicates (cf. domain of
Fig. 2 b) by comparison with the common platform series. Also the
Kupferschiefer is much more rich in organic carbon and, as underlined by
Wedepohl (1971), the Kupferschiefer of Germany and Poland is characterized
not only by high Cu grades, but also by high Zn and Pb grades exceeding 0.5%.
This is not the case in the Zambian Ore-Shale where Pb and Zn concentrations
around 50 to 100 ppm are observed.

The Shaban "RAT"

In Shaba, the RAT (argillaceous, talc-bearing rocks) forms the bottom of the
Roan (R) series. Francois (1974) distinguishes from bottom to top:
1. RAT Group (Rd formed of "micro-sandstones or hematite-bearing chlorite
pelites, lilac to purplish red" or red RAT. Its bottom is unknown. In Kolwezi,
Rl is identified over 235 m;
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 455

2. Mines Group (R 2 ) containing the copper and cobalt mineralization. It is com-

posed of two dolomitic formations enclosing a schistose-sandstone formation.
At the bottom of the first dolomitic formation, layer R 2.1.1. is composed of
rocks similar to Rt. but grey-coloured (grey RAT) where iron occurs in the
form of pyrite. The grey RAT are the host rocks of uranium mineralization
(Shinkolobwe, Kamoto, Kambove).
The separation between grey and red RAT is based on the presence of a
sedimentary conglomerate between the two formations. However, this separa-
tion is not constant. Cailteux (1977, 1983) shows that a continuous transition is
often observed (like in Kambove) with a variation of the deposition environment,
oxidizing (red RAT), then reducing (grey RAT).
The present study is concerned with the upper part of the RAT: 32 samples
from 42 m of red RAT and 3 - 5.6 m of grey RAT at Kamoko and Kambove-west
(see footnote, Table 4). These rocks mainly consist of chlorite, quartz and
dolomite. Grains of scaled or splintered quartz (Cailteux 1977) or with corrosion
pits (Audeoud 1982) are of probable volcanic origin. Chlorite forms a commonly
predominant matrix that is mainly sheridanite with X Mg around 1 (microprobe
analyses by D. Audeoud).
The chemical compositions are also very homogeneous and their variations
correspond to those of the quartz + chlorite and dolomite proportions (Figs. 2
and 3, Table 4). The three analyses presented in Table 4 clearly show the
dominant characteristics of the RAT. These are highly magnesian (chlorite and
dolomite), iron-poor and extremely alkali-poor rocks. The Li contents in the
most chloritic material are high (several hundred ppm). A cobalt anomaly (a few

Table 4. Some typical chemical composition of the "red RAT" from Kamoto (1) and Kambove (2, 3)

18 2 3 4b 5c
WtOJo RTl6 KB52 KB50 3535 ATl

Si~ 62.1 45.9 28.10 45.44 54.86

AI20 3 11.0 15.7 6.02 15.32 7.56
Fez0 3 t 5.9 1.6 0.52 5.45 3.69
MgO 13.5 20.8 20.99 14.19 10.22
CaO 0.2 0.9 14.81 0.34 0.62
Na20 tr tr 0.04 1.04 0.03
K20 tr tr 0.04 4.43 0.53
Ti~ 0.6 1.0 0.46 0.60 0.43
P20 S 0.09 0.05 0.17 0.44
I.L 6.5 13.0 27.67 11.70 20.1
Total 99.89 98.95 98.82 98.55 98.48
Li 390 666 157 320
Co 132 659 317 24 63

8 Analyses 1 and 2 by SNEA, the others by CRPG (see Table 2). The sample RTl6 is from the RT
drill hole (W 1328/Gecamines), at 38m under the grey/red RAT limit. The samples KB50 and
KB52 were taken from a cross-section in a gallery at Kambove-West, at respectively 0.1 and 5.7m
under the same limit.
b Argillite from evaporite series Triassic, Aquitaine Basin, France, in Jarousse (1978).
c Attagpulgite argillite (Eocene phosphate-bearing series, Senegal; sample M. Seyler).
456 B. Moine et al.

hundred ppm) is constant. Cailteux (1983) has published 15 new chemical

analyses of the RAT from Kambove. The two sets of data are very similar.
The highly magnesian character and prominence of Li are quite consistent
with our observations in the evaporitic series (Moine et al. 1981). Figure 2a and b
clearly shows that the RAT are located in the argillite and dolomitic marl domain
of the evaporitic series. However, a major problem is the nearly total lack of
alkalies contrasting with the abundance of K and the relatively high KI Al values
in the untransformed evaporite series (cf. analyse 4, Table 4). Various interpreta-
tions may be proposed.
In an isochemical hypothesis, clay and attapulgite marls as those identified in
the CenozoIc cover of the West African Shield (analysis 5, Table 4) may be con-
sidered as the initial sediment. Several observations contradict this interpreta-
tion. The attapulgite (with sepiolite) is mainly associated with calcite in marls
thus showing a trend different from that observed in Fig. 2 b. These rocks
contain low, but appreciable KzO contents (analyses 5, Table 4), whereas most
RAT show no KzO at all.
Above all, attapulgites are definitely richer in Si0 2 than chlorites. Their meta-
morphic recrystallization should lead to the formation of chlorite + quartz
rocks. But the Si - AI- (Mg + Fe) diagram of Fig. 7 shows that the RAT, in spite
of the contribution of detrital quartz, are not as siliceous as attapulgites and
palygorskites. The argillaceous fractions which originated the RAT should be
less siliceous than these rocks and should necessarily be plotted near the chlorites
(Fig. 7).
The presence of few mineral phases (here, mainly three) and very low concen-
trations of mobile elements such as the alkalies are often the signature of
exchanges between fluids and minerals (metasomatic alteration in the broad
sense of the word). This confirms the two hypotheses put forward to explain the
formation of the RAT: chloritization of detrital illite (Katekesha 1975) and
chloritization of fine-grained pyroclastic material such as cinerites (Cailteux
1977,1983).

\
PALYGORSKIT E S

SEP IOLITE «-,11 Fig. 7. Si-Al-(Mg+Fe) (milliatoms) showing

the discrimination between the RAT from Shaba
and attapulgites-palygorskite compositions
(circled and ATI, Table 4). Each rock from the
RAT is plotted as a large dot (assuming primary
dolomite) and also as an arrowhead (assuming
primary calcite). Triangles: some chlorite compo-
sitions
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 457

The chloritization of white micas may form chloritites in hydrothermal condi-

tions. However, this transformation should be evidenced by structures, such as
zonings intersecting the bedding. This chloritization is accompanied by a con-
siderable increase of the volume of the mineral which may be compensated by the
dissolution of some quartz (Moine et al. 1982). On the contrary, the petrographic
observations show well-preserved detrital quartz. The formation of aluminous,
magnesian phyllites (chlorite, corrensite) from detrital illites may occur during
early diagenesis as is the case in the evaporitic environments. However, as we
have already seen, this transformation seldom produces only chloritic rocks with
the total leaching of potassium. As a rule, high concentrations of K are related to
the presence of illites or authigenic feldspar.
Cailteux (1977, 1983) suggests that the top of the RAT is of pyroclastic origin.
He bases his hypothesis in particular on the observation of pyroclastic material in
certain parts of Shaba (Lefebvre and Cailteux 1975). The precise nature of this
volcanism is yet to be determined, but the presence of quartz and low Fe, Ti0 2
(usually from 0.3 to 0.6OJo) Cr, etc. contents suggests acidic volcanism. The
chloritization of the feldspathic ashes may have taken place in a sedimentary en-
vironment (analogy with the "Tonstein") or under hydrothermal conditions
during burial diagenesis. In this case, the moderate volume increase (13% for K-
feldspar) may have been compensated by the partial dissolution of the quartz.
The RAT have low Si0 2 contents, usually lower than 60%. The Si0 2 contents
may have been fairly higher (up to 70%) in the initial quartz-feldspathic material.
Audeoud (1982) presents the results of the quantitative analyses of REE by
neutron activation in six RAT samples. The spectra are varied, contrasting with
the rock mineralogical and chemical homogeneity. It seems that mobilization
may have taken place between different layers and that REE may have been
trapped in the monazite. The nearly total absence of an europium anomaly does
not confirm the feldspathic heritage (plagioclase). The spectra which appear to
be the less disturbed are rather identical to those of the shales. However, no clear
conclusion can be definitely reached on the origin of the RAT.
Whatever the precise origin of the RAT - argillaceous or volcano-sedimen-
tary - their mineralogical and chemical compositions have probably been
achieved in a restricted environment. This is in agreement with petrographic
observations such as dolomitic "patches" which are explained by the transforma-
tion of anhydrite nodules (Cailteux 1977, 1983).
Audeoud (1982) carried out a study by microthermometry and by Raman
spectroscopy with the MOLE microprobe of the fluid inclusions contained in the
dolomitic veinlets of the RAT and the identical inclusions associated with the
anchimetamorphic recrystallization. They are mainly aqueous with very high
salinities i.e. more than 60% of dissolved salts (Pirmolin 1970, has also shown
high salinities in the Kamoto dolomite). These salinities are mainly due to MgClz
and CaCI 2 , whereas the NaCI and KCI concentrations are comparatively low.
The origin of these brines probably lies in the series itself. They are quite consis-
tent with the RAT composition characterized by chlorite, dolomite and the
absence of alkali minerals. The metallogenic interest of these observations need
not to be emphasized.
458 B. Moine et al.

Conclusions

The approach in this study has been to define the primary mineralogical com-
position of metasediments from their chemical composition, thus revealing two
main features of the African Copperbelt:
1. The rocks which host the mineralization were deposited in an early evaporitic
environment of great extent whose signature is former magnesian clays.
2. Characteristics related to the alkali: high amounts of K in the Zambian Ore-
Shale, and in contrast, very low K and Na concentration in the Shaban RAT,
define original mineralogical compositions which do not correspond to sedi-
mentary or volcano-sedimentary material. They are very likely the signature
of pronounced diagenetic or hydrothermal transformations. The role played
by the reaction between solutions and minerals is supported in the RAT by the
observation of fluid inclusions whose composition is consistent with the min-
eralogical characteristics.
The role of the sedimentation environment and the post-sedimentary process
in the formation of the silicates is thus demonstrated and may contribute towards
a better understanding of the formation conditions of the associated sulphide
minerals. With this in mind, further studies should be carried out to determine
the precise relations between the sulphide content and the geochemistry of meta-
sediments at various levels of observation. Variations of the chemical charac-
teristics in unmineralized zones should be more thoroughly investigated than they
have been so far. The major variations observed in the compositions of the
Shaban RAT and formations of the same age in Zambia have to be explained.
They may be due partly to paleogeographic factors and partly to the nature of the
materials which underwent diagenetic transformation (for example, the role of
volcanic material in Shaba). These lithologies are likewise quite distinct from the
Kupferschiefer of northern Europe.
Finally, this study strengthens the view that the African Copperbelt deposits
can no longer be regarded as strictly syngenetic sedimentary deposits. In spite of
their apparent control by sedimentary structures, they are most likely related to
diagenetic or hydrothermal processes involving evaporitic brines and an
unknown source of metals.

Acknowledgements. Full facilities for sampling were given to L. Guilloux by the mining companies:
Nchanga consolidated Copper Mines Limited and Roan Consolidated Mines in Zambian; Gecamines
in Zaire; Kombinat of Sangerhausen in G.D.R. The study by L. Guilloux has received financial
support by the Delegation Generale Ii la Recherche Scientifique et Technique and by the Bureau de
Recherches Geologiques et Minieres and technical support by the Societe Nationale Elf-Aquitaine
and by the Centre de Recherches Petrographiques et Geochimiques, Nancy. The study by D.
Audeoud was supported by the Centre de Recherches sur la Geologie de l'Uranium, Nancy. The
manuscript was prepared with the assistance of the Laboratoire de Mineralogie, Toulouse. Mrs. J.
Guilloux is thanked for the translation of the text into English.
Major Element Geochemistry of the Host Rocks in Some Sediment-Hosted Copper Deposits 459

References

Annels AE (1974) Some aspects of the stratiform ore deposits of the Zambian copperbelt and their
genetic significance. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Soc
Geol Belg, Liege, pp 235 - 254
Annels AE, Vaughan DJ, Craig JR (1983) Conditions of ore mineral formation in certain Zambian
Copperbelt deposits with special reference to the role of cobalt. Miner Depos 18:71 - 88
Audeoud D (1982) Les mineralisations uraniferes et leur environnement a Kamoto, Kambove et
Shinkolobwe (Shaba, ZaIre). Petrographie, geochimie et inclusions fluides. These 3eme Cycle,
Lyon, 211 pp
Bartholome P (1974) On the diagenetic formation of ores in sedimentary beds with special reference
to Kamoto, Shaba, ZaIre. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes.
Liege, Soc Geol Belg, Liege, pp 203 - 213
Binda PL, Mulgrew JR (1974) Stratigraphy of copper occurrences in the Zambian Copperbelt. In:
Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Soc Geol Belg, Liege, pp
215-233
Bowie SHU, Dawson J, Gallagher MJ, Ostle D (1966) Potassium-rich sediments in the Cambrian of
Northwest Scotland. Trans Inst Ming Metal GB 75, 714:125 -145
Bowie SHU, Dawson J, Gallagher MJ, Ostle D (1967) Potassium-rich sediments in the Cambrian of
Northwest Scotland. Report of discussion at Nov. 66, general meeting, contributed remarks and
author's reply. Trans Inst Ming Metal GB 76, 723:60-69
Cailteux J (1977) Particularites stratigraphiques et petrographiques du faisceau inferieur du Groupe
des Mines au centre de Parc cuprifere shabien. Ann Soc Geol Belg 100:55 -71
Cailteux J (1983) Le "Roan" shabien dans la region de Kambove (Shaba, ZaIre). Etude sedimentolo-
gique et metallogenique. These, Univ Liege
Croxford NJW (1964) Origin and significance of volcanic potash-rich rocks from Mount Isa. Trans
Inst Ming Metal GB 74:33 - 43
Darniey AG (1960) Petrology of some Rhodesian Copperbelt orebodies and associated rocks. Trans
Inst Ming Metal GB 69:137-173,371-398,540-569
Drysdall AR, Johnson RL, Moore TA, Thieme JG (1972) Outline of the geology of Zambia. Geol
Mijnbouw 51 :265 - 276
Fleischer VD, Garlick WC, Haldane R (1976) Geology of the Zambian Copperbelt. Handbook of
stratabound and stratiform ore deposits. In: Wolf KH (ed) Elsevier, New York, 6:223 -352
Francois A (1974) Stratigraphie, tectonique et mineralisations dans PArc cuprifere du Shaba
(Republique du ZaIre). In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Soc
Geol Belg, Liege, pp 79 - 101
Garlick WG, Fleischer VD (1972) Sedimentary environment of Zambian copper deposition. Zambia
issue. Geol Mijnbouw, 51 :277 - 298
Guilloux L (1982) Etude chimique des series porteuses de quelques grands gisements de type Kupfer-
schiefer. Consequences metallogeniques. These d'Etat Lyon. Sciences de la Terre, Nancy, Mem 43
Guilloux L, Pellisonnier H (1974) Les gisements de schistes, marnes et gres cupriferes. In: Gisements
stratiformes et provinces cupriferes. Soc Geol Belg, Liege, pp 35 - 53
Hewitt DA (1973) Stability of the assemblage muscovite-calcite-quartz. Am Miner 58:785 -791
Hoschek G (1973) Die Reaktion Phlogopit + Calcit + Quartz = Tremolit + Kalifeldspat + H2 0
+ C~. Contrib Miner Petrol 39:231 - 237
Hoschek G (1980) Phase relations of a simplified marly rock system with application to the Western
Hohe Tauern (Austria). Contrib Miner Petrol 73:53 - 68
Jacobs GK, Kerrick DM (1981) Devolatilization equilibria in H2 0-C02 and H2 0-C02 -NaCI
fluids: an experimental and thermodynamic evaluation at elevated pressures and temperatures. Am
Miner 66:1135 -1153
Jarousse J (1978) Contribution au probleme de I'identification d'anciennes series evaporitiques dans
les ensembles metamorphiques. Approche principalement geochimique. These 3eme Cycle, Univ
Claude Bernard, Lyon-I, 232 pp
Jarousse J, Moine B, Sauvan P (1978) Etude chimique de series evaporitiques en vue de leur
identification dans les ensembles metamorphiques. CR Acad Sci Ser D 286:1057 -1060
Katekesha WM (1975) Conditions de formation du gisement cuprocobaltifere de Kamoto Principal
(Shaba, ZaIre). Mem Doct, Fac Sci Univ Liege, 237 pp
460 B. Moine et al.: Major Element Geochemistry of the Host Rocks

Kerrick DM (1974) Review of metamorphic mixed-volatile (H2 0-C02 ) equilibria. Am Miner

59:729-762
Kulke H (1976) Die salinare Trias des Atlassystems (Nordwest Afrika). Faziesverteilung, Tektonik,
Morphologie, Petrographie, Mineralogie und Geochemie, 213 S. Habil Schr Fak Geowiss, Univ
Bochum (unpublished)
Kulke H (1978) Tektonik und Petrographie einer Salinarformation am Beispiel der Trias des Atlas-
systems (Nordwest Afrika). Geotekt Forsch, 55 I-II:1-58
La Roche H de (1965) Sur l'existence de plusieurs facies geochimiques dans les schistes paleozoiques
des Pyrenees Luchonnaises. Geol Rundsch Dtsch 55:274 - 301
La Roche H de (1968) Comportement geochimique differentiel de Na, K, et AI dans les formations
metamorphiques et plutoniques. CR Acad Sci Ser D 267:39-42
La Roche H de (1974) Geochemical characters of the metamorphic domains: survival and testimony
of their premetamorphic history. Sci Terre, T XIX, 2:101 -117
Lefebvre JJ, Cailteux J (1975) Volcanisme et mineralisations diagenetiques dans Ie gisement de
l'Etoile, Shaba, Zaire. Ann Soc Geol Belg, Liege, 98(1):177 -195
Malan SP (1964) Stromatolites and other algal structures at Mufulira, Northern Rhodesia. Econ Geol
59:397 -415
Massonne HJ (1981) Phengite: eine experimentelle Untersuchung ihres Druck-Temperatur-Verhal-
tens im System K2 0 - MgO - Al2 0 3 - Si~ - H20. Dissertation, Ruhr-Univ Bochum
Massonne HJ, Schreyer W (1983a) A new experimental phengite barometer and its application to a
variscan subduction zone at the southern margin of the rhenohercynicum. 2nd E.U.G. meeting,
Strasbourg, 1983. In Terra Cognita, 3:187
Massonne HJ, Schreyer W (1983b) Stability of the talc-kyanite assemblage revisited. 2nd E.U.G.
meeting, Strasbourg, 1983. In Terra Cognita, 3:187
Mendelsohn F (1961) The geology of the Northern Rhodesian Copperbelt. Mac Donald, London,
523 p.
Moine B (1971) Caracteres de sedimentation et de metamorphisme des series precamrbiennes
epizonales a catazonales du centre de Madagascar (region d' Ambatofinandrahana). Sci Terre Fr
Mem 31:293
Moine B, Sauvan P, Jarousse J (1981) Geochemistry of evaporite-bearing series: a tentative guide for
the identification of metaevaporites. Contr Mineral Petrol 76:401 - 412
Moine B, Gavoille B, Thiebault J (1982) Geochimie des transformations metasomatiques a l'origine
du gisement de talc et chlorite de Trimouns (Luzenac Ariege, France). I. Mobilite des elements et
zonalite. Bull Mineral 105:62-75
Pirmolin J (1970) Inclusions fluides dans la dolomite du gisement stratiforme de Kamoto (Katanga
Occidental). Ann Soc Geol Belg 93(2):397 -406
Petitijohn FJ (1957) Sedimentary rocks. Harper and Row, New York, 718 p
Puhan D, Johannes W (1974) Experimentelle Untersuchung der Reaktion Dolomit + Kalifeldspat
+ H20 = Phlogopit + Calcite + CO2 , Contr Miner Petrol 48:23 - 31
Renfro AR (1975) Genesis of evaporite-associated stratiform metalliferous deposits - a sabkha
process. Econ Geol 69:33 - 45
Rentzsch J (1974) The Kupferschiefer in comparison with the deposits of Zambian Copperbelt. In:
Gisements stratiformes et provinces cupriferes. Soc Geol Belg, Liege, pp 395 -418
Shaw DM (1956) Geochemistry of pelitic rocks. Part III: Major elements and general geochemistry.
Geol Soc Am Bull 67:919-934
Skippen G (1974) An experimental model for low pressure metamorphism of siliceous dolomitic
marble. Am J Sci 274:487 - 509
Slaughter J, Kerrick DM, Wall VJ (1965) Experimental and thermodynamic study of equilibria in the
system CaO-MgO-Si02 -C02 -H20. Am J Sci 275:143 -162
Trauth N (1974) Argiles evaporitiques dans la sediinentation carbonatee continentale et epicontinen-
tale tertiaire. Sci Geol Fr Mem 49:195 p
Vrana S, Barr MWC (1972) Talc-kyanite-quartz schist and other high-pressure assemblages from
Zambia. Mineral Mag 38:837
Weaver CE, Pollard LD (1973) Developments in sedimentology. The chemistry of clay minerals.
Elsevier, Amsterdam, 213 p
Wedepohl KH (1971) "Kupferschiefer" as a prototype of syngenetic sedimentary ore deposits. Soc
Mining Geol Japan, Spec Issue 3:268-273
Zechstein Copper-Bearing Shales in Poland.
Lagoonal Environments and the Sapropel Model of Genesis
C. HARANCZYK 1

Abstract

The palaeogeographic situation of the Fore-Sudetic Monocline during the first

Zechstein evaporative subcycle, evolution of the metalliferous sedimentation and
subsequent subaerial geological processes are illustrated on a redox map, geo-
chemical profiles and diagram. The sequence of the sedimentation of the seaward
and lagoon ward slopes of the lagoonal barrier explains the differences between
copper- and lead-bearing shale sedimentation environments. Subaerial abrasion,
redeposition, oxidation, cementation and lagoonal brine infiltration contributed
to the redistribution of metals in the just-formed sediments before Werra main
dolomite sedimentation closed the first evaporative subcycle.

Introduction

Kupferschiefer as a stratigraphic term has for years been used in the Mansfeld
Ore District in Germany. According to Chronica Mansfeldensi written by Span-
genberg in 1572, mining operation (Bergbau) started in 1199 in this ore district.
However, the first geognostic modern descriptions we owe to Freiesleben (1815),
a pupil of the Werner school. These were followed by papers of several genera-
tions of geologists; let us mention only SchneiderhOhn (1921), Marowsky (1969),
Rentzsch and Knitzschke (1968) and Wedepohl (1971), namely, authors who
introduced new methods of laboratory investigations, and created the concept of
Kupferschiefer as a prototype of the sedimentary ore bed.
The discovery of new copper-rich areas on the southern margin of the Zech-
stein basin in Poland has greatly stimulated research on these sedimentary ore
deposits. The Fore-Sudetic Monocline in Lower Silesia is now not only the largest
European copper deposit, producing more than 350000 tons of copper and 700
tons of silver yearly, but it is the area of most complex and holotypical develop-
ment of sediments. Therefore, it must also be regarded as Locus Classicus, the
prototype of euxinic sedimentary ore beds.
Results of the recent investigations carried out here substantiate a revision of
some traditional concepts concerning this prototype. This review aims to high-

I Institute of Geological Sciences, Jagellonian University, 2a Oleandry St., Krakow, Poland

Geology and Metallogeny of Copper Deposits

(ed. by G.H. Friedrich et at.)
© Springer-Verlag Berlin Heidelberg 1986
462 c. Harailczyk
light some revised aspects of the genesis of these sediments, indeed, also to
explain the genesis of the richness of this unique Zechstein ore district. These
aspects are: Kupferschiefer as a sediment formed under regressive conditions of
the first Zechstein subcycle, the development of euxinic sediments in the lagoonal
area, the enrichment processes due to subaerial exposure, and finally, the
sapropelic model of genesis acting in lagoonal environment.

Early Zechstein Sea

The Zechstein Sea was formed when the northernmost rim of a chain of
interlinked sub-sea level inland drainage basins was breached or overflown by a
relative rise of sea level (Smith 1980), thus creating an inland sea, in places
initialy more than 250 m deep. It is contended that the rim remained as a barrier
near sea level and exerted a strong influence on the subsequent depositional
history of the whole basin (Smith 1980). A complex Zechstein inland drainage
basin the bottom of which lies well below contemporary sea level was thus
immediately established, drowning existing Lower Permian red sandstone desert
deposits, including aeolian sand dunes and rocky hills as well as the extensive
plains of the basin floor and widespread marginal peneplains (Smith 1980). The
Zechstein inland basin overlapped and in much of the area it might also be
genetically related to the foreland rift basins formed after Hercynian orogenic
collision and filled with Rotliegend sediments. Lorenz and Nicholls (1976) and in
this year also Jowett (1984) published an excellent plate-tectonic explanation of
the taphrogenic Permian evolution of Central Europe. The explanation is
important also for the discussion of the provenance of metals for the
metalliferous Zechstein sediments.

Palaeogeography of the Fore-Sudetic Monocline

The ingression of the Zechstein Sea came from the northwest along the line
Glogow-Polkowice (Fig. 1). Shortly after ingression and stabilization of the sea it
extended about 70 km farther toward Wroclaw, as is evidenced by a thin bed of
basal limestone (Eisentraut 1939, Neuhaus 1942, Rydzewski 1969). Formation of
the Zechstein metalliferous sediments is associated with the first regression of the
sea. A lagoonal topography was created when the sea retreated to the north,
leaving the Rudna lagoon, a set of barrier reefs around it (see Fig. 1) and a
narrow seaway from Gtogow to just west of Polkowice extending toward
Szklary. Three main Palaeotopographic wnes may be distinguished here (Fig. 1).
These are the western highs of oxidized sediments (Rote Faule) extending
westward from Sieroszowice (Rydzewski 1969, Haranczyk 1970, Oszczepalski
1982), the narrow Glogow-Polkowice-Szklary depression which ran southward in
the central part of the area, and the eastern coastal elevation extending eastward
from Lubin (Klapcinski 1964). Within this area a lagoon barrier separated the sea
from the adjacent northern marginal basin which because of increasing regres-
Zechstein Copper-Bearing Shales in Poland 463

GtOG6w

Cu Cu

RUONA

'A O+A

o1
2
I
~
,
km

9 Ore Flats
- - ' - Lagoon Barrier
~Bom'er Reef
- - - -Submarine Bar
D+A Do lomite Anhydrite
Fe"" Oxidized zone
• Town

Fig. 1. Palaeogeographic situation of the Fore-Sudetic Monocline during the first Zechstein
regression subcycle

sion and isolation evolved into a lagoon, named Rudna Lagoon after the main
locality (Figs. 1 and 2; see also Haranczyk 1970). More than 150 km 2 of this
lagoon have been explored by drilling, however, its eastern extension is still
unknown (Fig. 3). The southwestern marginal part of this lagoon is accessible in
the stopes of the copper mines. Apart from the main lagoonal barrier represent-
ing a zone of sand bars and elevations topped by reef material, some sandstone
bars which are also partly covered with a flat reef and are free from any copper-
bearing shales, due to erosion and reworking of the previously existing shales, are
recognized in the southwestern foreground of the lagoon (Fig. 4) . These bars did
not separate any isolated basin and are surrounded by flats covered by copper-
bearing sediments. Only the lagoonal barrier separates areas of the copper- and
lead-bearing shales. In the lagoon, the lead-bearing shales are underlain by a 2-
or 3-cm-thick layer of the copper-bearing shales formed at an early stage of
regression before isolation of the Rudna Lagoon and evaporative regime, and
also before stratification of the lagoonal brines was established. Rudna Lagoon
was a deeper lagoon than the shallow lagoonal pans at Mansfeld depicted by
Rentzsch (1974).
464 C. Haraitczyk

•
GtOG6w

EXPLANATION
O.....!...] km

- Lagoon Ba r rier
-- - Submarine Bar
l Lead lxlaring Sha l e s
TTII Erasion of Zechste i n b~ds
• Town

Fig. 2. Geochemical redox map of the Fore-Sudetic Monocline

Rudna Metallogenic Subcycle

The first Zechstein evaporative subcycle began shortly after ingression of the sea
and it is included in the Werra cyclothem. The most complete evaporative
sediment sequence is known from the locality of Rudna. Ingression of the
Zechstein sea is marked in places by basal conglomerates, and by basal limestone
or dolomite formed under oxidizing conditions. These sediments delimit the
largest extension of the Zechstein Sea (Eisentraut 1939, Klapcibski 1964,
Rydzewski 1969). They are followed by sediments formed during stagnation
associated with growing regression of the sea. Let us consider data substantiating
the existence of the first evaporative subcycle:
Zechstein Copper-Bearing Shales in Poland 465

I ~ z[]]]l) 3
Fig. 3. Geological sketch map of Fore-Sudetic area according to J. Krason and J. Sokolowski,
showing strata older than Cenozoic. 1 Cretaceous; 2 Triassic; 3 Zechstein; 4 Lower Permian red beds
(Rotliegend); 5 Palaeozoic; 6 Precambrian; 7 Hercynian igneous rocks; 8 dislocatiuns

1. The shoreline marked by transgressive basal limestone (locally dolomite) was

shifted almost 70 km westward in the flat peneplain area between Wroclaw
and Lubin.
2. According to Smith (1980) in England, marl slate with high metal content is
restricted to near-shore deposition in steep relief, however, in a flat marginal
peneplain, it is 40 - 50 km distant from the extensive shoreline.
3. A typical sub cycle sediment sequellce ranges in thickness from 2 m on the
internal slopes of the lagoonal barrier (Figs. 5 and 8) to almost 20 m in the
central part of Rudna Lagoon (Fig. 9). The Rudna sub cycle sediment
sequence within the lagoonal area includes shales, marly dolomites and
dolomites with anhydrite and limestones with anhydrite from bottom to top;
and it is a typical regression sequence with increasing evaporation. The
sub cycle sediment sequence is overlain by stylolithic dolomite. The sediment
sequence in the open sea was described by Beres et al. (1973).
4. Preserved lagoonal relief of more than 30 m is observed in the mine stopes of
the Lubin, Polkowice and Rudna mines in the Fore-Sudetic Monocline. The
466 C. Haranczyk

o
5-220

L s .t2J
o

Explanation

- - Borrier

== = :
5 -267
o
Sandbar o as II,,..

Fig. 4. Fragment of the map shown in Fig. 2 displaying area of subaerially exposed shales con-
sequently enriched in pyrite

difference is measured between the flat regressive reef capping the lagoonal
barrier and the tidal flats in the fore-shore which in their shallower parts are
free from shales, due to abrasion, and in the lower part are covered by the
copper-bearing shales burrowed by scavengers, and finally the lowest shore-
line marked by lenses of redeposited copper and organic-rich shales. Also the
thickness of the mineralized zones in the sandstones indicates a ca. SO-m-high
barrier in the southeastern part of the lagoon (Haranczyk 1970).
Taking into consideration the above-mentioned data, the fall of the sea level
during the Rudna subcycle is estimated to have been at least 70 m. It is highly
probable that it prevailed in the whole Zechstein basin. The Rudna subcycle
lasted a few hundred thousand years as inferred from the maximal thickness of
the lagoonal sediment sequence consisting mainly of seasonal rhythmic sedi-
ments.
Zechstein Copper-Bearing Shales in Poland 467

w E
82 5-86 5-11L 5- 215 5-208 5 - L18
5-116 5-120

/ \ I \ / ./ \ / \

.....
.. : .. :. ",

. '.:,' .'
RO LO
mJ[LZ]E3EHlimB[§]8ElD~~rn
2 3 L 5 6 7 8 9 10 11 12 13

Fig. S. Distribution of sediments around the Rudna lagoonal barrier. 1 Reef dolomite; 2 Stylolitic
dolomite; 3 dolomitic limestone with anhydrite inlayers; 4 dolomite with anhydrite nodules;
5 pyrite-bearing argillaceous dolomite; 6 lead-zinc-bearing argillaceous dolomite; 7 redeposited
copper-rich (smearing) shale; 8 copper-bearing shale (Kupferschiefer); 9 bioturbidite sandstones
hosting detrital biota of the reef talus; 10 Weissliegend sandstone; 11 the same with sulphide diffusion
laminae; 12 zone of sandstones with matrix replaced by anhydrite-hosting nests of massive ore;
13 regenerated and/or dolomitized sandstone

Evolution of the Sedimentation During the First Subcycle

The lagoonal relief of the surface of the Weissliegend sandstones in the Fore-
Sudetic Monocline was only partly inherited from desert topography. Sub-
aqueous formation of sandstone bars is suggested for most of the Weissliegend in
the area (Krason and Grodzicki 1965, Konstantynowicz 1965, Metzler 1975,
lerzykiewicz et al. 1976). An influence of synsedimentary tectonics is also
probable as the southern barrier runs parallel to the Sudetian tectonic direction.
During a short period of sea-level stability shortly after transgression, the top of
the lagoonal barrier and some other bars were in the euphotic zone and active
reef growth, locally with a stenohaline fauna, flourished. The reef is fairly flat -
rarely more than 2 m thick because of the regression of the sea caused the top of
the reef to approach sea level, limiting upward growth and encouraging lateral
expansion. Bivalves and crynoids are dominant, stromatolite structures are also
seen.
The next stage of the subcycle is marked by a sharp fall of sea level leaving all
reefs subaerially exposed. Separation of the Rudna Lagoon was established and
early dolomitization of the reef by refluxing lagoonal brines began.
In the subsequent stage a continuing fall of sea level initiated the growing
isolation of the lagoon, and the ensuing abrasion and washing out of the copper-
bearing shales lying on the steep, seaward slopes. Redeposition of the dis-
integrated material produced copper- and organic-rich shales (smearing shales),
468 c. Haranczyk

forming lenses along the lowest shoreline. Weathering of the exposed shales was
associated with cementation of the barrier sandstones by descending waters
(compare the barrier cross-section, Haranczyk 1970). Tidal flats in the fore-shore
of the barrier are covered with shales burrowed by scavengers. Cementation of
these shales has produced ore flats, a unique ore mineral form (Haranczyk
1984a). The burrowed shales from the tidal flats are also extremely rich in fish
fossils. Parenthetically, it should be noted that Freiesleben (1815) was right in
saying that the fish died of poisonous copper concentrations in the seawater,
however, it was true only for the zone adjacent to the supratidal flats where the
oxidation of the subaerially-exposed shales proceeded. Deposition of lead- and
zinc-rich shales in the Rudna Lagoon on the internal barrier slopes and of
nodular carbonates with intercalations of anhydrite in the central part developed,
indicating that lagoonal brines were already strongly concentrated and stratified.
During the lowest sea level, total isolation of the lagoonal basin produced
brines which invaded barrier sandstones and reef carbonates, replacing the
matrix and in places whole sandstone beds by anhydrite or gypsum and dolomite.
The replacement front is marked by regeneration of quartz grains and is followed
by massive concentrations of chalcocite, forming the richest ore in the district.
Full recovery of the sea level and uniform sedimentation of a beige stylolitic
dolomite or dolomitic limestone brought to a close the first regression and the
Rudna subcycle of the Zechstein Sea in the Fore-Sudetic Monocline (Figs. 5 and
6). Because the relief was not completely compensated by sedimentation, the
barrier formed a cupola and steplike faults developed along it.
The above presented evolution precludes a sabkha model suggested by Renfro
(1974) and others discussed by Lagny (1980).

Subaerial Ore Processes

Emergence of the metalliferous shales and prolonged contact with atmospheric

oxygen induced ore processes important for the redistribution of metals and their
concentration or dispersion. In the Fore-Sudetic Monocline, this process
developed on a large scale depending on the exposure time and contrast of the
relief. The following ore processes are recognized as important for metal concen-
tration:
1. Wave abrasion and washing out of the shale and sandstone beds lying on the
steep seaward slopes of the lagoonal barrier and on the slopes of submarine
bars causing redeposition (Blaszcyk 1982) and formation of lenses of shale
along the contemporaneous shoreline (Figs. 5 and 7). The redeposited shale
consists mostly of disintegrated shale matter which lost its rhythmic sedimen-
tation structure and sometimes its organic carbon and copper contents,
ranging up to 200/0. The disintegrated shale matter fills also scavenger trails.
2. Short, subaerial exposure of copper-bearing shales induces their partial oxida-
tion and the migration of metals mainly to the basal layers producing
enrichment in pyrite (Fig. 4) and formation of a contact layer of pure chalco-
Zechstein Copper-Bearing Shales in Poland 469

I I
55 o TF R I E5 I L5
I I

55 R L5 CD
.~
o <lI
.s::. c:
a. 0
:J N
W

800 600 l.00 200 a 200 l.00 600 800

m
Fig. 6. Evolution of sedimentation in the zone of lagoonal relief during the first regression of the
Zechstein Sea. Symbols of sedimentation zones: SS sea sedimentation; D redeposition of sediments;
TF tidal flats; R barrier reef; ES lagoonward slope of the barrier; LS lagoonal sedimentation;
1 - 4 stages of evolution. D A dolomite with anhydrite nodules

cite (Fig. 10). Covellite, digenite, native silver and rare cobalt minerals
frequently contribute to this layer (see also Haranczyk 1972).
3. Advanced oxidation of shales and descendent migration of solutions is the
most frequent cause for the genesis of the sandstone ore extending only under
subaerially-exposed shale areas of bars, barrier, highs and elevations. The
form of the mineralized sandstone zone was described in an earlier paper
(Haranczyk 1970). The sulphides disseminated in the sandstone, form
Liesegang-like parallel-asymmetrical diffusion bands built of sulphide grains
replacing the sandstone matrix.
470 C. Harailczyk

Rudno T -126- pl. %0

0.001 0.11.010.0 .S o -10 -20 -30 -l.0

2 1---:.........-<
v

o
o
o
/
o

Fig. 7. Sediment sequence of the sea shoreline, the location where lenses of redeposited shales are
formed. Localization is shown in Fig. 5. Additional explanations see Fig. 8

4. Surficial oxidation of the scavenger-burrowed shale beds lying on the tidal

flats in the fore-shore of the lagoonal barrier has induced the formation of the
ore flats within plywoodlike fissile shales. Some scavenger burrows are filled
with slightly diagenised sediments rich in foraminifera and other minute
fossils and subsequently replaced mimetically by sulphides, the same as those
that make ore flats (Fig. 11). Further details are given in another paper
(Haranczyk 1984a).
5. Prolonged emergence of the black shales occurring on eminences and highs
caused complete oxidation of the shales producing the so-called Rote Faule,
which means oxidized hematite-stained red shales. The metal content of the
shales was dispersed, or it was concentrated laterally in adjoining un-
weathered shales. Since Freiesleben's time (1815) the contacts of Kupfer-
schiefer and Rote Faule have been regarded as the most important sites for
rich ore.
6. Infiltration of the lagoonal barrier sandstones by concentrated lagoonal
brines induced the migration of metals, earlier distributed in the sandstones.
As a result irregular ore bodies of massive chalcocite (rarely galena) developed
in parts of the barrier sandstones where their matrix had been replaced by
anhydrite.
7. Sapropelic Model Acting in the Lagoonal Environment. Kupferschiefer or its
equivalent marl slate and, in lagoonal basins, lead- and zinc-bearing shales
may be regarded as the prototype of euxinic metalliferous sediments. Indeed,
Zechstein Copper-Bearing Shales in Poland 471

Rudna- -U-6-60G
0/00

o -10 -20 -30

513C
Carbonates

c
"iii
III
.s:
u
<II
N

Fig. 8. Sediment sequence on the internal slope of Rudna lagoonal barrier. 1 Stylolitic dolomite;
2 dark weakly laminated dolomite with pearls of anhydrite; 3 argillaceous laminated dolomite; 4 lead-
zinc-bearing dolomite with immature Brachiopods washed out from the barrier reef; 5 redeposited
copper-bearing shales; 6 copper-bearing shales showing rhythmic structure; 7 basal dolomite;
8 Weissliegend sandstone

these are kerogen-rich sedimentary rhythms or saprovarvites composed of

argillaceous-organic and dolomitic or calcitic laminae with a high or at least
above-average content of metals ~ccumulated by biological fixation
(Haranczyk 1984b). This fixation is evidenced by the correlation between C org
and S2- (Haranczyk 1961, 1972) and ipso facto with the metals saturating it.
Consequently, it is not possible that much of the organic carbon could be
oxidized during bacterial reduction of the sulphate ions, however, a consider-
able quantity of Hz in organic compounds was used for that purpose
(Haranczyk 1972, Rentzsch 1974). The organic substance of the shales is of
sapropelic origin as indicated by the low C/N ratio of 20 to 50 (Rentzsch
1974). Copper-bearing shales are usually kerogen-rich low-bituminous rocks
(Z. Sawlowicz, personal communication). There is strong evidence that
biogenic-sulphide sulphur originated in the shales. This has been proven by a
sulphur isotopic ratio <5 34s ranging from - 26 to - 32%0 (Marowsky 1969,
Wedepohl 1971, Haranczyk 1984b). Lead isotopes indicate that different
source material was involved in the genetic processes (Wedepohl et al. 1978).
The sapropelic model of biogenic fixation of metals is consistent with, and
can be better understood, if the process took place in a sea during its regression
stage. Marine regression stimulated stagnation and the seasonal bloom and fall
472 c. Haranczyk

Rudna S-1.08
0.01 0.1 1.0 o - 10 -20 -30
1068 ] : £1
~~~
- l -A-/
1069 ~7Fi:p,

Sulfates Sulfides

!
o

1071

1072

1073 a

1071.
1o

o
't
1075

1076

Fig. 9. Sediment sequence in the central part of the Rudna lagoon. For explanation of symbols see
Fig. 8

of phytoplankton in euphotic zones, as suggested by Brongersma-Sanders (1965),

thus developing a sea-bottom reduction zone in which the metalliferous anomaly
in seawater was effectively eliminated. Separation and growing isolation of the
Rudna lagoon produced brines with concentrated metals. When copper was
exhausted in the isolated lagoonal basin, lead, zinc and finally iron saturated the
biogenic-sulphide sulphur. Phytoplankton as well was blooming in euphotic
surficial zones, however, with increasing salt concentration its growth was
hindered, and consequently, beds of dolomarls, slightly laminated due to scarce
organic matter, were deposited, and finally organic growth almost ceased. It is
likely that the supply of heavy metals was exhausted earlier.

Discussion and Conclusions

The supergene processes involved during the first Zechstein subcycle produced
lenses of redeposited shales, introduced sulphide mineralization to the underlying
sandstones and contributed to the formation of the present size, shape and metal
concentration of the largest European copper deposits in Lower Silesia in
Poland. The process of supergene enrichment has so far been largely overlooked
Zechstein Copper-Bearing Shales in Poland 473

Fig. 10. Macrophotograph of a burrow perforated by scavenger in the copper-bearing shales from
tidal flats. The canal was filled with foraminifera-rich sediments and subsequently replaced with
bornite similar to that forming adjacent ore flats

in descriptions of copper deposits in sedimentary rocks (Tourtelot and Vine

1976), although the close relation between oxidation zones of so-called Rote Fau-
Ie and metal-rich Kupferschiefer beds has often been emphasized since Freies-
leben's time. Our knowledge of subaerial processes is still far from being com-
pletely understood. Especially the cementation processes of the unweathered,
deeper Kupferschiefer layers need additional sulphur isotopic data and
microscopic studies. It seems very probable that the parageneses including native
silver, digenite and covellite distinguished by Rentzsch and Knitzschke (1968) as
well the occurrences of rare cobalt, silver, mercury, palladium and platinum
minerals determined by Harailczyk (1972) . Kucha (1982) and others may be
regarded as parageneses of cementation minerals superposed on the primary
chalcopyrite, bornite and chalcocite mineralization of the Kupferschiefer during
its subaerial exposure.
The provenance of metals included in the Kupferschiefer is another intricate
unsolved problem. It is clear that the sapropelic model for the fixation of metals
dissolved in the early Zechstein Sea elucidates processes of efficient elimination
of the metal anomaly in seawater. Admittedly, different opinions on this ques-
tion are also held. Brongersma-Sanders (1965) proposed that the metals of
Kupferschiefer were supplied by normal water. This hypothesis does not explain
why the situation was never repeated. With the exception of the Kupferschiefer
itself, there are no increased metal contents in other cyclothemic basal Kupfer-
474 C. Haranczyk

Fig. 11. Macrophotograph of the basal chalcocite layer formed by cementation enrichment contact-
ing with bioturbidite sandstone with admixture of reef talus. Red stains (Rote Fiiule, II type) occur in
the Weissliegend sandstone. Red Stains developed when sandstone matrix was replaced with anhy-
drite and quartz grains were regenerated and produced and a porous water interface

schiefer equivalents, although the Stinkschiefer at the base of the second cycle
(Strassfurt-cycle) is the host rock of the Zechstein oil and gas deposits (Rentzsch
1974). Pertinent to the discussion is Wedepohl's suggestion (1971) that the high
manganesian content of carbonate rocks overlying the Kupferschiefer may in-
dicate hydrothermal submarine activity.
Areas of economic grade Kupferschiefer are not larger than 1070 of the total
area of Zechstein sediments (WedepohI1971), therefore, the question that now
arises is what has determined the localization of rich primary Kupferschiefer? It
must be added that a favourable euphotic zone and suitable environmental con-
ditions existed in much of the Zechstein Sea area. However, these factors alone
were insufficient. The author - due to the results of his studies and the forego-
ing considerations, has come to regard the presence of the Rotliegend rift basins
in the basement, explained by Lorenz and Nicholls (1976) and Jowett (1984) in
plate tectonic models, as the essential condition sine qua non for the area of sedi-
mentation of a primary copper-rich shale bed. The genetic relation might be asso-
ciated either with metal-remobilization action in a sabkha or in caliche style,
from the red beds filling these sedimentation basins, however, this is difficult to
reconcile with field evidence, or associated with reopening of rift dislocations,
which serving as feeding channels, enabled an ascension of Cu, Ag, Pb, Zn and
Co emanations (cf. Rentzsch 1974) from the resorbed oceanic lithospheric plate
in a partly frozen Benioff zone. This oceanic plate was earlier involved in
Hercynian collision and subsequent formation of the Rotliegend molasse and rift
Zechstein Copper-Bearing Shales in Poland 475

basins. However, other facts, e.g., with regards to Rucken Veins, must be ob-
tained before it can be said with certainty whether or not biogenically deposited
metals forming the Kupferschiefer, were of magmatic provenance associated
with waning Benioff zone activity. The sapropelic model of elimination of the
metal anomaly in the early Zechstein Sea, followed by enrichment during
subsequent subaerial exposure best explains the so far ascertained regularities of
ore distribution in old and new ore districts exploiting Zechstein black shales.

References

Beres B, Kienig P, Kijewski P, Tomaszewski JB (1973) Lithostratigraphical division of carbonate

rocks of the first Zechstein cyclothem i Fore-Sudetic Monocline. Przegl Geol 21: 17 - 20 (in Polish)
Blaszczyk JK (1982) Facies dichotomy in lowest part of Zechstein carbonate sequence in Lubin
copper district in Poland. Rudy Metale Nieielazne 27:151-156 (in Polish)
Brongersma-Sanders M (1965) Metals of Kupferschiefer supplied by normal seawater. Geol Rundsch
55:365 - 375
Brongersrna-Sanders M (1971) Origin of major cyclicity of evaporites and bituminous rocks: an
actualistic model. Merin Geol 11: 123 - 144
Eisenhuth K, Kautzsch E (1954) Handbuch fUr den Kupferschieferbergbau. Leipzig
Eisentraut 0 (1939) Der niederschlesische Zechstein und seine Kupferlagerstatten. Arch Lagersch
Forsch, Berlin, No 71
Freiesleben JG (1815) Geognostischer Beytrag zur Kenntniss des Kupferschiefergebirges, vol III, IV.
Freiberg
Harailczyk Cz (1961) Correlation between organic carbon, copper, and silver content in Zechstein
copper-bearing shales from the Lubin-Sieroszowice region Lower Silesia. Bull Acad Pol Sci Ser Sci
GeoI9:4-10
Harailczyk Cz (1964a) Investigation of copper-bearing Zechstein shales from the Wroclaw Monocline
Lower Silesia. Bull Acad Pol Sci Ser Sci Geol 12:13 -18
Harailczyk Cz (1964b) Petrographic classification of Zechstein copper-bearing shales from Lower
Silesia. Bull Acad Pol Sci Ser Sci GeoI12:19-24
Harailczyk Cz (1970) Zechstein lead-bearing shales in the Fore-Sudetian Monocline in Poland. Econ
GeoI65:481-495
Harailczyk Cz (1972) Ore mineralization in the Lower Zechstein euxinic sediments in the Fore-Sudetic
Monocline. Arch Miner 30: 13 - 173 (in Polish)
Harailczyk Cz (1984a) Ore flats in the copper-bearing shales of the Fore-Sudetic Monocline Lower
Silesia Poland. Amstutz Band (in press)
Harailczyk Cz (1984b) Lagoonal relief controlling distribution and isotopes content of the Zechstein
metalliferous sediments and hosted sulphides. Inst Geol Warszawa Pap (in Polish) (in press)
Jerzykiewicz T, Kijewski P, Mroczkowski J, Teisseyre AK (1976) Genesis of the Weissliegendes
sediments of the Fore-Sudetic Monocline. Geol Sudetica 11 :57 - 97
Jowett EC (1984) The Rotliegendes in Central Europe: basin development controlled by plate
tectonics. Przegl Geol 4:196 - 201 (in Polish)
Klapciilski J (1964) Palaeogeographical characteristics of the Zechstein of the Fore-Sudetic
Monocline. Ann PTG 34:551- 573 (in Polish)
Konstantynowicz E (1965) Mineralization of Zechstein sediments in Northern Sudetic Trough. Prace
Geol PAN, Krak6w, 28 (in Polish)
Krasoil J, Grodzicki J (1965) Sedimentationsentwicklung und Mineralisation des Weissliegenden
Niederschlesiens. Coli Pap, Eisleben, DDR
Kucha H (1982) Platinum-Group Metals in the Zechstein Copper Deposits, Poland. Econ Geol
77:1578-1591
Lagny P (1980) Les gisements stratiformes associes aux evaporites. Position dans Ie temps et place
dans I'espace des basins sedimentaires evaporatiques. Bull Centr Rech Explor Prod Elf-Aquit
4:44-478
476 C. Haranczyk: Zechstein Copper-Bearing Shales in Poland

Lorenz V, Nicholls A (1976) Permocarboniferous Basin and Range Province of Europe: an applica-
tion of plate tectonics. In: Falks H (ed) The continental permian in central, west and south Europe.
Riedel, Dortrecht
Marowsky G (1969) Schwefel-, Kohlenstoff- und Sauerstoff-Isotopenuntersuchungen am Kupfer-
schiefer als Beitrag zur genetischen Deutung. Contr Miner Petrol 22:290 - 334
Metzler M (1975) Quelques donnes sur les depots cupriferes de Basse Silesie Pologne. Bull Bur Rech
Geol Min 2 II 5:407-422
Neuhaus A (1940) Ober die Erzfiihrung des Kupfermergels der Grtiditzer Mulde in Schlesien. Z
Angew Miner 2, 3
Oszczepalski S (1980) Palaeogeography, sedimentation and mineralization of the Z 1 carbonate series
(Zechstein) in the western part of the Fore-Sudetic Monocline Western Poland. Contr Sedim
9:307-323
Renfro AR (1974) Genesis of Evaporite-Associated Stratiform Metalliferous Deposits - A Sabkha
Process. Econ Geol 69:33 - 45
Rentzsch J (1974) The Kupferschiefer in comparison with the deposits of the Zambian Copperbelt.
In: Cenenaire de la Societe Geologique de Belgique. Gisements Stratiformes et Provinces
Cupriferes. Liege, pp 395 - 418
Rentzsch J, Knitzschke G (1968) Die Erzmineralparagenesen des Kupferschiefers und ihre regionale
Verbreitung. Freiberger Forschungsh C231 :189 - 211
Rydzewski A (1969) Petrography of the copper-bearing shale from Fore-Sudetic Monocline. Inst
Geol Warszawa Bull 217:113-170
Schneiderhohn H (1921) Chalkographische Untersuchung des Mansfelder Kupferschiefers. Jahrb
Min Beil 47:1- 38
Smith DB (1980) The evolution of the English Zechstein basin. Contrib Sedimentol 9:7 - 34
Spangenberg C (1572) Mannsfeldische Chronica. Eisleben
Tourtelot B, Vine JD (1976) Copper Deposits in Sedimentary and Volcanogenic Rocks. Geol US Surv
Prof Pap 907 C
Wedepohl KH (1971) Kupferschiefer as a Prototype of Syngenetic Sedimentary Ore Deposits. Soc
Min Geol Jpn Spec Iss Proc IMA-IAGOD Meetings '70 IAGOD, vol III, pp 268 - 273
Wedepohl KH, Delevaux MH, Doe BR (1978) The potential source of lead in the Permian Kupfer-
schiefer bed of Europe and some selected Palaeozoic mineral deposits in the Federal Republic of
Germany. Contrib Miner Petrol 65:273 - 285
Formation Conditions of Copper-Sandstone
and Copper-Shale Deposits
A. M. LUR'YE t

Abstract

The concept introduced here suggests a common provenance of copper origin for
all cupriferous sandstones and shale-type deposits, the source of copper being
terrigenous red-bed formations deposited under arid conditions. Copper was
leached from these rock deposits by subsurface waters. Copper precipitation oc-
curred at hydrogen sulphide barriers, among which two types are recognized:
syngenetic (in unconsolidated sediments) and epigenetic (in lithified sediments).
The latter are subdivided, according to the origin of their organic matter, into
autochthonous and allochthonous groups. The decisive factor controlling the
formation of sediment-hosted copper deposits with large reservers is the specific
nature of the paleohydrogeological and geochemical environment.
Copper-sandstone and copper-shale deposits represent a highly specific group
of deposits, distinguished by the following characteristic features: (1) invariable
spatial association with red beds formed in an arid environment; (2) ore localiza-
tion in grey sedimentary rocks in close proximity to the red beds; (3) consistent
mineral composition and associations of major metallic minerals; (4) a zonal dis-
tribution pattern of the sulphides in the ore bodies.
Three hypotheses have been discussed by researchers for the origin of the
metallic substances in this group of deposits: hydrothermal (endogenetic solu-
tions), sedimentary (provenance area), and exogenetic-epigenetic (subsurface
pore waters of arid red-bed formations). The validity of the above-cited hypo-
theses can readily be weighed: if it is true that they should provide a lucid under-
standing for the known spatial association of copper mineralization with arid red
beds. This relationship, repeatedly verified by field observations, can be regarded
as a well-founded empirical law; nevertheless, it is commonly ignored by the pro-
ponents of the hydrothermal hypothesis. A similarly adequate solution of the
source problem has been offered by the supporters of the sedimentary hypo-
thesis. According to the latter, advocated by Strakhov (1963), copper was
leached from ore deposits of humid provenance by surface waters, and rede-
posited in arid regions in the form of carbonates that subsequently underwent
conversion to sulphides during early diagenesis. However, the original copper
carbonates have never been encountered. Consequently, it must be assumed that

I Institute of Lithosphere of the USSR, Academy of Sciences, Moscow, 109180, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et a\.)
© Springer-Verlag Berlin Heidelberg 1986
478 A.M. Lur'ye

these carbonates were invariably precipitated only in those areas where the
oxidizing environment in the sediments underwent a subsequent change to a re-
ducing, hydrogen sulphide environment.
The general spatial association of copper mineralization with arid red-bed
formations is feasibly explained by the epigenetic hypothesis, but from this view-
point the characteristics of certain deposits, undoubtedly of sedimentary origin
(for example, Mansfeld), has in past years remained incompatible. According to
the epigenetic hypothesis, copper was leached from red beds by subsurface
waters, and therefore its accumulation in marine sediments was regarded as in-
consistent with this concept. In fact, the contribution of copper from bottom
sediments to sea basin waters is quite natural, considering that influx to the seas
is derived not only from the run-off of surface waters, but also from inflow of
subsurface waters. If the waters of red-bed formations drained into the paleosea,
then these waters could have been responsible for the copper supply. Hence, the
most warranted hypothesis is that copper and associated metals were leached by
subsurface waters from red-bed formations. In this respect the deposits under
study are a group of closely related deposits, having in common the origin of
their metal content.
Sulphide mineralization is usually restricted to grey sediments, occurring im-
mediately adjacent to the red beds. This association indicates that the concentra-
tion of copper occurs at hydrogen sulphide barriers that are formed at the
interface between sediments of markedly different characteristics. The distinc-
tions between deposits are largely due to the conditions under which the
hydrogen sulphide barriers are formed. According to the time of formation of
the host rocks relative to mineralization, two types of hydrogen sulphide barriers
can be distinguished: syngenetic and epigenetic types. The first occur in uncon-
solidated sediments, the second in lithified rock deposits.

Syngenetic Barriers

Syngenetic barriers are most frequently formed as a result of the transgression of

the sea over an area of arid red molasse sediments. As a consequence of changes
in the drainage pattern caused by the advance of the sea, the discharge area of
subsurface waters in the red beds becomes the sea bottom, where interactions
occur between these waters and the hydrogen sulphide-bearing bottom silts
(Fig. 1).
All marine sediments that have formed under reducing conditions are poten-
tial hydrogen sulphide barriers. The real barriers are restricted to local areas
where hydraulic communications exist between the subsurface waters of the red
molasse sediments and the water bodies draining them. This hydraulic com-
munication is a most essential factor in the ultimate control on the formation of
copper deposits.
Confirmation of this is found in the following geological features of the
deposits:
1. Copper mineralization is restricted exclusively to those basins of marine
sediments where the basal section is comprised of red-bed formations. In the
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 479

pze-oee sta.~e

a o · ... a

0_

Dee-deposition sta.fe sea·t~"et

FeS 2 ,CoI, .LH 2S-II1HJ<2-
Cu-S

Fig. la, b. Formation of copper deposits in marine sediments. Pre-ore stage: a period of
accumulation to terrestrial red molasses in intermontane troughs (active water interchange); b period
of peneplanation (low hydrodynamical activity). Ore-deposition stage: period of transgression and
submarine discharge. 1 Cupreous chloride complexes (copper in water solutions); 2-4 dispersed
copper in sediments (2 adsorbed copper; 3 intercrystalline copper; 4 oxides and dissolved carbonates).
5 Water table; 6 boundary between infiltration and connate waters

Western Ural region, the Kazanian marine sediments, containing copper sulphide
accumulations in the basal section, are directly underlain by the red beds of the
Ufimian (Lur'ye 1972). The ore-bearing deposits in the Lower Zechstein of
Central Europe lie on the Rotliegend red molasse. The Nonesuch shales in the
White Pine deposit, U.S.A., are superposed on the red beds of the Copper
Harbor Formation (Ensign et al. 1972). The marine sediments of the Roan basin
within the Zambian Copper Belt overlie terrestrial sediments that accumulated
under arid climatic conditions; due to metamorphic alterations, the rocks lost
their original coloration (Mendelsohn 1963).
 480 A. M. Lur'ye

N
I
o.

0,001)
0,05

0,(10') ---------..,I-+-(~
0,05

o
,
to
,
0\0 u.

Fig. 2. Projection of ore zones in the basal section of the

Kazanian marine sediments on the color distribution
plan in the top of Ufimian sediments. For the plan,
color of rocks at the top of the Ufimian stage: I grey
sediments; 2 alternating red and grey sediments; 3 red
sediments; 4 ore zones in the basal sections of the
Kazanian marine sediments (circled numbers 1 Kandri-
kulsk; 2 Belebe; 3 Urshak; 4 Fedorov-Sterlibash; 5 ~I O"· ""
Salmish). - For the profile A - B: 1 clay; 2 sandstone; 3
argillite; 4 limestone; 5dolomite. For the columns at the 03
right, the rock colors: red (diagonal pattern), grey (no
pattern). Location of drill holes on the profile is without §S
scale
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 481

2. Copper mineralization is typically restricted only to the lower section of the

marine sediments. The remaining portion of the marine section, frequently
reaching hundreds and thousands of metres in thickness, is barren. Copper-
bearing rocks occur in immediate proximity to the underlying red beds. If the red
beds thin out or undergo facies changes to grey sediments, then the mineraliza-
tion is not seen in the overlying marine sediments. This is reflected very well in
the Bashkirian ore zones where the red beds occur at the top of the U fimian grade
laterally into grey sediments (Fig. 2). The copper-bearing zones in the basal
members of the Kazanian invariably overlie red beds, and have not been encoun-
tered outside the red-bed areas. The same observation applies to the Zechstein of
Central Europe: copper mineralization occurs only directly above the red
molasse. On uplifts of basement complexes, where red beds are lacking, copper
mineralization has not been recorded.
Sections composed of alternating terrestrial red beds and grey marine
deposits reveal the presence of several ore horizons e.g. the Kartamyshskaya
Formation of Donbass (Lur'ye 1970), Upper Kazanian of the West Fore-Urals
(Lur'ye 1973). Those depositional cycles which lack red beds are barren of
copper ores. This relationship is exemplified in the Zechstein cycles, where
deposits of non-ferrous metals are known to occur only at the base of the Werra
series overlying the Rotliegend. The basal layers of all higher depositional cycles,
which directly overlie evaporites, are barren.
3. The position of copper mineralization'in the marine basin reveals a distinct
dependence upon the structural features of the underlying red-bed deposits. The
ore bodies at the base of the Kazanian marine section are located above "shoe-
string" sandstones at the top of the Ufimian (Fig. 3). The top of the Ufimian
seems to be comprised chiefly of silty-argillaceous deposits. The sandstones are
evidently ancient channel deposits. In those cases, where the sandstones do not
extend along the entire channel of the Pre-Kazanian surface, but instead form
isolated "windows", a particularly sharp increase of the copper content can be
observed above these "windows".
From the paleogeographical sketch map of the Early Permian, showing the
projections of the ore bodies in the Zechstein marine sediments, it is apparent
that copper has concentrated chiefly above the red molasse sediment of the Rot-
liegend that fills the Variscian troughs (Fig. 4). Such superposed copper deposits
are known in the Saale, Saar, North-Sudeten and Oder troughs. The distribution
and configuration of the ore bodies depend to a large degree upon the structural
features of the red-beds filling the troughs. The longitudinal axis of the ore-runs
in the Mansfeld, Sangerhausen and Edderitz deposits (the Gellitzer line)
coincides with the boundary where the porphyritic conglomerates at the top of
the Rotliegend undergo lateral facies changes to arenaceous argillites (Hoynin-
gen-Hiine 1%3). Above this interface, the copper mineralization is characteristi-
cally at its maximum vertical extent, involving not only the cupriferous shale
beds, but also the lower Zechstein limestone section. With increasing distance
from this interface, the roof of the ore body drops further downwards into the
lower beds of the cupriferous shales (Fig. 5).
The Richelsdorf deposits are located in the northeastern part of the Saar
trough (Fig. 6). The closure of the trough resulted here from its conjugation with
482 A.M. Lur'ye

E::J1 •9

Ol .'0
f::;>ll .,1
[IIn~ .11
§r e 1J

~,

I ~*
0·

6AlAL LAYE U
.•
LIN G ULA
..
• ;
8

Fig. 3
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 483

the Buchenau Transversal Uplift. The copper mineralization is restricted exclu-

sively to the area overlying the conglomerates at the top of the Rotliegend. The
contours of the ore bodies in the cupriferous shales match very closely the inter-
face where the conglomerates grade laterally into sandstones (Richter 1941).
In the Zambian Copper Belt, the pre-Katangan erosional relief provided the
control on the distribution of the deposits in the Roan basin. Topographically
low areas in the basin are filled with terrestrial rock deposits that are similar to
those comprising the pre-ore horizon. High areas reaching to the base of marine
deposits are overlain by barren rocks, which are directly contiguous to the richest
ores (Mendelsohn 1963).
In the White Pine deposit, the copper mineralization extends over a ridge
buried beneath the red beds of the Copper Harbor Formation. This ridge had no
influence on the facies in the Nonesuch Basin, but, nevertheless, an ore zone
occurs above it (Ensign et al. 1972).
4. No changes in the position of the ore zone are apparent even when the
environment of sedimentation underwent abrupt changes. The transition from
basal sandstones to limestones and Lingula clays in the Kazanian section did not
effect the distribution pattern of the copper mineralization. All the units of the
section are ore-bearing in some places, and barren in others (Lur'ye 1972). The
same observation applies to the White Pine deposits, where the ore section, 15 m
thick, exhibits marked changes in its sedimentary environment, involving the
uplift of the seafloor, partial draining of the basin, and subsequently more wide-
spread transgression of the sea. These variations had no effect on ore emplace-
ment: mineralization does not show any relationship to the regression and trans-
gression of the sea (Ensign et al. 1972).
Notwithstanding the fact that the facies type does not control the position of
ore localization in marine basins, this factor does determine the morphology of
the ore body. Copper concentrates in those beds that are rich in carbon and
reduced sulphur, and do not accumulate in facies of oxidizing environments, due
to sections composed of several interbedded lithofacies, exhibit corresponding
alterations of beds with different copper contents. In cases where lateral changes
from reducing to oxidizing environments occur within the deposits, these facies
changes provide the ultimate control on the distribution of the mineralization
and so delimit the configuration of ore runs.
5. The orientation of the sulphide zones in ore bodies depends upon the
position of the red-bed deposits. At the geochemical barrier, the redox potential
undergoes a gradual change from high values in the red beds to low values in the
grey sediments. The Eh - pH diagrams (Garrels 1962) show that decreasing Eh

Fig. 3. The Belebei area. Paleogeographical map of the end of the Ufimian and copper content in
basal layers of the Kazanian marine sediments. Terrestrial condition of sedimentation (1- 2):
1 coastel plain (red clays and silts with a trace of soil forming); 2 "shoe strings" and alluvial fans (red
sands). Conditions of basin sedimentation (3 - 5): 3 bars (grey sands); 4 parts of the basin with high
salinity (dolomite, gypsum, silts and clays with gypsum); 5 lagoons of low salinity (alternations of red
clays, silts and more rarely, sands). 6 Areas where the top of the Ufimian stage was eroded; 7 dumps
of former mine workings; 8 ore zones in the profile (copper content more than 0.1070). Copper
content in basal layers of the Kazanian marine sediments (according to drill hole data) in kg m - 2: 9
less than 2.5; 10 more than 2.5; 11 more than 5; 12 more than 10; 13 more than 20; 14 outcrop
 00
I
1~ 6~ """"
21Cj-1 1E::j
0606
31 1 b E;ZI
91 . . . 1

~
3::
._ '- ;:::""'?~?,,-:/ Z: ~
...,
'1;"
Fig. 4. Beds of Zechstein basin and position of the ore zones. 1 Source area in upper Rotliegend; 2 limits of sedimentation for
regions of Upper Rotliegend; 3 - 6rock composition of the Upper Rotliegend (3 conglomerates; 4 sandstones; 5 argillites; 6 salts);
7 boundary of replacement of psephitic rocks of Rotliegend by argillaceous rocks; 8 boundaries of the Zechstein Sea; 9 copper de-
nnoito iT Ri,.h.,lotl",f· "S"n".. rh,,,,.en: TTT Mansfeld: TV Edderitz: V Sorembenz-Weil3wasser: VI North Sudetes: VII Lubin-
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 485

...•...
;

B 'B S
1...........11~6
1"'::<4J ~'
cm·~,
,..
Fig. 5. Distribution of non-ferrous metals in the copper-shale llOrizon southeast of the Harz moun-
tains, with elements of the paleography of the Upper Rotliegend. Compiled from papers of C. Freese
asnd W. Jung (1965), E. Hoyningen-Hiine (1963), O. Knitzschke (1965), Autorenkollektiv (1968).
1 Outlines of present-day rock masses; 2 limits of sedimentation regions of Upper Rotliegend;
3 boundary of replacement of psephitic rocks of Rotliegend (in the southeast) by argillaceous rocks
(in the northwest); 4 hematitic zones (Rote Faule). Areas of copper mineralization (more than 5.5 kg
tonne-I) in beds of the copper-shale horizon (5-7): 5 in beds 1 and 2; 6 in beds
1 to 4; 7 in beds 1 to 6 (beds: 1 fine clay; 2 coarse clay; 3 Kammschale; 4 Schiefer Kopf; 5 Schwarze
Berge; 6 Dachklotz); 8 areas with lead mineralization. Deposits: I Sangerhausen; II Mansfeld; III
Edderitz

values in the eu - Fe - S - 0 - H system are coincident with a change in the

stability fields from chalcocite to bornite and then chalcopyrite. The changes in
the sulphide composition of all ore bodies reveal similar sequential order of
sulphide zones. The "chalcocite end" in the zonal sulphide series invariably faces
the red beds. The copper content in the ores decreases with the distance away
from the latter (commonly the most enriched are the chalcocite-bornite ores).
The widely held consensus which postulates that the supply of copper is
derived directly from the continents in the process of ore genesis, does not
provide a satisfactory explanation for the regularities outlined above. They can
be comprehensibly interpreted only through the concept of leaching of copper
and associated metals from subjacent red beds by subsurface waters. The inflow
of subsurface waters into the Kazanian Sea could have occurred along the ancient
channels as the top of the Ufimian (Fig. 3). The outflow of subsurface waters
486 A. M. Lur'ye

A
/
/ ----- B
O Zontro

C\l8tbro

C Htntboch

• ',I . 2.
'h U ~ 28

Fig. 6A, 8 . Plan of the top of Rotliegend in the Richelsdorf deposits area (A), and copper content in
basal layers of the Zechstein (8). (From Richter 1941). 1 Sandstones (a from drill hole data; b from
outcrop data); 2 conglomerates (a, b the same as above). Copper content, in kg m- 2: 3 less than 10;
4 more than 10; 5 more than 20

from the Saar through into the Zechstein Sea would have been evoked by the
impounding effect of the Buchenau Uplift (Fig. 6) . The ore-bearing solutions
would emerge as seeps within the conglomerate area at the top of the Rotliegend.
This concept affords a feasible explanation for the matching of the contours of
ore bodies in the Richelsdorf deposits with the boundaries of facies changes from
conglomerates to more fine-grained rock types. In the Saale trough, the zones of
facies changes from conglomerates to arenaceous-argillaceous deposits occurring
at the top of the Rotliegend where the discharge areas for the subsurface waters
flow out from the troughs into the sea (Hoyningen-Hiine 1963). The argillaceous
sediments, as well as the basement complexes, served as obstructions impeding
the flow of the subsurface waters along the seaward-inclined Saale trough
(Figs. 4 and 5).
In the Roan basin, the subsurface waters at first oozed out within vast areas,
forming extensive ore deposits in the basal marine sections. Then, as the perme-
ability of the bottom sediments decreased under compaction, the discharge areas
became increasingly more restricted and localized around the highs in the pre-
Katangan relief where reservoir rocks of high permeability were preserved in the
overlying sediments (disintegration products of the uplands). This resulted in the
formation of concordant ore bodies in combination with flat-dipping crosscut-
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 487

ting ore bodies. The formation of the copper ore zones at the bottom of the
Nonesuch series can be explained by the presence of a burial ridge of rhyolites
beneath the sediments of the Copper Harbor Conglomerate. Above the ridge the
subsurface waters reached the sediments of the Nonesuch basin, where copper
sulphides were precipitated (Ensign et al. 1972, Brown 1971).
It is common knowledge that the origin of hydrogen sulphide in the sediments
is associated with biogenic sulphate reduction and also that copper solutions can
exert a toxic effect on the barrier microflora. The inhibiting effect of the copper
is not apparent as long as the amount of copper in the solution does not exceed
the equivalent amount of hydrogen sulphide (Temple and Le Roux 1964). Con-
sequently, it is conceivable that for the normal function of the hydrogen sulphide
barrier, a definite sequence of events is required: first, the development of
sulphate reduction processes in the bottom silts, and then the influx of copper-
bearing solutions. If the influx of ore-bearing solutions into the sediments occurs
prior to change of the oxidizing environment to a reducing environment, then no
sulphide barriers can be formed. This is confirmed by interrelations of ore
minerals: if there is an intergrowth of iron sulphides and copper sulphides in the
ores, the latter consistently replace the first. If copper was originally present in
the interstitial waters, then this copper would prevent the initial precipitation of
iron sulphide until the almost complete extraction of copper from the solution.
The sulphate reserve is limited, insofar as the supply from bottom waters to
the sediments ceases at depths between 4- 6 m from the surface (Shiskina 1972).
However, the high sulphur content in the ores, as compared to that in rocks of
similar depositional conditions, can be explained by the presence of sulphates in
the ore-bearing solutions, which is consistent with the concept of the formation
of copper-bearing solutions under arid conditions and supports the sulphur
isotopic data: the latter excludes the possibility that the entire supply of hydrogen
sulphide could have been contributed by ocean waters. From the sulphur isotopic
ratios, according to Vinogradov (1973), the inference can be drawn that the sub-
surface waters were the source of sulphate sulphur.
Taking into account the restrictions imposed by the toxicity of copper on the
development of biogenic sulphate reduction, a rough approximation can be made
of the rate of filtration of the ore-bearing solutions through the barrier. Accord-
ing to Kuznetsov (1976) the production of hydrogen sulphide by bacteria in silts
can achieve several mg 1- 1 day -1. Let us accept an average amount equal to
0.1 mgl- 1 day-1. This amount of HzS can precipitate O.4mg of copper in the
form of CUzS. Precipitation of copper from an ore-bearing solution with the con-
centration of 10 mg 1- 1 requires the work of bacteria during 10: 0.4 = 25 days.
With 100/0 water in the sediments (1 litre of solution for 10 dm 3 of sediment)
copper toxicity wjll not be apparent if within 25 days the ore solution advances
no more than VW
= 2.15 dm, i.e. the filtration rate should not exceed
21.5: 25 = 1 cm day-1.z
As a speculation it can be assumed that the formation of barren areas occurs
where the influx of copper exceeds the allowable values. No matter how

2 In all these calculations, the significance is in the order of magnitude of parameters but not in their
absolute values.
488 A. M. Lur'ye

paradoxical this may seem, it is possible that the more enhanced copper influx
was the cause of the barren areas. In particular, this phenomenon can explain the
fact that no correlation can be established between copper-bearing zones in
marine silts and tectonic fault zones in the underlying red-bed formations . While
the deposits are being formed, the discharge of copper-bearing subsurface waters
occurs not in the form of submarine springs, but is distributed over the total area
of the geochemical barrier by means of slow filtration through marine silts.

Epigenetic Barriers

Epigenetic barriers can be subdivided into two groups: allochthonous and

autochthonous barriers. The first group involves those barriers where organic
matter, which is the sulphate-reducing agent, was brought into the red-bed
formation from nearby oil and gas source beds; the second group, where organic
matter accumulated together with the sediments.
An obvious example of deposits formed at an allochthonous geochemical
barrier is the Central ore zone of the well-studied Dzhezkazgan copper deposits
(Fig. 7). The ore rocks of the minable series in these deposits accumulated in
oxidizing environments. The grey color of the rocks is secondary in nature
(Germanov 1962, Lur'ye et al. 1978, Feoktistov et al. 1971). The change in the

/recent eeoSiQn
GU2fau level

D a.
OS
I7ZJ c
1:.-=1 cI
ES:9 e
f
~9
(!:;] h
Gu= j

-i

Fig. 7. Formation of copper deposits in lithified rocks as exemplified in the Dzhezkazgan ore
deposits. Allochthonous hydrogen sulphide barrier after Lur'ye and Gablina (1978) . a Red sand-
stones, conglomerates; b grey sandstones, conglomerates; c red siltstones, argillites; d grey siltstones,
argillites; e limestones; fore deposits; g disseminated copper sulphide; h disseminated pyrite; j flow of
waters containing copper; j flow of waters containing organic matter
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 489

Dzhezkazgan deposits from oxidizing to reducing conditions and the formation

of the hydrogen sulphide barrier, according to Germanov (1962), is associated
with the penetration of hydrocarbons into the ore rocks from underlying Lower
Carboniferous marine deposits. The geological and geochemical data available to
date gives supporting evidence for this concept. In the Dzhezkazgan deposits a
lithological window was identified through which the hydrocarbons could have
penetrated into the ore rocks. This window is located near the paleo-uplift, where
the minable series was not deposited. Approaching the paleo-uplift, all the red-
colored rock types are replaced by grey sandstones. The intensity of reduction in
the host rocks tends to increase towards this lithological window: the relict red-
stained rocks completely disappear, and the sulphide composition of the ores
undergoes a gradual change towards minerals that require lower redox values for
their formation. Hydrocarbon compounds of the type associated with petroleum
have been encountered in the reduced rocks (Poplavko et al. 1977, Susura 1980).
Numerous examples can be offered of epigenetic deposits with autoch-
thonous hydrogen sulphide barriers. Among these are the Zhilandinskaya group
of deposits in the Dzhezkazgan area (Esenov et al. 1975), and the Kargala
deposits in the West Fore-Ural region (Lur'ye 1973). The first group of deposits
is restricted to rock formations, which, according to their position in the section,
are transitional from marine to terrestrial deposits. The copper mineralization is
confined to sandstone layers containing coal detritus. The grey marls and argil-
lites enclosing these beds, contain copper only immediately adjacent to the
contact with the ore body.
The ore bodies in the Kargala deposits are confined to channels infilled with
clastic material. The channels are incised in the "marly bed". The higher copper
contents tend to occur near the accumulations of plant debris in the channel
deposits. Copper-bearing marls occur only around the fringes of the ore bodies.
The origin of hydrogen sulphide at epigenetic barriers is more intricate, due
to the fact that it involves the introduction of an important new factor - tem-
perature. Pecularities of the mineral composition of the copper sulphides 3 infer
that the deposits were formed at temperatures below 75 DC and therefore
hydrogen sulphide may be of biogenic origin. From this approach to the
problem, the resemblance of pecularities of the copper sulphides, accumulating
in the sediments to those of lithified rocks, is a quite expectable feature, since it
reflects similarities in the conditions of ore genesis.
As Rose (1976) showed, the most probable form of copper transport in the
ore solutions is the form of cupreous-chloride complexes: [CuC121- and
[CuCI 312 -. The copper content in the form of chloride complexes in the solution
can reach 100 mg 1- 1 at 0.5 m Cl- and pH 7. Let us take a copper concentration
in the ore solution of 10 mg 1- 1• If rocks of a specific gravity equal to 2 have a
water content of 10070, then for the conversion of this water into a mineralized
solution, the amount of copper that should be extracted from 1 dm 3 (2 kg) of
rocks would equal 1 mg, on 5 X 10 - 5% of its weight. This figure is lower than the

3 Monomineral inclusions of djurleite, bornite with dispersed phases of chalcopyrite and chalcocite,
as well as x-bornite, are peculiarities of sulphides of copper in these deposits (Lur'ye 1978). x-
bornite is not stable and decomposes at 75°C.
490 A.M. Lur'ye

abundance of copper in the red-bed formations by a factor of 10-102• During

multiphased water interchange, the amount of copper extracted from the source
beds could be considerably greater.
Taking the rate of flow of subsurface waters equal to 1 cm day-t, a total po-
rosity of 100/0 and an effective porosity of 6%, then the calculated amount of ore
solution flowing through an area of 1 m 2 of the hydrogen sulphide barrier would
average 200 I yr -t, from which 2 g of copper could be precipitated in the form of
sulphides. The maximum mineralization of the ore bed in the Mansfeld deposits
(250 kg m -2) could be attained in a mineralization period lasting 125000 yrs.
As shown above, for normal functioning of the hydrogen sulphide barrier,
the amount of copper supplied per time unit (the product of copper concentra-
tion in the ore-bearing solution and its flow rate) cannot significantly exceed the
production of equivalent amounts of hydrogen sulphide. As there is no reason to
suggest considerable increases in the sulphide reduction process, time remains the
controlling factor, defining the growth of ore zones within the epigenetic barrier.
Since the linear dimension of epigenetic barriers in the direction of solution flow
is greater by a factor of 102, compared to that of syngenetic barriers, the mineral-
ization period will be millions of years in duration rather than tens of thousands
of years. The possibility of hydrogen sulphide influx from outside should not be
neglected for epigenetic barriers. In this case limitation in the amount of copper
supplied to the barrier, and thus the duration of formation of deposits, will be
defined by other factors.
From the foregoing calculations it becomes immediately apparent that the
source of copper is not related to the content of copper in red beds; it is only es-
sential that copper should occur in a form that can be released from the rocks in-
to the pore solution, and also that from such an immense reservoir as red-bed
formations, the subsurface waters should flow towards a restricted area where a
hydrogen sulphide barrier functions over a long period of time. From this, it is
conceivable that the formation of large deposits requires a specific paleohydro-
geological and geochemical environment. These genetic considerations should
furnish the basic concepts for prospecting for copper-sandstone and copper-shale
deposits.

Acknowledgements. I am indebted to Prof. A. W. Rose from Pennsylvania State University (USA)

for his comprehensive comments on the first version of this paper.

References

Ackermann E, Forster A (1960) Grundziige der Stratigraphie und Struktur des lrumiden Troges. In:
Int Geol Congr. Rep 21 Session. Norden, part 18. Copenhagen, pp 182-192
Brown AC (1971) Zoning in the White Pine copper deposit, Ontonagon Country, Michigan. Econ
GeoI66,3,4:540-572
Ensign KO Jr, White WS, Wright JC et al. (1972) Copper deposits in the Nonesuch Shales of
Michigan. In: Ore Deposits of the United States. Izd Mir 660 pp
Esenov Sh E, Zaizev YA (eds) (1975) Geology and mineral resources of the Dzhezkasgan ore district.
Nedra, 284 pp (in Russian)
Feoktistov BP, Kramarenko LE (1971) Bearing on the gray color of the rocks in the Dzhezkazgan
deposits. Litologia Polez Iskopaem 3:119-125
Formation Conditions of Copper-Sandstone and Copper Shale Deposits 491

Freese G, Jung W (1965) Uber die Rotfarbung der Basalschichten des Zechsteins (Rote Faule) und
ihre Beziehung zum Nebengestein im siidostlichen Harzvorland. Freiberg Forschungsh, Leipzig,
193:9 - 23
Garrels R (1962) Mineral equilibria. Mineralniye ravnovesia. M Izd-vo Inostr Lit, 306 pp
Germanov AI (1962) Hydrodynamical and hydrochemical formation conditions of certain hydro-
thermal deposits. Gidrodinamicheskiye i hidrokhimicheskiye uslovia obrazovania nekotorikh
gidrotermalnikh mestorozhdenii. Izv AN SSSR Ser Geol 7:79 - 98
Hoyningen-Hiine E (1963) Zur Palaohydrologie des Oberrotliegenden und des Zechsteins im Harz-
vorland. Ber Geol Ges DDR 1:201 - 220
Knitzschke G (1975) Die wichtigsten Erzminerale des Kupferschiefers sowie seines unmittelbaren
Liegenden und Hangenden im Siidostlichen Harzvorland. Z angew Geol 12:625 - 637
Kuznesov SI (1976) The results and outlook of development of geological microbiology. In: Ecology
and geochemical activity of microorganisms. Izd-vo ONTI Nauchnogo Centra Biologicheskikh
Issledovanii. Puschino, 179 pp (in Russian)
Lur'ye AM (1970) Regularities in location of copper mineralization in Lower Permian variegated
formation of Donbass. In: State and problems of Soviet lithology. V kn.: Sostoyanie i zadachi
Sovetskoi lithologii. Vol II Izd Nauka, 248 pp (in Russian)
Lur'ye AM (1972) Mansfeld-type mineralization in Upper Permion Sediments of Western Urals
Foreland. Academii Nauk SSSR vol 203, 3:658 - 661 (in Russian)
Lur'ye AM (1973) The types of copper deposits in redbed formation of the Russian platform. T. XV,
6:79 - 88 (in Russian)
Lur'ye AM (1978) Alteration of ores during epigenesis and metamorphism in the copper deposits of
the redbed associations. Intern Geol Rev 20, 6:627 - 636
Lur'ye AM, Gablina IF (1978) Principle scheme for the formation of Exogenetic copper deposits.
Printsipal'naya skhema obrazovania ekzogennikh mestorozhdenii medi. Dokladi AN SSSR. Vol
241,6:1402-1405
Mendelsohn F (ed) (1963) The copper belt of Northern Rhodesia. Mednii poyas Severnoy Rodezii
(pod redaktsiyey F. Mendelsona). M Izd-vo Inostr Lit, 472 pp
Poplavko EM, Ivanov VV, Serkis YuT (1977) On the geochemistry and formation conditions of
cupriferous sandstones and shales. 0 geokhimicheskikh osobennostyakh i uslovia obrazovania
medistikh peschanikov i slantsev. Geokhimia 88:1217 -1233
Richter G (1941) Geologische GesetzmaBigkeiten in der Metallfiihrung des Kupferschiefers. Arch
Lagerstattenforsch 73:61
Ridge TD (ed) (1972) Copper deposits in the Nonesuch Shales of Michigan. In: Ore deposits of the
United States. AIME, New York, pp 602-627. Medniye mestorozhdenia v slantsakh Nonesuch,
shat Michigan. V kn.: Rudniye mestorozhdenia Soedinyennikh Shtatov. Izd Mir, pp 602 - 627
Rose AW (1976) The effect of cupreous chloride complexes in the origin of red-beds copper and
related deposits. Econ Geol 71,6:1036
Strakhov NM (1963) Types of lithogenesis and their evolution in the history of the earth. M.
Gosgeoltekhizdat, 535 pp (in Russian)
Shish kina OV (1972) Geochemistry of interstitial waters in bottom sediments of seas and oceans.
Geokhimia morskikh i okeanicheskikh ilovikh vod. Izd Nauka, Moscow, 238 pp
Srakhov NM (1963) Types of lithogenesis and their evolution in the history of the earth. Tipi
litogeneza i ikh evolyutsia v istorii zemli. M Gosgeoltekhizdat, 535 pp
Susura BB (1980) Geochemistry of epigenetic ore-deposition in the red-bed formations of the Chu-
Sarysui depression in Kazakhstan (copper-sandstone type). Geokhimia epigeneticheskovo rudo-
obrazovania v krasnotsvetnikh formatsiakh Chu-Sarysuiskoy deressii Kazakhstana (tip medistikh
peschanikov). Avtoref Diss Na Soiskaniye Uchen. St Kand Geol Miner Nauk M:27 pp
Temple KL, Le Roux NW (1964) Syngenesis of sulfide ores: Sulfate-reducing bacteria and copper
toxicity. Econ Geol 59, 2:271- 278
Vinogradov VI (1973) The source of sulphur in ore-deposits, as reflected by sulphur-isotope data. In:
Int Geochem Congr 2:45 - 53. Istochnik seri rudnikh mestorozhdenii po izotopninm dannim.
V kn.: Mezhdunarodnii geokhimicheskii kongress. M T 2, pp 45 - 53
Major Types of Copper-Bearing Zones in the
Soviet Union
Y. v. BOGDANOV 1

Abstract

The territory of the Soviet Union is characterized by orogenic, platform and geo-
synclinal, copper-bearing zones. Major epochs of sedimentary copper accumula-
tion are: Early and Late Proterozoic and Late Paleozoic, favouring generation of
orogenic copper sandstone, and Middle Mesozoic, favouring the emplacement of
copper - pyrrhotite mineralization in early geosynclinal black shale trough
zones.
Different ways of evaluating areas with good potential are shown, based on
the cyclic recurrence and facies features of copper-bearing sediments, as well as
structure of productive strata in deposits.
The cyclic pattern of copper-bearing molasse sediments determines the
morphology of productive strata, their simple or multicyclic structure. Flyschoid
sediments contain copper sandstone with a simple structure of productive strata,
having relatively small size.
Ore-bearing deposits of arid red beds are divisible into the following zones:
(1) copper and polymetallic mineralization in red and variegated sediments of
greywacke composition and (2) copper mineralization in variegated and red
sediments of arkose composition and displaying thick cycles.

Introduction

The extensive territory of the Soviet Union is one of the few regions of the earth
characterized by widespread, copper deposits of sedimentary and sedimentary-
volcanogenic origin of different age. Well known are such copper-bearing
provinces in our country as Central Kazakhstan (Dzhezkazgan, Sary-Oba, etc.)
and the North Transbaikal area (Udokan, Unkur, etc.), which are similar in
many characteristics to such foreign ore-bearing areas as the Copperbelt in
Central Africa, the Fore-Sudetic Trough in Lower Silesia (Lubin-Sieroszowice),
Poland, and Central Afghanistan (Ainak, etc.).
Also well known is the great economic significance of copper sandstone
deposits. Economically, ores of this type rank second among copper deposits

1 VSEGEI, Leningrad, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
 Major Types of Copper-Bearing Zones in the Soviet Union 493

after ore bodies of the porphyry copper type. Copper sandstone is economically
valuable due to the morphology of its ore bodies, which is favourable for exploi-
tation (large sheet-form deposits), and a simple concentration process for most
ores, which have a rather high (> 1.5OJo Cu) copper content and contain admix-
tures of easily extractable precious and rare metals.
The potential for discovering new copper sandstone deposits is far from being
exhausted. This potential is confirmed by the fact that during recent decades a
number of intersting ore bodies were discovered in the USSR (the Igarka area,
Sette-Daban, Western Darvaza) and abroad (Ainak in Afghanistan). Thus, it
seems appropriate to characterize the copper-bearing zones and to discuss a more
detailed classification in order to solve urgent problems of discovering additional
deposits and evaluating areas with high copper potential.

Major Types of Copper-Bearing Zones

In the Soviet Union, several major copper-bearing zones are classified on the
basis of their geotectonic position into orogenic, platform and geosynclinal
formations. The stratiform copper deposits occurring within these zones differ
greatly in many parameters (Fig. 1).

Fig. 1. Distribution of major copper-bearing zones in the territory of the Soviet Union. I-III Cop-
per-bearing zones of fold belts: I Proterozoic; II Paleozoic; III Mesozoic-Cenozoic; IV - V copper-
bearing zones of platform mantle. IV Early Paleozoic; V Middle and Late Paleozoic. Zones:
°
1 Kodar-Udokan (PR I ); 2 Talas (PR 2 ); 3 Igarka (PR2 ); 4 Yenisei (C 2 - 3 ); 5 Lena (C 3 - 1 ); 6 Dnestr
(D I ); 7 Minusa (D-C I ); 8 Tengiz-Sarysu (D2 - 3 -P); 9 Sette-Daban (D 3 -C I ); 10 Donbas (PI);
11 West Darwaza; 12 Fore-Urals (P z); 13 Mangyshlak (P - T); 14 Near-divide (T - J 2 ); 15 Tadjic
(Kd, 16 Fergana (Ng)
494 Y. V. Bogdanov

The specific character of copper mineralization in orogenic and platform

zones which are underlain by red and variegated carbonate-terrigenous deposits,
is determined mainly by the composition of the clastic material and its cement.
Copper-bearing carbonate-cemented red beds of arkosic composition are charac-
terized by a limited suite of accessory components (minor Ag, Mo, Bi). Carbon-
ate-cemented red beds of greywacke composition concentrate, along with
copper, admixtures of Ag, Co, Pb, Mo, Re and other elements. Quite different
ore elements are accumulated in humid carbonate-free red beds. These are
primarily, iron, bauxites and phosphorites.

Orogenic Copper-Bearing Zones

This type of extensive ore-bearing area is characterized by the largest number of

stratified copper sandstone ore deposits. Copper-bearing red and variegated car-
bonate-clastic and volcanogenic-clastic rocks occur in early and late orogenic
troughs of varied shape and size. Depending on the tectonic regime and the
features of orogenesis, these rocks are characterized by a cyclic recurrence on dif-
ferent scale. The lithologic composition of these deposits is mainly determined by
two factors, viz., the structure of the eroded basement blocks and features of
preceding or synchronous volcanism, as well as by specific paleogeographic and
paleotectonic conditions in the source and sedimentation areas. The above
factors, to a large extent, predetermine the special metallogeny of the red and
variegated formations.
In the North Transbaikal area, within the Olekma-Vitim mountain land,
numerous troughs of Karelids are developed, filled in by metamorphosed Early
Proterozoic red beds. The most interesting structure within this ore-bearing
province is the Kodar-Udokan zone, containing the Udokan and several minor
copper sandstone deposits (Fig. 2). The regional distribution of copper-bearing
sediments displays a distinct zonal pattern relative to source areas. The strati-
graphic and formational control of mineralization have been recorded. This
evidence, along with the great areal extent of ore mineralization, provides the
main basis for assigning the mineralization in question to a primary sedimentary
origin (Bogdanov 1966). It should be noted that rich ores display many features
of epigenetic concentration, viz., (1) an association with the most coarse-grained
sandstone; (2) the presence of strongly recrystallized carbonate or carbonate-
mica-quartz cement; (3) the development of various stringer - impregnated
structures and texture in ores; (4) a crosscutting character of mineral zoning
relative to the rock bedding (Bogdanov and Feoktistov 1982).
In the rather complicated accumulation history of copper-bearing sediments
in the Kodar-Udokan trough, two major stages are distinguished; (1) early
orogenic - accumulation of flyschoid greywacke deposits of the China Series
(3000 - 4000 m) and (2) late orogenic-sedimentation of arkosic molasse sedi-
ments of the Kemen Series (4000 - 5000 m).
Ore bearing deposits of these series from two belts, extending up to 150 km
and 20 - 25 to 40 km wide, which surround denudation areas or paleouplifts in
the orogenic trough. In each belt, there is a gradation up the section from eche-
Major Types of Copper-Bearing Zones in the Soviet Union 495

,...".~_ Sul ban

Fig. 2. Distribution of copper sediments of the Kodar-Udokan zone. 1Sediments of the molasse silt-
stone-sandstone series; 2 sediments of the flyshoid sandstone-mudstone-siltstone series; 3 shale
strata, enclosing schistose rock block of the flyshoid series formation; 4 copper-bearing belts, con-
firmed (0) and assumed (b) (I Aleksandrovsky-Chitkanda; II Naminga-Sakukan); 5 coastal-marine
sediments with some copper-bearing horizons; 6 lagoonal-deltaic copper sediments. Areas circled
numbers: Sulban (1); Unkur (2); Ikabya (3); Hani-Sakukan (4); Burpala (5); Pravo-Ingamakit (6);
Udokan (7); Saku (8); 7 Baikal-Amur magistral

lonlike lenses of copper sandstone to more continuous copper slate layers

(Fig. 3).
The outer belt is most uniform and corresponds to an older Aleksandrovsky-
Chitkanda group of the China Series in the Udokan carbonate-terrigenous
deposits. The ore-bearing part of the section is characterized by flyshoid cyclic
recurrence and a greywacke composition of the clastic rocks. Copper mineraliza-
tion is confined to coastal-marine sediments comprising lenses of silty-sandy
lagoonal-deltaic deposits (at Krasnoe, Pravyj Ingamakit, etc.). The thickness of
productive strata in these mineralized areas amount to tens of metres. Lenticular
ore bodies occur in an echelon. Ores are dominated by chalcopyrite, bornite and,
less frequently, chalcocite, occurring together with pyrite and pyrrhotite. In-
creased content of silver and molybdenum, and less frequently, of lead, nickel,
cobalt, arsenic, and bismuth, are recorded. Ores are characterized by a compara-
tively narrow range of isotopic composition of sulphide sulphur (from -15%0 to
+ 10%0 t5S 34 ).
The well-known Udokan deposit and a number of cupreous sandstone ore
occurrences are confined to the inner belt, corresponding to the younger
Naminga-Sakukan group. The productive strata at Udokan, about 400 m thick,
are composed of metamorphosed clastic sediments of an arkose composition,
having thick cycles. Lenticular ore bodies display an echelonlike occurrence in
the productive sequence. Stringer-impregnated bornite-chalcocite and bornite-
chalcopyrite ores dominate, pyrite-chalcopyrite ores being found much more
496 Y. v. Bogdanov

Fig. 3. Genetic model of copper sandstone deposits of the Kodar-Udokan zone. I Area of denudation
- blocks of Archean rocks: Chara (CH) and Kalar (K) ; 2 - 3 China Series: 2 flyshoid sandstone-
mudstone-siltstone sediments; 3 lenses of siltstone (a), sandstone (b), and carbonate rocks (c);
4 - 6 Kernen Series: molasse siltstone sandstone sediments; 5 lenses of sandstone (a) and siltstone (b) ;
6 mudstone-siltstone sediments of Naminga formation; 7copper-bearing horizons, enclosing copper
sandstones of Dzhezkazgan (a) and Dzhilanda (b) types and copper slates (c); 8 faults; 9 directions of
migration of elision waters

rarely. These ores are characterized by low contents of accessory elements (silver
and bismuth). The isotopic composition of sulphide sulphur is highly variable (a
range of 50%0 ~S34), with light isotopes dominating, which points to the presence
of biogenic sulphur in these cupreous deposits.
The small areal extent of deposits of the inner copper-bearing belt testifies to
the fact that the prospects for discovering cupreous sandstone deposits of the
molasse type, similar to the Udokan deposit, are limited. The potentials of the
outer belt, which has a great areal extent, are much higher (Bogdanov et al.
1973).
In Central Kazakhstan, Upper Paleozoic cupreous deposits are localized
within the extensive Teniz-Sarysu zone which has a north-south strike. The most
interesting copper sandstone deposits have been discovered in the Dzhezkazgan
area. There, two long ore-bearing belts of a north-eastern strike have been out-
lined, viz., the northern one - Taskuduk, and the southern one -
Dzhezkazgan. These belts are composed of variegated lagoonal-deltaic silty-
sandy deposits. Within the above belts, several copper sandstone deposits have
been discovered, differing in size, and morphology and in the structure of the
productive strata and certain ore bodies. The Taskuduk flyschoid belt includes
the Sary-Oba, Itauz and a number of smaller deposits and ore shows, which are
characterized by a simple structure.
In the Dzhezkazgan molasse belt, apart from the Dzhezkazgan deposit, only a
few ore shows of copper sandstone are recorded. The productive strata of
Dzhezkazgan, about 650 m thick, have a multicyclic structure with an echelonlike
occurrence of lenticular ore bodies. Up the section of the productive strata, the
morphology of ore bodies and mineral zones becomes more complicated
(Bogdanov et al. 1973, Narkelyun et al. 1983).
The mineral composition of the ores is relatively simple. Impregnated copper
ores are dominated by chalcocite, bornite and chalcopyrite. Pyrite, arsenopyrite,
tennantite and other more rare minerals are also found. Ores are dominated by
Major Types of Copper-Bearing Zones in the Soviet Union 497

light sulphur isotopes, probably of biogenic ongm (OS34 from + 1.00 to

- 4.02%0). The economic potential of the more uniform Taskudak belt seems
much higher. Since this belt was outlined, several minor cupreous sandstone
deposits have been discovered within it. South of Dzhezkazgan, in the Chu-
Sarysu Depression, Carboniferous variegated rocks on the slopes of various up-
lifts should be studied.
Among orogenic copper-bearing areas of a simpler structure, the Igarka area
in the Lower Yenisei River should be mentioned. Not long ago, Vendian
cupreous sandstone was discovered there. The copper-bearing red-bed formation
is confined to the junction area of the Igarka protrusion of Baikalids and a
piedmont trough. Stratified copper mineralization, localized on several horizons,
is represented by stringer-impregnated bornite-chalcocite and chalcopyrite min-
eralization in grey clastic sediments (Narkelyun et al. 1983).
The economic potential of copper in the Igarka area has not been adequately
studied. First, the Vendian variagated deposits, occurring near paleouplifts
should be studied in detail and prospected.
With a certain degree of conventionality, the ore areas of Western Darvaza in
Central Asia (P 2 ), Sette-Daban in East Siberia (C I ) and the Minusink Basin in
the southern Krasnoyarsk area (D - C I ) can also be assigned to these types of
copper-bearing areas. These copper-bearing areas display an increased content of
pyroclastic rocks in the section and include sheetlike bodies of basaltic or
andesitic volcanics. The evaluation of the economic potential of the above
copper-bearing areas has not yet been completed.

Copper-Bearing Areas in Platform Regions

Within the sediments mantling the platform, there are practically no economic
copper-bearing areas. Numerous ore shows of cupreous sandstone are recorded
near large arched paleouplifts on the eastern margin of the Russian Platform, in
the Fore-Uralian area and in the southern Siberian Platform in the Upper Lena
River (Bogdanov et al. 1973, Narkelyun et al. 1983).
In the Fore-Urals, malachite and chalcocite mineralization is confined to
Lower Permian alluvian and lacustrine continental deposits, displaying highly
variable facies. There are small bodies of rich epigenetic copper ores, confined to
fluvial facies. There is also a progressive decrease in age of cupreous sandstone
from north to south.
Lower Permian sandstone is also reported from the southern Russian Plat-
form, the slopes of the Bakhmut and the Kalmius-Torets basins. In this latter
area, a rhythmic alternation of variegated mudstone, marl, siltstone and sand-
stone of the Kartamysh Formation, assigned to shallow marine sediments, is
recorded. Ore-bearing horizons of small thickness are rather uniform along the
strike and dip. Cupreous rocks contain low-grade impregnations of chalcocite,
and less frequently of chalcopyrite, bornite and pyrite.
The economic potential of copper presence in red formations of marginal,
tectonically activated parts of the Russian Platform have not been adequately
studied. As a first priority, the slopes of synsedimentary paleouplifts (domes,
498 Y. v. Bogdanov

arches, swells, etc.) should be studied, for they frequently control the distribu-
tion of copper sandstone deposits (Fore-Sudetic Block in Lower Silesia, the
Igarka protrusion in the Krasnoyarsk area, etc.). Of certain interest in this
respect is the Fore-Urals area near Orenburg, where lagoonal-deltaic variegated
rocks underlie slopes of such paleostructures.
In the southern Siberian Platform, cupreous deposits occur within the lower
Paleozoic of the Lena and Yenisei areas. These are thin horizons of cupreous silt-
stone, sandstone and marl of shallow, marine facies.
The most interesting, the Lena copper-bearing zone, 600 by 200 km, extends
northeastwards and includes areas, underlain by the Verkholensk and Ilga
formations (C 2 - 3), as well as by lead-zinc-bearing biogenic carbonate rocks of
the Ust-Kut Formation (0 1). The thickness of copper-bearing horizons ranges
from 1 to 10 m, rarely more. Towards the central parts of synsedimentary
troughs, the thickness of the Ilga Formation increases from 60 - 70 to 300 m, and
the number of copper-bearing horizons grows from one to three. Cupreous rocks
are dominated by malachite. In places, chalcopyrite, bornite and chalcocite, as
well as pyrite, galena and tennantite, occur.
The economic potentials of the Lena and Yenisei copper-bearing areas are not
quite clear. Further studies on the structural control of cupreous sandstone
should be carried out. Thus, of great significance is the study of the distribution
pattern of cupreous sandstone, particularly of lagoonal-deltaic facies, on the
slopes of various paleouplifts (the Nepa Arch, etc.). Of certain interest with
regards to the copper presence are variegated deposits of the so-called
podsolevoj 2 complex (V - Ct>.

Copper-Bearing Zones of Geosynclinal Areas

In geosynclinal areas, stratiform copper deposits of a sedimentary-volcanogenic

origin are mainly found among black shale formations of the initial evolution
stages of troughs. Pyrite-polymetallic deposits, found among sedimentary-vol-
canogenic formations in such provinces as the Urals and the Ore Altai, are com-
monly assigned to the family of pyrite deposits (Savin 1978). Along with these de-
posits, an isolated group is frequently formed by copper-pyrrhotite deposits,
localized in black shale formations. The Near-divide metallogenic zone of the
Great Caucasus can be named among the areas with extensive showings of cop-
per-pyrrhotite mineralization. Within this zone, a very thick (3.5 to 12 km) se-
quence of Lower-Middle Jurassic flyschoid clastic rocks occurs and encloses
comprising copper-pyrrhotite deposits (Kizil-Dere, Katsdag, etc.). The exhala-
tive-sedimentary copper-pyrrhotite mineralization is commonly localized in
supravolcanogenic parts of the section. Ore bodies at the Kizil-Dere deposit are
lenticular, less frequently vein-shaped, with characteristic pinches and swells.
Bodies of massive copper-pyrrhotite ores are accompanied by aureoles of
stringer-impregnated mineralization of the same composition. Rocks of the foot-
wall of ore deposits are strongly quartzified, sericitized and carbonatized. The

2 Podsolevoj complex - rocks underlying salt strata.

Major Types of Copper-Bearing Zones in the Soviet Union 499

thickness of these zones of wallrocks alteration is 20 to 40 m. The ores, along

with pyrrhotite, pyrite and chalcopyrite, also contain sphalerite, galena, and less
frequently, cobaltite and magnetite (Savin 1978).
The presence of copper-pyrrhotite mineralization in shale structural-forma-
tional zones points indirectly to a possible presence of pyrite-polymetallic miner-
alization. Thus, many early geosynclinal black shale formations should also be
studied.

Major Epochs of Copper Accumulation

The copper-bearing areas discussed are distinguished mainly by an extensive de-

velopment of the following three types of structural-formational complexes, viz.,
orogenic and platform red and variegated carbonate-clastic and geosynclinal
black shale deposits. Accumulation of the above deposits of copper-bearing
complexes took place during certain evolution stages of geosynclinal-folded,
orogenic and platform areas of different age. Thus, major ore epochs are dis-
tinguished, strongly differing in scale and character of mineralization. These are,
primarily, the Early Proterozoic, Late Proterozoic and Late Paleozoic copper-
bearing epochs.
The Early Proterozoic epoch of sedimentary copper accumulation was mani-
fested during the stage of Karelian orogeny in the western Stanovoj Mobile Belt.
There, the oldest copper sandstone and slate deposits were emplaced. In this
region, cupreous red beds were accumulated in a system of north-south and east-
west orogenic troughs. Cupreous sediments are commonly confined to the
coastal-marine belt of an extensive ancient arid region. A multistage generation
of cupreous sandstone took place during orogenic movements due to erosion of
Archean blocks of the Aldan Shield.
Copper sediments are characterized by certain regular changes in time: grada-
tions of greywacke into arkoses, flyschoids into molasses, increase of the thick-
ness of productive strata of cupreous sandstone deposits, and a simpler composi-
tion of ores. The initial red colouring of rocks has almost disappeared due to
regional metamorphism, corresponding to te greenschist facies or metagenesis
stage.
The Late Proterozoic epoch of copper accumulation was manifested most
intensively in red beds of foredeeps on the Siberian Platform. This is, primarily,
the Igarka Trough, where in Riphean time (youngest Precambrian) several accu-
mulation stages of cupreous sediments are distinguished, covering the marginal
parts of arid areas and manifested most intensively near paleouplifts. Copper
accumulation is much weaker in the Western Baikal area, where in Vendian, the
emplacement conditions of cupreous sediments were characterized by a less arid
climate.
The Late Paleozoic ore-bearing epoch was manifested in the development of
copper-bearing arid carbonate-clastic and volcanogenic-clastic red beds in
Caledonian and Hercynian orogenic troughs of the Sarysu-Teniz zone in Central
Kazakhstan (C 2 - 3), Sette-Daban in East Siberian (C t ), Mangyshlak (Pt),
500 Y. v. Bogdanov

Western Darvaza (P 2), and Minusinsk depressions (D - C 1), as well as platform

red beds of the Fore-Uralian area (P 1). This epoch is characterized by an absence
of economic concentrations of cupreous sandstone among Mesozoic and Ceno-
zoic red beds.
Of major interest are copper sandstone concentrations in the Dzhezkazgan
area. They are characterized by: (1) association with coastal-marine (lagoonal-
deltaic, etc.) greywacke deposits and (2) gradation of flyschoids into molasse
sediments in time and increased size and grade of copper-bearing sequences.
Geosynclinal copper accumulation was restricted to occurrences of flyschoid
black shale formations with copper-pyrrhotite mineralization (Kizil-Dere etc.) in
the Mesozoic.

A Prognostic Evaluation of Copper-Bearing Areas

Regional prognostication is a process of successive differentiation of the regions

evaluated, based on their potentials, with a view to establishing ore-bearing areas
which are potential ore fields of deposits of the leading economic types. The
lithological-petrological composition and the inner structure (rhythmic recur-
rence, etc.) of formational units predetermine their ore content. It is this very
principle that determines the only correct oroer of regional prognostication:
from studying the environment to the discovery of ore. A specific feature of
regional prognostication is its areal character. This suggests that the standard
objects chosen for prognostic evaluation should not be deposits, but ore-bearing
areas (zones, areas, etc.).
In the course of evaluating metallogenic provinces for deposits of the
cupreous sandstone and copper-slate type, primarily the geotectonic, forma-
tional and stratigraphic ore-controlling factors should be taken into account.
Within large regions under consideration, of particular importance are orogenic
troughs, depressions, and troughs with abundant mafic volcanics in the basement
and flanks uplifts. Copper-bearing carbonate-terrigenous sediments are charac-
terized by widespread red and variegated formations of a carbonate composition.
They display a thick molassic bedding, or a medium, or less frequently, thin
flyschoid bedding, the deposits of an arkose or greywacke composition being
very thick. Copper mineralization is commonly confined to the upper part of
megacycles.
Epochs of sedimentary copper accumulation are different for provinces of
different age. In the North Transbaikal area, at least two copper-bearing time
intervals are distinguished in Early Proterozoic; in Central Kazakhstan, two
periods in Late Paleozoic.
In the course of evaluating metallogenic zones, the taxon prognosticated is an
ore area, and only in certain instances, a potential ore field, characterized by a
favourable combination of ore-controlling and ore-localizing factors, as well as
the development of ore-bearing rocks. For the evaluation of zones, stratigraphic,
paleotectonic and lithological-formational criteria are used (Bogdanov et al.
1973, Kutyrev 1984).
Major Types of Copper-Bearing Zones in the Soviet Union 501

In the section of structural-formational complexes, several (two to three)

copper-bearing intervals are distinguished, forming in plan extensive belts of
copper-bearing coastal-marine deposits (Kodar-Udokan zone in the North Trans-
baikal area, Teniz-Dzhezkazgan zone in Central Kazakhstan, Copperbelt in
Central Africa, etc.). Cupreous sandstone is commonly confined to lagoonal-
deltaic deposits, accumulated on the slopes of paleouplifts, defined on the basis
of thickness variations, changes in lithological composition, gradation from red
beds to grey beds and other facies transitions. In the Kodar-Udokan zone, six ore
areas are distinguished within two belts.
The main prognostication criteria for large-scale exploration are
stratigraphic, paleotectonic, lithological, and mineralogical-geochemical ore-
controlling and ore-localizing factors. A more detailed study of copper-bearing
belts allows the exploration team to distinguish and trace the ore-bearing horizon
and to give their preliminary evaluation, as well as to outline sites where
productive strata of deposits, i.e. potential ore fields, are formed. Particularly
favourable for copper accumulation are lenses of lagoonal-deltaic deposits. They
are characterized by a rhythmic recurrency, expressed in alternation of cross-
bedded sandstone and conglomerate-breccias with siltstone and mudstone.
Relative to paleouplifts a progressive decrease in age of cupreous sandstone and a
mineral zoning - chalcocite-bornite-chalcopyrite-pyrite - are observed. The
following expected morphological and petrological types of copper sandstone
deposits are established: Dzhezkazgan, Lubin-Sieroszowice, Dzhilanda and Cis-
Uralian.
Copper sandstone of an arkose composition is characterized by monometallic
ores. Besides copper, they also contain minor admixtures of silver, molybdenum,
bismuth and other metals (Udokan, Ainak in Afghanistan, etc.). In cupreous
sandstone of a greywacke composition, ores frequently have a polymetallic com-
position. Along with copper, they often contain appreciable concentrations of
cobalt, lead, molybdenum, rhenium, silver and other metals (Dzhezkazgan,
Unkur in the North Transbaikal area). Copper mineralization is commonly
accompanied by authigenic mineralization in the form of hydromica, carbonate,
zeolite and pyrite.
The cyclic pattern of molasse cupreous sediments determines the morphology
of productive strata, their simple or multi cyclic structure. Flyschoid sediments
contain copper sandstone with a simple structure of productive strata, having
relatively small size. The latter are determined by the dimensions of lenses of
lagoonal-deltaic or other coastal-marine deposits.

Conclusions

1. The territory of the Soviet Union is characterized by copper-bearing zones,

differing greatly in their geotectonic position. Orogenic, platform and geosyn-
clinal zones are distinguished. Copper-bearing zones in orogenic and platform
areas are characterized by variegated and red deposits, in arid environment
sediments; those in geosynclines, by black shale formations.
502 Y. V. Bogdanov

2. Major epochs of sedimentary copper accumulation are: Early and Late

Proterozoic and Late Paleozoic which favour the emplacement of orogenic
cupreous sandstone, and Middle Mesozoic, which favours the generation of
copper-pyrrhotite mineralization in early geosynclinal black shale trough zones.

3. Copper-bearing zones of arid red beds are divisible into the following types
and subtypes based on predominance of deposits with different lithological com-
positions and cyclic recurrence.

Copper and Polymetallic Mineralization in Red and Variegated Sediments of a

Greywacke Compostion

Based on cyclic recurrence:

1. Thin and medium cycles (flyschoids) of Sary-Oba, etc.; Dzhilanda group,
Unkur, etc.; Kodar-Udokan zone, etc.
2. Thick cycles (molasses).

Based on facies:
1. Lagoonal-deltaic (Dzhezkazgan, etc.).
2. Lacustrine-alluvial (Fore-Urals, etc.).

Based on the structure of productive strata of deposits:

1. Multicycle (Dzhezkazgan, etc.).
2. Simple structure (Western Darwaza, etc.).

Copper Mineralization in Variegated and Red Sediments (molasses) of an Arkose

Composition, Having Thick Cycles
Based on facies:
1. Lagoonal-deltaic (Udokan, etc.).
2. Lacustrine-alluvial (Central Asia, Fore-Urals, etc.).

Based on the structure of productive strata:

1. Multicycle (Udokan, Ainak, etc.).
2. With a simple structure (Igarka area, Copperbelt in Central Africa, etc.).

Types of copper-bearing zones with a two-stage structure are established. Up

the section, flyschoid sediments of a greywacke composition grade into molasse
sediments of a greywacke (Dzhezkazgan area) or arkose composition (Kodar-
Udokan zone).

4. A prognostic evaluation of copper-bearing areas is based on formational

analysis accounting both for specific features of lithological composition and
rhythmic recurrence of formational units. The areal character of regional prog-
nostication determines the following procedure: from establishing and studying
Major Types of Copper-Bearing Zones in the Soviet Union 503

favourable geological settings and formational combinations to discovering

potential or real ore-bearing areas.

References

Bogdanov YuV (ed) (1966) Copper deposits of the Olekma-Vitim mountain land. Geology and the
regularities of their distribution. Nedra, Leningrad, 385 pp (in Russian)
Bogdanov YuV, Feoktistov VP (1982) A genetic model of copper sandstone deposits of the Udokan
type. Trans USSR Acad Sci 263,4:949-952 (in Russian)
Bogdanov YuV, Bur'yanova EZ, Kutyrev EI, Feoktistov VP, Trifonov NP (1973) Stratified copper
deposits in the USSR. Nedra, Leningrad, 312 pp (in Russian)
Kutyrev EI (1984) Geology and prognostication of conformable copper, lead and zinc deposits.
Nedra, Leningrad, 248 pp (in Russian)
Mendelsohn F (ed) (1961) The geology of the Northern Rhodesian Copperbelt. Mc Donald, London
Narkelyun LF, Salikhov VS, Trubachev AI (1983) Copper sandstones and slates of the world. Nedra,
Moscow, 414 pp (in Russian)
Savin SV (1978) Geological structure of the Near-divide metallogenic zone of the Great Caucasus.
Geol Ore Dep 20, 6: 13 - 23 (in Russian)
Volnov BA, Leleshus VL, Provotorov NG (1978) Copper presence in the south-western Darvaza.
Trans Acad Sci Tadj SSR 3:37 - 40 (in Russian)
Red-Colored Terrigenous Sediments
Specific Copper-Forming Systems
B.B. SUSURA, V.O. GLYBOVSKY, and A. V. KISLITSIN t

Abstract

Remobilization of the geochemical background of the ore- and rock-forming

elements of red-colored copper-bearing sediments in the Dzhezkazgan ore region
has been studied. Potentialities of these deposits as specialized ore-forming
systems are shown along with the paragenesis of copper and oil formation, with
the biofactor actively participating. The systems elements are clearly dis-
tinguished in the geochemical and geophysical fields. The metallogenic classifica-
tion of the copper-bearing, red-colored formations and the stratiform copper
ores of the copper sandstones and slates type connected with them has been
worked out. It served as the basis for the formation-metallogenic analysis of the
ore content of the red-colored formations and stratolevels of various age (from
Upper Proterozoic to Neogene) in Kazakhstan. Areas of occurrence of the com-
merically perspective formations of the Dzhezkazgan type have been dis-
tinguished.
The stratiform copper mineralization of the sandstone and slate type reveals a
distinct spatial connection with a group of red-colored (variegated) terrigenous
sediments and their detritus in volcanogenic and carbonate deposits of different
age (from Early Proterozoic to Cenozoic), forming in practically all geote~tonic
zones of arid terrestrial sedimentation. The red-colored molasses retains, mainly,
a metallogenic copper enrichment, irrespective of subsequent transformations up
to the stages of deep metamorphism.
The stability of red-colored, cupreous-sedimentary rocks identifies them as
specific metal-specialized ore-forming systems and ores derived from the
remobilization of the geochemical background concentrations in the sediments.
As for the hydrogenic-epigenetic concept of the genesis of copper sandstones and
slates, it is explained by oxidizing conditions favorable for the migration of
copper and other chalcophile elements (Gablina 1983, Lur'ye 1979, Perelman
and Borisenko 1962, Rose 1976, Susura 1980). The mobilizing agents are fluid
systems of various natures which are characteristic of the ore-forming process.
The ore-concentrating factors in minable strata are geochemical hydrogen
sulphide barriers. Such barriers occur rather commonly in the red-colored
deposits. They are the syngenetic barriers, argillite and limestone intercalations

I The Kazakh Research Institute of Mineral Products, Alma-Ata, USSR

Geology and Metallogeny of Copper Deposits

(ed. by O. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Red-Colored Terrigenous Sediments - Specific Copper-Forming Systems 505

and lenses, enriched in organic matter, coal seams and scattered organic remains.
Accumulations of hydrogen sulphide, hydrocarbons, bitumens and oils generat-
ed within the sediments themselves or introduced from the adjacent sediments
serve as the epigenetic barriers.
Most detailed investigations of the hydrogenic-epigenetic concepts are exem-
plified by the Permian-Carboniferous red-colored strata in the Dzhezkazgan Re-
gion of Kazakhstan. Together with the underlying and overriding grey-colored
terrigenous-carbonaceous deposits, the latter fills the Dzhezkazgan Sarysu de-
pression. It is laterally adjacent to evaporitic sediments.
The stratiform deposits of the region occur in the lenticular zones of the grey-
colored polymictic sandstones and conglomerates alternating with the red-col-
ored aleuroargillites. Several metallogenic types of mineralization are distin-
guished. They differ in the position in the minable formation, likewise in scope,
in their ore elements, as well as in mineralogical and element zoning.
In the lower part of the red-colored sequence, minor mineralization of the Ji~
landin types has a widespread areal extent. Some medium- and small-sized depos-
its and dozens of copper showings are traced in line for more than 100 km along
the outcrop of this ore horizon at the edges of the Dzhezkazgan-Sarysu depres-
sion. The ores of this type are essentially cupriferous. The mineralogical zoning is
represented by the replacement down-dip of chalcopyrite-bornite mineralization
by chalcocite.
An ore bed of the Mansfeld type is established where the zone of the red-col-
ored deposits contacts the overlying marl deposit of Late Permian age. The mono-
metallic deposits of copper sandstones and limestones of some small-sized deposit/!
and subeconomic ore occurrences are confined to this stratigraphic level.
In the sequence of ore zones, a replacement of chalcopyrite ores by bornite-
chalcocite is observed stratigraphically upwards.
A multistage copper-polymetallic mineralization of the Dzhezkazgan type in-
volves, as a rule, almost the entire sedimentary section. Archlike deposits with
the proper outer and dissected inner outlines of boundaries encircle the consedi-
mentation brachy-anticlines. The element and mineralogical zoning is expressed
by a replacement of lead and copper-polymetallic ores by purely copper ones
down the section and down-dip. A wide scope of admixed elements is typical of
the ores. These regularities have been established for the Dzhekazgan Region and
some of its new deposits.
The background contents of the ore-forming elements (Cu, Pb, Zn, Fe, S) in
the minable red-colored deposits of the Dzhezkazgan region are close to Clark
concentrations in sedimentary rocks. The transformation of the geochemical
background is evident in the conjugate regions of the elements leaching and accu-
mulation (Susura 1980). The lenselike leaching zones evident in the red-colored
deposits of the consedimentation troughs encircle the ore accumulations within
the paleo-anticline structures. The leached rocks are characterized by the low
background (200/0 of clark) contents of copper, lead, zinc and are almost sterile
with regard to sulphur. The reverse pattern exists for iron, implying that leaching
from the cement of ore-bearing grey sandstones was followed by the formation
of the grey border along their contact with the enclosing red-colored aleuroargil-
lites (Perelman et al. 1962).
506 B. B. Susura et al.

Partitioning of organic matter is exhibited by an increase of its content from

0.05070 to 0.3% in rocks of ore zones (Arustamov et al. 1969) due to an increase
of the mobile bitumen component (the bituminosity ratio increases from 15% to
90%) (Susura 1980). Different compositions and spectrophotometric charac-
teristics of bitumens in the ore-bearing horizons at their irregular change in the
deposits ore zone are indicative of the stratified mobilization and paragenesis of
copper- and oil-forming components in the minable deposits.
Differentiation of rock-forming elements is established in the form of various
deuteric changes of the cement of granular intercalations, i.e. silicification, albit-
ization, carbonatization, chloritization and others. They are evident mostly in
the ore zones and are not typical for unmineralized red-colored sections. Silicif-
ication followed by the deposition of essentially copper ores prevailed in the ore
zone under acidic conditions. Albitization and carbonatization prevail under
neutral and alkaline conditions. Lenticular zones of an intensive epigenetic car-
bonatization or anhydritization of the cap-rock type of the oil- and gas-bearing
salt domes are clearly observed in the cement of deposits overlying the ores
zones.
The zoning revealed in the remobilization of geochemical background con-
centrations identifies the red-colored deposits of the Dzhezkazgan region as an
epigenetic stratified copper-forming system. The elements of this system (mobil-
ization, transport and concentration) are well recognized in the field (Smelov et
al. 1983). The mass transfer balance of the ore elements leached from the source
rocks is comparable to the ore reserves.
A comparison of the paleo-hydrodynamic conditions with the isopleth maps
of the elements and with deuteric changes of the rocks of the minable complexes
indicates a major role in the ore genesis of the eliasite stage in the development of
artesian structures (Susura 1980). The post-lithification folding and dislocations
with a break in continuity intersect the ore zones without influencing the intensity
of mineralization on the mineralogical and element zoning.
A thermobarogeochemical study of the fluid inclusions of the mineral para-
geneses of ores and rocks of the Dzhezkazgan region indicates the unique gas
regime of different formation areas during ore deposition. The aqueous inclusion
phase is geochemically analogous to the current formation waters of the red-
colored exogenetic deposits (J 180: -11 to -15.5 per mil: D: - 88 to -120 per
mil). From the temperatures of homogenization, a trend is established in the
stratigraphic section in good agreement with the geothermal gradient of the
Earth. At the depth of maximal burial of the minable deposits (about 2 to 3 km),
the geothermal gradient probably determined the temperature of ore formation
(from + 100° to + 140°C). The hypergenic nature of mineralization is confirmed
by the argon isotope (prevalence of the gaseous component) of the fluid inclu-
sions of the ore sulphides, as well as by the biogenic nature of the sulphur (J 34s
- 4.0 ± 52.0 per mil).
Specific conditions for the development of particular types of biocoenoses
form in different zones of the red-colored formation. Heterotrophic micro-
organisms participate most commonly in the ore formation (Susura 1980). Under
the oxidizing conditions of the red-colored zones, their aerobic biocoenoses
produce organic acids and promote the leaching and dispersion of the ore ele-
Red-Colored Terrigenous Sediments - Specific Copper-Forming Systems 507

ments in the form of organometallic compounds and complexes. Under reducing

conditions of the ore depositional zones, the sulphate-reducing and mineral-
degenerating heterotrophic microorganisms are most common. Their activity rise
to the formation of hydro sulphuric barriers and mineralization of the
organometallic compounds.
Figure 1 gives a probabilistic model of the ore strata in the Dzhezkazgan
region. During the lithification of the sediments and under geostatic loads, the
gas and liquid component segregates into the granular headers and migration
occurs from the consedimentation troughs to the uplifts. The organic compounds
of the oil series, responsible for the change from oxidizing conditions to reducing
conditions, accumulated stage by stage in the relaxation centres. The organic
matter resulted in iron reduction (grey color appeared) and an active biogenic
reduction of sulphate. A sequential sedimentation of chalcophile elements from
the metal-bearing formation waters was due to the concentration growth of H~
according to the solubility products of the sulphide. The latter caused the element
and mineralogical zoning of the ore zones. The mineralization of the Jilandin ore
horizon is affected markedly by the underlying carbonate deposits which acted as
a probable source of organic matter and! or hydrogen sulphide. The localization
of ore of the Mansfeld type was determined by the syngenetic barrier in the zone
of facies transition of the red-colored deposits into the overlying carbonate beds.
It may be assumed from the general metallogeny of the red-colored forma-
tions of the Dzhezkazgan type that the genetic concepts are generally similar for
other specific stratified accumulations of economic minerals, i.e. manganese,
strontium, native sulphur, asbestos, oil and gas. Most of these deposits are
derivatives of the selective remobilization of the geochemical background con-
centrations of elements in the enclosing sediments.

Fig. 1. Hydrogenic-epigenetic copper ore in the Dzhezkazgan region. 1 Directions of geostatic

pressure; 2 directions of movement of gas-liquid ore-forming solutions; 3 directions of inflow of
bitumens and hydrogenic sulphide; 4 clay-carbonate deposits; 5 red-colored terrigenous deposit; 6
terrigenous-carbonate bituminous deposit; 7 provenance area; 8 ore zones; 9 region of epigenetic
rocks growing grey; 10 evaporites
508 B. B. Susura et al.

In contrast to the terrigenous deposits of the Dzhezkazgan type, there are a

large number of small, copper depositional centres in the red-colored carbonate-
terrigenous sediments. The concentrating components are carbonified organic re-
mains (the Urals type of the copper-bearing sediments and ores) scattered in the
minable deposits, as well as intercalations and lenses of limestones, coaly argil-
lites and coals (Donbas type of formations and ores). These accumulations fall
short of an industrial scale. Such sediments do not attain the copper contents of
the Jilandin and Mansfeld ore beds. Evidently, the ore horizons of a series of
copper showings of the above-mentioned types are attributed to continental
sabkha processes (Remfro 1974) and accumulation of copper-bearing peat bogs.
The copper-bearing character of the red-colored volcanogenic-sedimentary
mola~se is mainly defined by the metallogenic type of their terrigenous compo-
nent. The influence of the effusive strata is shown in native copper mineraliza-
tion.
At a later stage in the development of artesian structures, many copper-bear-
ing beds (e.g. weathering crusts, karsts, eolian and vein deposits, oxidized and
secondary sulphide enrichment zones) are related to infiltrations and reactions in
the red-colored deposits. In most cases, they are indicators of the primary ore,
but some may be of independent commercial importance (weathering crusts,
karsts).
The copper-bearing metallogenic strata occupy a very definite position in the
stratigraphic sequence of the red-colored deposits. The geotectonic series reflects
a sedimentogenetic change from geosynclinal regions to the platform areas. The
position in the facies change defines the formation structure, i.e. a set of sedi-
ment-forming rocks in a typical spatial relationship and, consequently, forming
stratified ore runs. Asymmetrical and symmetrical facies changes within the sedi-
ments are distinguished. The former are formed in depressions within closed ter-
minal basins, the latter occur at the transition into the sea basin. A climatic series
reflects climatic conditions relating to the erosional and sedimentation regions. A
mutual intersection of geotectonic and facies conditions forms a metallogenic
type of deposit and defines specific features of its ore content (Fig. 2).
Such classifications of red-colored, copper-bearing sediments and ores con-
nected with them is necessary for predictive metallogenic analysis, as only some
of the mineralization is of commercial importance. The formations of the Dzhez-
kazgan type, including the major commercial types (namely, the Dzhezkazgan,
Man~feld and Jilandin deposits), are most prospective. Weathering zones and
volcanogenic-sedimentary formations are of special concern.
From the metallogenic point of view, Kazakhstan is a unique region of strati-
form copper deposits. Red-colored, copper-bearing formations of various ages
(from Upper Proterozoic to Neogene), and of variable size (some large and a
great number of stratified ore showings) are widely spread in the region. Time
and again, the red-colored sedimentary areas ranged for hundreds of thousands
square kilometres to cover different geotectonic zones. A complete metallogenic
variety of sediments and related ores is known in the region.
Generally, Upper Proterozoic red-colored deposits form asymmetrical series.
South of the volcanogenic complexes of the Kokchetav anticlinorium and via the
region of variegated volcanogenic-sedimentary and terrigenous-carbonate depos-
Red-Colored Terrigenous Sediments - Specific Copper-Forming Systems 509

n 0
~~,,---------------A'
nn____________ ~r- ______ ~ ______

Fig. 2. Metallogenic series of red-colored copper-bearing strata. 1 - 6 Metallogenic types of forma-

tions: 1 volcanogenic-sedimentary; 2 evaporite; 3 Dzhezkazgan; 4 the Urals region; 5 Donbas;
6 weathering zones and karsts; 7 terrigenous-carbonate deposits; 8 volcanogenic strata; 9 basement
rocks; 10 copper mineralization. Types of mineralization: I Dzhezkazgan; II Mansfeld; III Jilandin;
IV volcanogenic-sedimentary; V weathering zones; VI the Urals region; VII Donbas. Geotectonic
position of sedimentary zones: BII volcanic belt; n n foredeep; 0 orogene; n platform

its, a change occurs in the thick terrigenous red-colored deposits of the Karatau-
Talas zone and Chinghiz region. The latter resembles the Dzhezkazgan type
formation.
The major areas of the red-colored sedimentation in the Silurian include
appreciable zones of the Earth's crust, stabilized at the late geosyncline stage of
Caledonides development.
In Early Silurian, they delineated the Balkhash massif which retained a
marine regime. In Late Silurian, this epicratonal sea closed and the marine sedi-
mentation changed to a continental type. As a whole, it may be stated that the
areal constriction of the Silurian red-colored sediments coincided in time with the
final epicentre in the Balkhash region. The lithofacies of rocks and the concen-
trating character for copper mineralization suggests that these deposits belong to
the adjacent Donbas-Urals region - volcanogenic-metallogenic type.
The occurrence of the red-colored Devonian molasse was defined by the
development of orogenic-volcanic belts of the Urals and Central Kazakhstan.
These deposits form symmetric series of sediments in the orogenic intermountain
area of the Urals-Mongolian folded belt and an asymmetrical series in the Urals
trough and in the Western European platform (Fig. 3a). In the central part of the
orogenic area, in the terminal basin (epicontinental marinelake), terrigenous-
evaporite strata were deposited in the form of a giant lense occupying a central
part of the Chu-Sarysu depression. At the periphery, they grade into the ter-
rigenous strata of the Dzhezkazgan and then of the Urals region and Donbas
types. These regions are overlain by the red-colored volcanogenic-sedimentary
complexes of the Devonian volcanic belt of Central Kazakhstan and Mugodjary.
In the Urals-Caspian region, a change occurs from the thick greywacke series of
the Urals mountain structures to the fine-grained terrigenous-carbonate deposits
of the Caspian region depression.
Several hundreds of stratiform copper showings of various types are found in
the Devonian strata. In the upper part of the strata which contact the overlying
510 B. B. Susura et aI.

Fig. 3a, b. Areas of (a) Devonian and (b) Carboniferous-Permian red-colored copper-bearing
deposits of Kazakhstan. 1 Grey-colored sea deposits; types of red-colored terrigenous sediments:
2 Dzhezkazgan; 3 the Urals region; 4 evaporite; 5 coal-bearing; 6 volcanogenic-sedimentary; 7 vol-
canogenic deposits; 8 regions of no sedimentation

carbonate beds of Late Devonian age, a copper-bearing horizon of the Mansfeld

type is traced everywhere. Favorable conditions for the localization of com-
mercial mineralization are observed in strata of the Dzhezkazgan and volcano-
genic-sedimentary types in Ulutau, in Karatau, in the Chu anticlinoria and in the
Shyderti-Chinghiz Tarbagatei region.
In the Famennian-Tournaisian period, red-colored sedimentation diminished
sharply. This beds of the Donbas and the Urals type with insignificant copper
showings are known only in the Chu depression and its margins.
Red-Colored Terrigenous Sediments - Specific Copper-Forming Systems 511

The area of red-colored Carboniferous-Lower Permian sediments includes

vast areas of Kazakhstan. Generally, these areas form a symmetric series of the
intermountain sediments between the Valerianovsk and Balkhash-Late Paleozoic
volcanogenic belts (Fig. 3 b). In the central part of the area, evaporites of the
inner zones of Dzhezkazgan-Sarysu and Chu depressions are isolated. They are
bordered by the sediments of the Dzhezkazgan metallogenic type. In the west and
southeast the latter grade into volcanogenic-sedimentary units, and in the north-
east, they grade via the Urals region-Donbas strata into the deposits of the
Karaganda and Kuzbas coal-bearing basins. In the north and in the west, the red-
colored deposits are displaced by the terrigenous-carbonate facies of the Caspian
region and the West Siberian platform.
The Dzhezkazgan type of sediment of the area mentioned encloses the
Dzhezkazgan deposits and a number of medium-sized deposits of the Jilandin
and Mansfeld types. The major commerical prospects are attributed to this ore
horizon, particularly in the Chu-Sarysu depression and in the Turgai trough.
In the Permo-Triassic, the main areas of the red-colored sedimentation have
shifted westward to the Kyzyl Kum, Caspian and West Urals regions. Red-
colored sedimentation has been observed in the Teniz, Chu and South-Balkhash
depressions and in the Bakanas synclinorium. These deposits resemble the
Donbas and the Urals type including a great number of insignificant copper
showings. Only Lower Triassic deposits in Mangyshlak and Permian red-colored
deposits in the intersalt dome troughs of the Caspian region depression are close
to the Dzhezkazgan type of the copper-bearing sediments.
For the red-colored Meso-Cenozoic deposits of Kazakhstan, one may outline
a metallogenic analogy with the coeval formations of the Tajik zone and Fergana
depression known for some copper showings of the Donbas the Urals type.
The Pre-Cretaceous Period is indicative of a widespread development of
weathering zones. The processes were most intensive in the Turgai region which
is a unique province characterized by the development of bauxite, nickel, copper-
bearing and polymetallic weathering zones and karsts. A number of karst cavities
in the Arkalyk region have formed a copper-bearing stratolevel of the Mansfeld
type. In the products of re-deposited weathering crusts filling these funnels,
copper ore accumulations have been revealed.
The metallogenic aspects considered of the red-colored formations and of
various types of specialized copper-forming systems of the typification-metal-
logenic analysis of the red-colored deposit areas may be used in the study and
evaluation of the stratiform copper content of other regions.

References

Arustamov AA, Esenov ShE, Parshin OB, Shtifanov VI (1969) Copper sandstones of Northern
Dzhezkazgan. Nauka, Alma-Ata, 137 pp (in Russian)
Bogdanov YuV, Burianova EZ, Kutyrev EI, Feoktistov VP, Tifonov NP (1973) Stratified copper
deposits in the USSR. Nedra, Leningrad, 311 pp (in Russian)
Oablina IF (1983) Conditions of copper accumulation in red-colored continental formations. Nauka,
Moscow, 110 pp (in Russian)
512 B. B. Susura et al.: Red-Colored Terrigenous Sediments

Lur'ye AM (1979) Exogenetic copper deposits in sea deposits. Author's abstract of a doctor's of
geology and mineralogy. Thesis, Moscow, 40 pp (in Russian)
Perelman AI, Borisenko EN (1962) Essays on copper geochemistry in hypergenesis zone. Proc IGEM
AN SSSR 70:30-99 (in Russian)
Remfro AR (1974) Genesis of evaporite associated stratiform metalliferous deposits - a sabkha
process. Econ Geol 69, 1:33 - 45
Rose AW (1976) The effect of cupreous chloride complexes in the origin of red-bed copper and
related deposits. Econ Geol 71, 6:1036 -1048
Smelov AA, Kolmogorov VYu, Junusov TD (1983) Petrophysical model of the ore-forming system of
the Dzhezkazgan region and its prediction- and search importance. In collection: Metallogeny
problems in Kazakhstan. Nauka, Alma-Ata, pp 194-207 (in Russian)
Susura BB (1980) Geochemistry of the epigenetic ore formation in the red-colored formation of the
Chu-Sarysu depression in Kazakhstan (type of copper sandstones). Author's abstract of a
candidate's of geology and mineralogy thesis. Moscow, 26 pp (in Russian)
Cupriferous Sandstones and Shales of the Siberian
Platform
F.P. KRENDELEVt, L.F. NARKELYUN 2, A.I. TRUBACHEV 2, V.S. SALIKHOV 2,
P.N. VOLODIN 3, V. V. KUNITSIN 4, V.S. CHECHETKIN 4, and N.N. BAKUN 5

Abstract

The Siberian platform is host to a variety of mineral deposits, including strati-

form copper deposits in sandstones and shales. Six major copper provinces are
recognized (Prieniseyskaya, Prisajanskaya, Pribaikalskaya, Prialdanskaya, Pri-
verkhojanskaya, Prianabarskaya), and within these at least 15 major zones of de-
posits and occurrences, including the Udokan district. The cupriferous sediments
occur in Proterozoic and Paleozoic detrital sediments ranging from sandstone to
shale, usually either varicoloured or black. Some of the sediments contain carbo-
nate (cement) and some contain volcanic material. The cupriferous sediments are
localized around the border of the platform in basins filled with sediment shed
from adjacent uplifts. More central parts of the platform contain lead-zinc,
barite, fluorite, gypsum-anhydrite, and other evaporite minerals, whereas gold
occurs in conglomeratic zones closer to the uplifts. The size and shape of ore
zones depend closely on sedimentary facies. On a basin scale, the cupriferous
sandstones and shales transgress into younger sediments toward the center of the
basin, and are mineralogically zones from underlying pyrite and chalcopyrite in-
to overlying bornite-chalcocite zones. Similar circumferential and beltlike zones
of cupriferous sediments are recognized around and adjacent to other platforms
on all continents. The deposits were formed during sedimentation and diagene-
sis, but have since been subjected to various combinations of epigenesis, meta-
morphism and weathering. The sources of copper are considered to be in the up-
lifted blocks that supplied the sediment, and in volcanic effusions.
The Siberian Platform and its folded border zones is one of the most impor-
tant metallogenic regions in the world. It is formed almost entirely of sedimen-
tary formations ranging in age from Archean to Cenozoic and including various
mineral deposits (Bogdanov et al. 1973, Narkelyun et al. 1977). Cupriferous
sandstones and shales are confined to Proterozoic and Paleozoic formations of
the platform where the Prieniseyskaya, Prisajanskaya, Pribaikalskaya and Prial-
danskaya copper provinces have been established (Fig. 1).

Institute of Natural Resources, Butin Street 26, Chita, USSR

Chita Polytechnic, Chita, USSR
Central Research Institute of Geological Prospecting, Moscow, USSR
4 Chitageology, Mingeo RSFSR, Chita, USSR
5 All Union Institute of Research in Petroleum, Leningrad, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G~ H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
514 F. P. Krendelev et al.

~
~
~-=-
---
-:.-_---
--------)
-c~
J:s~ ..~
,-- 1150\1

~
~ -~'

.. ~

!-.:
~
~

'.
.~

= ' ''"l ~ ,\
~
==>.:.1:'
iotn::l ~ ~~ ;X;

~
~.j

~
''', ~
~-
~.

~1 EE3l .....;:. '~-~

'- - - -i

~~412ZJ5
1:::1, [ll]1 ga
.....~ ...,..' ...... '

. 0,
km 200 zoo, 400~m
1

!S:]lo[IJlIl- 112~I~rn14
[YJ'5~16017BI0E39 I IV, l lzo
Fig. 1. Location of cupriferous sandstones and shales on the Siberian platform. The tectonic base is
simplified after Yanshin (1966).1 Basement uplifts; 2 cover of ancient platform; 3 Baikalides; 4 Cale-
donides; 5 Paleozoic basins; 6 Hercynides; 7 cover of epipaleozoic platform; 8 Mesozoides; 9 volca-
nogenic sedimentary complex; 10 zones of cupriferous sandstones and shales. 11 -17 Deposits and
occurrences of copper mineralization according to different metallogenic epochs: 11 Proterozoic; 12
Vendian; 13 Cambrian; 14 Ordovician; 15 Silurian; 16 Devonian; 17 Carboniferous. 18 Boundaries
of copper provinces; 19 copper provinces (Roman numerals) and zones (Arabic numerals) within the
provinces. I Prieniseyskaya: 1 Igarskaya, 2 Priangarskaya. II Prisajanskaya. III Pribaikalskaya: 1
Zapadno-Pribaikalskaya, 2 Verkhne-Lenskaya, 3 Angaro-Tungusskaya. IV Prialdanskaya: 1 Koda-
ro-Udokanskaya, 2 Kondo-Karengskaya, 3 Nuyjsko-Berezovskaya, 4 Olekmo-Tokkinskaya, 5 Vos-
tochno-Aldanskaya. V Priverkhojanskaya. VI Prianabarskaya: 1 Prianabarskaya proper, 2 Keneli-
kanskaya, 3 Olenekskaya. VII Severo-Zemelskaya zone

Early Proterozoic copper deposits are the most ancient on the platform (the
Kondo-Karengskaya, Olekmo-Tokkinskaya, Kodaro-Udokanskaya and Vos-
tochno-Aldanskaya zones). Copper-bearing minerals are widely distributed in
varicoloured detrital and volcano-detrital sediments in Riphean and Vendian
strata of all the provinces. Early Cambrian cupriferous sediments are known
throughout the platform. Middle and Late Cambrian copper-bearing sedimen-
tary rocks occupy considerable areas. During Ordovician time copper accumulat-
ed in the Priangarskaya, Nyujsko-Berezovskaya and Angaro-Tungusskaya prov-
inces. In the Priverkhojanskaya province copper is confined to the Silurian effu-
sive-detrital suite. Devonian carbonate-detrital and effusive-detrital sediments
Cupriferous Sandstones and Shales of the Siberian Platform 515

have been established in the Priverkhojanskaya province as well as in the Vos-

tochno-Aldanskaya zone and in the Ribinskaya depression of the Prisajanskaya
province. The copper-bearing section terminates with Early Carboniferous effu-
sive-detrital formations.
Thus, cupriferous sandstones and shales of the Siberian platform are to be
found in sedimentary formations ranging in age from Lower Proterozoic to Car-
boniferous, their vertical stratigraphic thickness being 50 km. The mineralization
is localized within narrow age intervals and is characterized by distinct strati-
graphic control.
Accumulation of the cupriferous sedimentary formations took place within
the platform depressions such as those of Kodaro-Udokanskaya and Ugujskaya,
as well as in orogenic structures such as the Prieniseyskaya, Prisajanskaya, Pri-
verkhojanskaya and Angaro-Lenskaya foredeeps and in marginal parts of the
platform structures. All ore-bearing structures are located mainly in transition
zones from geosynclinal to platform formations. This localization in transition
zones is the essence of geotectonic control in the distribution of cupriferous sedi-
ments.
The copper deposits are closely connected with the host sediments. Various
cupriferous sedimentary formations can be distinguished by their lithological as-
sociations. Varicoloured, low-carbonate sandstone, varicoloured calcareous
shale, grey low-carbonate sandstone, and varicoloured volcanogenic-detrital cu-
priferous sedimentary formations are the most important lithological types. That
the copper mineralization is confined to certain formations is the most important
control on the location of copper deposits in the sedimentary suites of the Siberi-
an platform.
Cupriferous sedimentary formations have vertically and laterally a genetic
connection with other sedimentary formations of the platform containing metals
and minerals such as lead, zinc, gold, rare metals, phosphorus, gypsum, anhy-
drite, barite, fluorite, celestite and rock and potash salts. The coarse detrital sedi-
ments are confined to the peripheral parts of the platform, adjacent to mountain
structures that were the regions of erosion. Away from the mountains towards
the internal parts of the platform, there are in sequence: detrital, carbonate-detri-
tal, argillaceous-carbonate, carbonate-sulfate and saline formations. Gold min-
eralization is mainly restricted to the coarse detrital sediments; copper mineral-
ization is confined to low-carbonate and carbonate-argillaceous sandstones;
lead-zinc mineralization is restricted to argillaceous carbonate and carbonate;
and mineral salts to sulfate-saline formations (Fig. 2).
On the Siberian platform copper has been recognized in conglomerates 6,
gravelites, sandstones, aleurolites, argillites, argillaceous limestones, dolomites
and tuffaceous rocks. According to lithological characteristics we differentiate
three main types of copper deposits: (1) cupriferous conglomerates and sand-
stones - subcontinental deposits containing carbonate material up to 15010, in-
cluding allochthonous organic substance (Early Cambrian formations); (2) cu-
priferous sandstones and siltstones of subcontinental, lagoonal, deltaic and lit-
toral-marine types containing autochthonous and allochthonous organic matter

6 Probably pebble conglomerates.

516 F. P . Krendelev et al.

A B C
I II ill IT V. 'ill 1lII v.m IX X XI XII x:m
c, v v
it@. it ~
D R-Cc~· v v :-@:-:
S vvC~vv
O2-3 [#. Mj::±:·
0, ~ PII-Zn-C~J Y"l'y
el •
lil-Z·Bl ~.
£2
q=.:± ;::{Cll-
• •r r r r
(, <> 0 0 ALl 0 0
:~.pHi1
r r r
Pe-ln
V :'.::-.:.:.- :- ~:'-:.""". - ."-:. :' ~
:-®.ti1'8-zn .6: - ~@~ ~

~
· Cu. ~-· . ~-P'I-Zn

R ~. Ifi .PJ~n· ••

[email protected]:
o
oAu.•• •~ 'o®
A '-v@/
. Cu v. o 0 V m
Pit!
<>

o
•
0

0
@
0

0
0

•
0

0
..-@::
~
. : - '.'@ -- ... .:
- . . -
•. _ ' . .
.~
A

I:-@·: Il ~ 2 IvV@)/I~ ~ 4 ~~ ~6
lo.@)o 017 1:=@~.J8 1:@:J9 ~ !O ~ 1 1

Fig: 2A - c: Stratigraphic-tectonic location of copper-sedimentary formations of the Siberian plat-

form and their paragenetic associations with other mineral deposits. Sedimentary formations, cuprif-
erous formations : 1varicoloured-detrital; 2 varicoloured, carbonate-detrital; 3 volcanogenic detrital;
4 detrital-carbonate (with Pb and Zn), Lead-zinc; 5 carbonate; 6 argillaceous-carbonate. Gold-bear-
ing: 7 conglomeratic-sandstone; 8 carbonaceous-argillaceous (black shale); 9 rare metallic-flyschoid;
10 sulphate-carbonate (with fluorite , celestite, barite); 11 sulphate-saline. A Interplatform-fold ba-
sins of the Olekmo-Vitimskaya mountain country. Zones: I Kondo-Karengskaya; II Olekmo-Tokkin-
skaya; III Kodaro-Udokanskaya. B Orogenic structures framing the platform: IV Igarskaya zone; V
Priangarskaya zone; VI Prisajanskaya province; VII Pribaikalskaya province; VIII Vostochno-Al-
danskaya (Eastern Aldan) zone; IX Priverkhojanskaya province. C Platform structures (edge zones
of the platform): X Vostochno-Aldanskaya (Eastern Aldan) zone; XI Verkhnelenskaya (Upper Lena
River) and Angaro-Tungusskaya zones; XII Priangarskaya zone; XIII interior parts of the platform

(Kodaro-Udokanskaya zone); (3) cupriferous shales, marls, argillaceous lime-

stones and dolomites of lagoonal and littoral-marine origin containing about
40% carbonate with autochthonous organic matter in the form of algae, plank-
ton, bitumen and microorganisms (Late Cambrian and Ordovician formations):
Sedimentary facies have a decisive influence on the location, morphology, pa-
rameters and composition of ore bodies in cupriferous sandstones and shales.
For example, the thick ore bodies of the Udokan copper deposit are associated
with the submarine-deltaic facies, whereas the smaller ore bodies, containing py-
rite and chalcopyrite, occur in a bay-lagoonal facies.
Cupriferous Sandstones and Shales of the Siberian Platform 517

A B c
II: illN II TIf N 11 ruN

Fig. 3A - C. Location of lithological-facial types of deposition in the ore-bearing section of Udokan

copper deposits (cross-section). Ore-synclinal: A south flank; B axial zone; C North flank . I Strati-
graphic units: sk3 Upper Sakukanskaya. Horizons: sk} under ores; skJ ore-bearing; sic} over ores; nm
Namingskaya. II Lithological types of rocks: 1quartzitic sandstones, vaguely layered and massive; 2
sandstones with magnetite; 3 cross-bedded quartzitic and calcareous sandstones; 4 sandstones with
thin cross-bedding, or wavy or lenticular-wavy layering; 5 coarsely cross-beded calcareous sandstones
with fragments of argillaceous rocks; 6 horizontal and cross-wave argillites; 7 siltstones. III Deposi-
tional environments of sediments: 8 periodically dried bays and lagoons; 9 shallow littoral zone of sea
with relatively calm sedimentation; 10 lagoons; 11 bays; 12 subaerial part of delta; 13 submarine part
of delta; 14 shallow seawaters. IV Mineralogical types of ores: 15 bornite-chalcocite; 16 chalcopyrite;
17 pyrite mineralization; 18 boundaries of ore-bearing horizon; 19 boundaries of depositional en-
vironments

Together with the transgression of favourable facies from regions of erosion

towards the basin of sedimentation, there is a migration of mineralization up-
wards in the section (Fig. 3). A close connection of copper mineralization with
certain facies and the age transgression of the copper mineralization has been de-
termined for many stratigraphic levels and zones of the Siberian platform and
other regions of the world.
Mineralogical and geochemical features of the copper deposits are quite spe-
cific, and that is why they differ distinctly from other genetic types of copper de-
posits. More than 50 ore minerals have been identified in cupriferous sandstones
and shales. Most widely developed are chalcocite, bornite, and more rarely chal-
copyrite, which generally forms zonal patterns, both vertically and laterally.
 VI
00
-

Table 1. Mineralogical and geochemical zoning of copper sandstones and copper shales deposits

Redox potential Ore zones Chemical forms of iron Content of re- Ore minerals Impurity - ele- Lithological and
of ore-forming sidual organic ments and their facial types of
facies In nonmetamor- In metamorphic carbon (0/0) mineral forms ore-bearing sedi-
phic rocks rocks mentary rocks

Oxidizing Barren zone Limonite is Magnetite and From O.On to Chalcocite (in Absent Red and red-
widely distribut- hematite are O.n small amounts); grey argillaceous
ed quite common native copper continental la-
(sparse) goon sandstones,
often indicating
traces of salinity

Slight reduc- Chalcocite Limonite; hema- Magnetite, From O.n to n Chalcocite; bor- Native elements: Grey, dark-grey
tion tite (rare) hematite, chal- nite (rare) Ag, Au, Hg, Bi, and reddish grey
cocite platinum group sandstones,
elements dark-grey silt-
stones, argillites
Reduction Bornite Limonite and Magnetite, Bornite, chalco- Sulphides of Pb, and carbona-
~
hematite are rare hematite; bornite cite; betechtinite Mo, Re, Bi (in ceous rocks :-c
(sporadic) and Cu-sulpho- wittichenite) As, (rare), deposited ..,::0:::
salts are rare Sb in delta or bay- CD
::s
Po
lagoon environ-
ments in active ~
<
hydrodynamic ~
conditions ~
 (j
~
Chalcopyrite Protoxides of Fe Sulphides of Fe Chalcopyrite; Sulphides of Zn, Light-grey, grey
(carbonaceous, predominate; bornite and Co, Hg. Isomor- and dark-grey
"8.
biotite, magnetite and pyrite (more phic impurities fine-grained [
glauconite) and hematite are in rarely); galena of Cd, Ge, Zn sandstones, silt- '"Vl
sulphides of Fe small amounts and sphalerite stones, argillites §
are widely (rare) and carbona- ~
spread; limonite ceous rocks de-
is practically ab- posited in deltas, ~
'"
sent bays and la- [
goons in quiet Vl
Strong reduction Pyrite Sulphides of Fe Sulphides of Fe n. Pyrite; chalcopy- Isomorphic hydrodynamic
predominate; red are widely rite and pyrrho- impurities of Ni, conditions ~...,o
rocks with limo- spread; magne- tite (more rare- Co, Se, Fe, As
nite are absent tite and hematite ly); sphalerite in arsenopyrite, g-
associated with (rare) Au Vl

pyrite are not 5'

(I)
...Si·
present P
."
Very strong re- Pyrrhotite Sulphide iron predominates From n to nO Pyrrhotite Co, Ni: sulphide Dark-grey, often ~
duction associated with and isomorphic almost black 0'
pyrite graphite-bearing 8
siltstones, argil-
lites, sandstones,
and quartzites,
deposited in lit-
toral zones of
the sea under
quiet hydrody-
namic conditions

v.
IC
-
520 F. P . Krendelev et aI.

...;
~--------------------------------------~ ~
Cupriferous Sandstones and Shales of the Siberian Platform 521

Copper ores also contain various minor elements such as Pb, Zn, Mo, Ag, Cd,
Co, Ni, Bi and Au, which are forming separate minerals or occur as isomorphic
substitutions within the principal ore mineral zones (see Table 1).
Thus, the localization of cupriferous sandstones and shales on the Siberian
platform is controlled by stratigraphic, geotectonic, formational, lithologic-faci-
al, paleoclimatic and paleogeographic factors. A favourable combination of the
above-mentioned ore controlling factors has led to the formation of ore-bearing
provinces stretching along almost the entire perimeter of the Siberian platform
and making up one circumferential East-Siberian metallogenic belt of copper
sandstones and shales (Narkelyun et al. 1983).
The analysis of published data on copper sandstones and shales of the world
shows that they form circumferential and regional belts. We offer the following
classification of the distribution of the copper belts concerned (Fig. 4): (A) An-
cient platforms. (a) Closed belts on the circumference of ancient platforms [East-
Siberian, East-European, Gondwana (South America and Africa), North Ameri-
can); (b) Closed and semi-closed regional copper belts near shields (Aldansky,
Anabarsky, Ukrainian, Canadian shields); (c) Linear belts of foredeeps and the
marginal parts of the platform (foredeeps of the Siberian platform, Pricarpat-
sky, Appalachian and other foredeeps; marginal parts of the Siberian, East-Eu-
ropean and North American platforms). (B) Young platforms: regional copper
belts embracing geosynclinal massifs and prominences on the platforms (Czecho-
slovakian, French and others). (C) Geosynclinal folded regions: regional copper
belts of internal troughs and basins (Minusinskaya, Tuvinskaya, Djezkazganska-
ya, Middle Asian and other basins).
With regards to the genesis of copper sandstones and shales of the Siberian
platform, the following hypotheses exist: original-sedimentary, volcanogenic-
sedimentary, hydrothermal-postmagmatic, diagenetic-epigenetic and connate-
hydrothermal. The above-mentioned factors controlling the location of mineral-
ization as well as its genetic features such as the concordant nature of ore bodies
in the layered sedimentary suites, the multi-layer structures of mineralization
caused by the cyclicity of the rocks, the close connection of mineralization with
textural and structural characteristics of host rocks (cross-bedding, parallel-bed-
ding and lenticular bedding), concretionary and socket-shaped segregations of
ore minerals, sedimentary-diagenetic deformations of weakly consolidated ore
sediments in landslide processes and the presence of asymmetric ore zoning (Bez-
rodnikh et al. 1977), testify to the original sedimentary genesis of copper sand-
stones and shales of the Siberian platform. However, later epigenetic and meta-

Fig. 4. Location of cupriferous sandstones and shales in the principal geotectonic structures of the
world (on the basis of tectonic maps of the Physical-Geographical Atlas of the World, 1964, simpli-
fied). Ancient platforms: A Siberian; B East European; C African; D Indian; E Chinese; F North
American; G South American; H Australian. 1 Exposure of folded basement of ancient platforms
(shields); 2 sedimentary cover of ancient platforms. Baikalides: 3 exposure of folded structures; 4
foredeeps and aulacogens. Caledonides: 5 exposure of folded structures; 6 depressions. Hercynides: 7
exposure of folded structures; 8 marginal depressions; 9 sedimentary cover on Caledonian and Her-
cynian folded basement. Mesozoides: 10 exposure of folded structures; 11 intermountain depres-
sions; 12 foredeeps; 13 central massifs. Alpides: 14 regions of folding; 15 intermountain depressions;
16 marginal depressions; 17 zone of intensive Upper Cretaceous and Tertiary volcanicity; 18 copper
ore belts (sandstones and shales). 18 Established; 19 assumed
522 F. P. Krendelev et al.

morphic processes led to the rock transformation, sulphide recrystallization and

redeposition, the formation of Alpine-type veins and other changes (Bogdanov et
al. 1973, Krendelev et al. 1983, Narkelyun et al. 1977).
Popov (1964) has stated that the process of mineralization and its later defor-
mation and metamorphism comprised the following stages: sedimentogenesis,
diagenesis, epigenesis, metamorphism and weathering. The principal stages that
led to the formation of mineralization are sedimentogenesis and diagenesis; min-
eralogical-geochemical peculiarities of their development especially relating to
the deposits which distinctly marked zoning in ore minerals have been considered
in Table 1. Thus, the formation of sulphides was controlled mainly by redox re-
actions, quantity and quality of the buried organic matter and lithological-facial
conditions.
The geotectonic position of cupriferous sedimentary suites close to the anci-
ent massifs, which have above normal abundance of both copper mineralization
and copper deposits of various genetic types testifies to the fact that these massifs
may have been the source regions from which copper was transported in detrital
remains or dissolved into the basins of sedimentation where the copper-bearing
sediments accumulated.
We have come to certain conclusions concerning the sources of copper for the
specific districts of copper sandstones and shales on the Siberian platform. It has
been shown that the principal sources of copper for the Kodaro-Udokanskaya
zone were the Archean suites of the Aldansky shield and Charsky block (Bogda-
nov et al. 1973, Krendelev et al. 1983, Narkelyun and Yurgenson 1968). Convinc-
ing evidence of a close geochemical connection of the regions of erosion and the
basins of sedimentation in Pribaikalskaya, Prisajanskaya and Prienisejskaya
provinces has been given in a number of publications (Bezrodnikh et al. 1977,
Bogdanov et al. 1973, Narkelyun and Yurgenson 1968).
In all probability the source of copper for the formation of Devonian cuprif-
erous sandstones in the Priverkhojanskaya zone and Sette-Davan mountains,
may have been underwater volcanic exhalations, and for overlying copper-bear-
ing carboniferous deposits the source of copper may have been eroding effusive-
sedimentary rocks.
Thus, it can be stated that the principal sources of copper and accompanying
metals in cupriferous sandstones and shales of the Siberian platform were the ad-
jacent regions of erosion. This fact can be convincingly proven by a close metal-
logenic connection of the regions of erosion and the basins of sedimentation. Un-
derwater volcanic effusions may be considered the sources of metals in certain
zones of the platform.

References

Bezrodnikh YP, Narkelyun LF, Salikhov VS, Trubachev AI (1977) Conditions of formation of cu-
priferous sandstones in the south of the Siberian Platform. In: The basic problems of sedimentary
ore formation. Him, Frunze, pp 112 - 136 (in Russian)
Bodganov GV, Burjanova EZ, Kutirev EI (1973) Stratified copper deposits in the USSR. Nedra, Len-
ingrad, p 312 (in Russian)
Cupriferous Sandstones and Shales of the Siberian Platform 523

Krendelev FP, Bakun NN, Volodin RN (1983) Cupriferous sandstones of the Udokan Range. Nauka,
Moscow, 248 pp (in Russian)
Narkelyun LF, Yurgenson GA (1968) On sources of ore material in forming the deposits of copper
sandstone type. Lithol Ore Mineral 6:114 -123 (in Russian)
Narkelyun LF, Bezrodnikh YP, Trubachev AI, Salikhov VS (1977) Cupriferous sandstones and
shales of the southern part of the Siberian Platform. Nedra, Moscow, 223 pp (in Russian)
Narkelyun LF, Salikhov VS, Trubachev AI (1983) Cupriferous sandstones and shales of the world.
Nedra, Moscow, 414 pp (in Russian)
Popov VM (1964) Stratified deposits of nonferrous metals and the problems connected with their gen-
esis. In: Problems of ore genesis. Nedra, Moscow, pp 350 - 368 (in Russian)
Yanshin AL (ed) (1966) Geotectonic map of Eura-Asia. Moscow
Genetic Types of Copper Mineralization in the Igarka
Area, West of the Siberian Platform
I. F. GABLINA 1

Abstract

This paper discusses the general regularities of the distribution of the copper min-
eralization associated with the Vendian red-bed formation. Lean stratified min-
eralization is restricted to the zones transitional from the red-beds to the underly-
ing and overlying grey-colored rock deposits. Rich ore deposits are spatially asso-
ciated with epigenetic carbonaceous matter in the highly-permeable coarse-clastic
red-beds near the paleo-uplifts.
The epigenetic host rock alterations associated with ore deposition, and the
copper mineral zonation are discussed. A genetic classification of the copper
mineralization in the area is suggested.

Geologic Setting

The Igarka copper-ore area is located on the western margin of the Siberian Plat-
form (see Fig. 1 in Krendelev paper). According to its tectonic setting, it is posi-
tioned at the intersection of two large Precambrian structures: the Igarka Uplift
of the Baikalides, and the foredeep, surrounding it at the east. The Igarka Uplift
is a large block structure (up to 90 -100 km in length) composed of highly de-
formed and metamorphosed terrigenous-carbonate rocks of the Riphean series.
The foredeep was developing at the end of the Late Riphean. It is filled with a
red-colored carbonate-terrigenous series of the Vendian, 1300 m thick, the Izlu-
chinskaya Formation. The latter is underlain by marine terrigenous-calcareous,
carbonaceous rocks of the Riphean Chernorechenskaya Formation, which is
considered to be the major complex of oil and gas source-beds in the Precambri-
an of the Siberian Platform. At the end of the Vendian and beginning of the
Cambrian, as a result of the transgres~ion of the sea, the foredeep was over-
lapped by the Sukharikhinskaya Formation. From this period onward the area
evolved under platform conditions (Rzhevskii et al. 1980).
The Igarka Uplift is divided from the downfold by a marginal suture, along
which the Riphean basement rocks are gently overthrust into the Vendian sedi-

1 Institute of the Lithosphere, USSR Academy of Sciences, 109180 Moscow, USSR

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 525

ments that fill the foredeep (Fig. 1). The Igarka Uplift represents a folded block
structure, dissected by narrow basins into several blocks. One of these blocks
constitutes the Chernorechenskaya anticline, which is divided from the Igarka
Uplift by a trench-shaped intermontane trough (Fig. 1).
The Izluchinskaya red-bed formation is characterized by considerable facies
changes. In a northeastern-southwestern direction, from the center of the fore-
deep toward the Igarka Uplift, the facies undergo a change from calcereous-ar-
gillaceous flood-plain-lagoonal sediments to alluvial-lacustrine-silty sandstones
and then - to coarse deposits to river channels and piedmont detrital fan sedi-
ments (conglomerates, gritstones). In the same direction, the thickness of the for-
mation decreases from 1200 - 1300 m to 200 - 300 m and less.
The Izluchinskaya Formation consists typically of red-colored rocks with a
high ferric oxide content (Fepy'FeO > 1) and low organic-C values (common-
ly <0.1 0/0, seldom reaching 0.280/0), which tend to suggest that oxidizing condi-
tions prevailed in the environment of their deposition. Only the flood-plain-la-
goonal sediments, occurring northeast of the mineral estate, contain layers of
grey dolomites, limestones, and argillites that in contrast to the former, are of
high organic-C content (0.44 -1.80%).
The marine sediments of the Sukharikhinskaya Formation (450- 500 m)
which lie conformably on the red-beds, are composed of limestones and dolo-
mites, often silty and argillaceous, with sandstones and argillites intercalated in
the bottom and middle sections. At the base is a quartz-sandstone member. The
Sukharikhinskaya Formation consists characteristically of grey rocks with a high
organic-C content (0.13 to 0.75%) and a predominance of ferrous oxides
(Fepy'FeO<1).
This formation is of uniform composition. The only exception is within the
Graviika ore locus near the eastern flank of the Igarka Uplift, where all the car-
bonate facies of the lower section are replaced by sandstones of shallow water-
coastal facies. The formation is of reduced thickness in this area.

Copper Mineralization

Copper mineralization is restricted to three stratigraphic levels: the base and top
of the Izluchinskaya red-bed formation (horizons A and B) and grey rocks en-
closed in the red-beds.
Horizon A is associated with the transitional zone from the Izluchinskaya
red-beds to the underlying Riphean marine sediments. The contours on its out-
crops beneath the Quaternary encircle the Chernorechenskaya anticline (Fig. 1).
The copper-bearing horizon is of relatively uniform composition along its entire
extension, and is composed of interbedded argillites, argillaceous limestones,
marls, and black argillaceous-calcareous shales. Horizon A typically contains
uniform mineralization along the strike and low metal content (although several
times in excess of the normal Cu abundance). High copper concentrations are ex-
tremely scarce.
The mineralization is represented by finely disseminated impregnations of di-
genite (CuuS), bornite (CusFeS~, and chalcopyrite (CuFeS~ which are zoned
 1+ +11
+ + +
+ , +
+ , ~Z
I...,,] J
~,
m:] 5
rsI 6
[ZJ 7
C2J B
[Z] 9
~IO
[ ] ; ] II

12

b., I I L I . . ..
 ."
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 527

Cu-content. %
,- 3 ,- 2 0.' 1
185 -,....--~--""'--~ E2]1 [.;(·:/16

_9
~ 190
f3212 1;=;=;=1 7 . ,2
o
-"
III
B3 1:·-:.18 ~ 13
.~ c 195 CO[l t====~ L IIIllD lL
-" 0
~5 CJl0
.~ ~ 200 +-""rr"'lrrl
-" ...
~2 Horizon "8"
205 -t::n:~~ 8 Cu-content . %
di
1-3 ,- 2 0.1 1
M 100
210~~1
o
~ 215 ~ 105
~ c o
III
c:.=
0 -" c
'E 0220 ~ ,g 110
g~ 'E 0
E
o::J ...
iii 0
- - 225 ~ 2115

120
o,..,
o Horizon "A"
~ 125 -Ir:::CCJ
c
'"o
.J::.
c 130-{::;:;::~~
~ 0
0''':
C 0
~ ~
62
Fig_ 2, Vertical mineral zonation in horizons A and B. 1-8 Lithologies: 1 limestones; 2 dolomites;
3 marls; 4 siltstones, argillites; 5 carbon-bearing shales; 6 quartz sandstones; 7 gritstones; 8 conglom-
erates. 9- 10 Rock colour: 9 red, brown; 10 grey, black, green. 11-/4 Metallic minerals: 11 digenite
(dl); 12 bornite (B); 13 chalcopyrite (Cp); /4 pyrite (PI)

vertically in the sequence away from the contact with the red-beds (downward in
the section). The copper content of the ore rocks decreases in the same direction
(Fig. 2). The vertical range of copper mineralization does not extend more than
5 -15 m from the base of the red-beds. The metallic minerals occur in the form
of small (up to 1 mm) isometric concretionary aggregations of metasomatic type,
frequently surrounded by diffusion aureoles of the same sulfides (Fig. 3). The
observed relationships between the metallic mineral aggregations and the host
rocks serve to indicate that the mineral clusters formed within consolidated
sediments.
Copper-bearing horizon B is restricted to the base of the grey marine sedi-
ments of the Sukharikhinskaya Formation that rest on the Izluchinskaya red-

Fig. 1. Map and cross-section showing geologic structure and distribution of copper mineralization in
the Igarka area (adapted from V. F. Rzhevskii). 1 Rocks of the basement fold structure; 2 Chernore-
chenskaya Formation (PR2 cr); 3 Izluchinskaya Formation (PR2 is); 4 Sukharikhinskaya Formation
(V - el sh); 5 Cambrian (el-2); 6 copper-bearing horizons A and B; 7 deep-seated faults of the margi-
nal suture; 8 other faults; 9 overthrust zone; 10 exploration wells; 11 ore occurrences; I Sukharikha;
II Izluchinsk; III Rudnii; IV Chernorechensk; 12 Graviika ore deposits
 VI
~

.....
~
o
~
~.

Fig. 3a-d. The detailed features of sulfides in horizon A (a, b) and horizon B (c, d, white-digenite and bornite). Photo-
micrograph of the polished section. Photo by N. Inyashkin. Abbreviations: Di digenite; B bornite; Cp chalcopyrite
 i
n'

o
i...,
(j

~(1)
...
~
~
~
~.
o
::;
Er
Er-
(1)

f:><"
I»

~
(1)
,?'
~
1&
o
...,
Er-
(1)
(/)
CT
(1)
::1.
I»
::;
"tj

~
8'
3

v.
~
Fig. 4a - d. Photomicrographs of the specific features of the epigenetic organic matter (black) in sandstones (a, b, c) and dolomites (d).
Photo by E. Zenkova
530 I. F. Gablina

beds (Fig. 1). A distinctive feature of this horizon is the meager copper content,
persistent along the entire extent of the horizon, and local enrichment only within
discrete areas. The horizon is 10- 30 m thick. The sulfide mineralization exhibits
an irregular distribution pattern: the concentrations tend to increase towards in-
tercalated sandstones or limestones with a high arenaceous, silty content. The
.metallic minerals are represented by the same paragenetic sulfide associations as
in horizon A, but, in contrast, occur in the section in reversed order of succession
(Fig. 2). The regular distribution of the metallic minerals in the vertical section in
both horizons exhibits a symmetrical zonation in relation to the red-beds. The
metallic minerals occur in horizon B in the form of finely disseminated impregna-
tions in the ore rocks. Most typical are tiny (not exceeding 10 ~m) globular forms
or spherolites of radial-concentric structures. The rock texture is the controlling
factor in their distribution (they form pore fillings in sandstones and inclusions in
the authigenic cement), from which it can be inferred that their time of formation
was closely related to the host rocks (Fig. 3).
Several ore occurrences in the Graviika copper deposits are associated with
horizon B.
In the Graviika ore deposits horizon B shows greatly increased copper content
in the ore rocks immediately above the coarse-clastic alluvial-deltaic facies of the
underlying the red-beds. The ore body here is dome-shaped and, in general, con-
figuration matches the contours of the sandstones of the coastal, shallow water
facies. Its thickness varies from 50 - 100 m at the crest of the dome to a few centi-
meters on the flank.
The rocks of the Sukharikhinskaya Formation in the area of the Graviika ore
deposits exhibit intense alteration. These secondary alterations are reflected in si-
licification, dolomitization, and calcitization of the rocks, as well as occurrences
of epigenetic organic matter (BOM). The BOM is impregnated in the sandstones,
occurring in the form of pellicular-pore cement and discrete oval-shaped aggre-
gates, and it also penetrates into the quartz sandstones and carbonate rocks, per-
vading the numerous minor fissures, crush zones, and stylolitic sutures (Fig. 4).
The BOM is of greatest vertical range in the shoaly sandstone deposits. As a con-
sequence the latter become dark grey and in some places black in color. The pig-
mentation intensity, caused by the impregnated BOM, shows a conspicuous de-
crease upward in the section. When the sandstones undergo facies changes to
dolomites the depth of penetration of BOM into the rocks of the Sukharikhin-
skaya Formation shows a sharp decrease in different directions away from the
"sand bank", being limited, as a rule, to the basal sandstone layer (1 m thick).

Fig. 5 a, b. Stratified section of the Graviika ore deposits, contoured on the basal sandstones of the
Sukharikhinskaya Formation. a Epigenetic zonality in the wall rocks near ore. 1-3 Zones of epige-
netic alterations: 1 quartz zone; 2 dolomitic; 3 calcitic. 4 Relative content of major oxides (Si02 ;
MgO; CaO) in different alteration zones; 5 contour lines showing distribution of shallow-sandstone
facies. 6 - 9 Wells with data on organic-C content in basal sandstones of the Sukharikhinskaya For-
mation: 6 <O.lOJo; 70.1-0.5OJo; 80.5 -1 OJo; 9 >1 OJo. 10 Occurrences of epigenetic organic matter in
rocks underlying the Izluchinskaya Formation; 11 limit of altered, lighter-colored rocks in the Izlu-
chinskaya Formation; 12 limits of epigenetic alteration zones; 13 faults; 14 overthrust zone; 15 wells
(with no data on organic-C). b Metallic mineral distribution pattern. 1-5 Metallic minerals:
1 sulfides of the chalcocite series; 2 bornite; 3 chalcopyrite; 4 galena; 5 sphalerite
 Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 531

m
"4-

~.,
.:: ....... ::::::::::::::::::
.. . ........... , ...... .
.. ':
D
' "
:' .
...... . ...... . ..
'"
.. ~
..
.. I
1 f 1 )
'
d 1"

..,
B
... ~

EJ[j
Bg
'0 'It

"l !:l

rurJ
. ~

[]
., :,

[]rrJ
.. lil

~EJ
- 0,

~~ IF>

.~
'"'
532 I. F. Gablina

BOM is rare in the overlying dolomites. In the rocks of the Sukharikhinskaya

Formation, containing BOM, the organic-C content is several times higher than
normal abundance, reaching in single samples to 1.09 -1. 71070.
A distinct zonation can be observed in the distribution pattern of other rock
alterations, which are also controlled by lithofacies: the zone of silicification co-
incides in area with the shoaly sands facies: eastward it is replaced by a zone of
predominant dolomitization, and further east of calcitization (Fig. 5A).
Commercial ore deposits are typically represented by the following highly-cu-
priferous association of djurleite (Cu1.geS) + pink bornite (CU5+xFeS4-J, which
is superimposed on earlier paragenetic associations. Djurleite and pink bornite
form lenticular veinlet deposits that crosscut the rocks and the disseminated di-
genite, bornite, and chalcopyrite globules (Fig. 6A). Of wide occurrence are cop-
per sulfide pseudomorphs as replacements of pyrite.
The rich djurleite and djurleite-bornite ores occur in a band extending along
the eastern margin of the Igarka Uplift. This ore zone changes eastward to a lean
ore band of low-grade chalcopyrite and chalcopyrite-galena mineralization (Fig.
5B). A similar change in the mineral zones can be observed vertically upward
from the top of the red-beds. The described distribution of the sulfides is evi-
dence for a decrease in the copper concentration and Bh values of the environ-
ment, occurring upward in the vertical section from the top of the red-beds and
in an east-west direction away from the Igarka Uplift.
The copper mineralization in the Izluchinskaya Formation of the Graviika
mineral estate is restricted to highly permeable rocks - river channel conglomer-
ates, consisting of poorly rounded coarse-grained pebbles of carbonate rock
types (dolomites, limestones), and practically lacking sedimentary in-fillings. In
the zone where the Chernorechenskaya anticline abuts the Igarka Uplift, the cal-
careous conglomerates and some of the wedge-shaped polymictic argillaceous
gritstones are of lighter colours and intense carbonatization. These bleached con-
glomerates, in contrast to their red-colored analogs that replace them toward the
north and east, are characterized by the presence of solid carbonaceous sub-
stances, similar in type to the BOM (Fig. 5A) that is impregnated in the rocks of
the Sukharikhinskaya Formation. In the Izluchinskaya conglomerates the carbo-
naceous matter occurs in the form of large (up to 1 -1.5 cm) oval- and drop-
shaped inclusions in the carbonate cement, or in the form of a black mass densely
pervading the entire rock section. According to data available from X-ray exami-
nations, the carbonaceous matter of the inclusions can be assigned to the medium
anthraxolite type. Its elemental composition, as determined in the VIMS labora-
tory, is as follows: C = 88.96; H = 3.64; (O+N+S) = 7.41 wt%, all on a dry,
ash-free basis. Its carbon isotopic composition (013C = -2.98%) is closely simi-
lar to that of petroleum of the Siberian platform. The specific forms of BOM oc-
currence, its restriction to highly permeable rock types in the Vendian deposits,
and localization near the marginal suture in the zone of conjunction of two up-
lifted blocks, and its crosscutting position in relation to the lithofacies and strati-
graphic boundaries - all these factors taken together give evidence for the mi-
grational nature of the carbonaceous matter. These accumulations of carbona-
ceous substances in the permeable rocks of the Vendian can probably be regarded
as traces of destroyed oil deposits.
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 533

Fig. 6a, b. The detailed features of sulfides on the Graviika ore deposits: a in sandstones of the Suk-
harikhinskaya Formation; b in conglomerates of the Izluchinskaya Formation. Photo by N. Inyash-
kin. Abbreviations: Dj djurleite; Di digenite; B bornite

The lighter-coloured conglomerates, containing anthraxolite, typically reveal

a pronounced increase in their organic-C content (up to 1.5 - 3.5%) and the pre-
dominant occurrence of ferrous oxides as compared to ferric oxides (Fepy'
FeO < 1). They form a body, triangular in plan and lenticular in vertical section,
that in the central part is comprised of a dolomite zone ("dolomite core"). The
conglomerates resemble here dolomitic metasomes with ghost-relicts of primary,
clastic structures and rare corroded quartz grains or silicate rock fragments. Cal-
cite plays an insignificant role, occurring at times as pore-fillings in the dolomitic
conglomerates. The calcite content tends to increase from the dolomite core to-
ward the periphery of the bleached body. According to the data available from
534 I. F. Gablina

chemical analyses of their composition, the conglomerates of the dolomitic zone

consist almost entirely of calcium and magnesium carbonates.
The calcitic zone surrounds the dolomite core, being transitional from the
practically unaltered red-colored rocks to the dolomitized zone. This zone, on
tracing the color boundary, curves around the bands of weakly permeable red-
colored rocks, revealing an intricate flexuous configuration. The major epigenet-
ic mineral in this zone is coarse crystalline calcite, occurring as metasomatic re-
placements of the cement and dolomitic pebbles. The maximal development of
the calcitization process results in the conversion of the calcareous conglomerates
into secondary limestones, containing ghost-relicts of original pebbles. In con-
trast to the metasomes of the dolomitic zone, they characteristically have a lower
content of organic-C (0.09 - 0.270/0, in occasional instances - up to 0.45%) and
of MgO, the latter decreasing in amount at the transition from the dolomitic to
the calcitic zone.
The sulfide mineralization coincides spatially with the calcitic zone and in
part involves the dolomitic zone. The ore bed is of intricate multilayered configu-
ration, matching the configuration of the interfaces of the transition from the
red-colored rocks to secondary grey-colored types. The sulfides are represented
principally by late paragenetic associations of highly-cupriferous sulfides (djur-
leite + pink bornite) and to a lesser degree - by chalcopyrite and galena. The sul-
fides occur as crusts and pore-cement in the conglomerates and frequently form
corrosion borders in pebbles. In some cases three corroded-metasomatic types re-
veal relicts of the primary-globular structure (Fig. 6B). The above-cited charac-
teristics suggest the long duration of the mineralization period, beginning in sedi-
ments with open pore-spaces and continuing during the process of lithification of
the rocks.
The lateral zonation of the metallic minerals is reflected in the change of djur-
leite ores to bornites and then to chalcopyrites, occurring in the direction of the
change in color of the host rocks from red to grey-colored types. In the same di-
rection the zone of copper mineralization is replaced by a lead-bearing mineral
zone. The mineral zones are repeated in the vertical section, revealing an orderly
symmetrical arrangement relative to the red-beds.
Besides the previously described type of mineralization characterized by me-
tallic mineral concentrations of commercial value, and associated with the grey-
colored modifications of coarse-clastic facies of the Izluchinskaya Formation,
metallic mineral occurrences are also associated with the calcareous-argillaceous,
flood-plain, lagoonal facies (Chernorechenskaya ore occurrences). The latter are
located northeast of the Graviika ore locus, within the zone transitional from the
continental red-bed facies of the Izluchinskaya Formation to the grey lagoonal
sediments. The ore mineralization is restricted to the layers of grey-colored dolo-
mites, limestones, and argillites, which are typically of high organic-C content.
The mineralization is of the stratified multilayered type (according to the number
of grey-coloured rock layers), and shows similarity in its major features to the
stratified ore deposits of horizon B.
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 535

General Regularities

The described types of copper mineralization in the Igarka area exhibit character-
istics in their geologic position, mode of occurrence, configuration, and magni-
tude of the ore deposits and also in the time of ore deposition as related to the
time of deposition of the host rocks. Notwithstanding, they are closely related,
demonstrating affinities in their major genetic factors. The latter are as follows:
(1) general association with red-bed formations; (2) localization of the ores in the
area where the red-beds are directly marginal to the grey rock types (rich in or-
ganic carbon) which differ markedly from the former in their geochemical prop-
erties; (3) persistence of typical metallic mineral assemblages, reflecting the gen-
eral similarity in the geochemical environment of ore deposition; (4) zonal distri-
bution pattern of the metallic mineral species, the mineral zones ex~ibiting uni-
formity in their orderly arrangement in relation to the red-beds; (5) the joint oc-
currence in the ores of djurleite 2 together with non-equilibrium mineral associa-
tions [djurleite (or digenite) + pyrite; bornite + pyrite 3J which imply low-temper-
ature ( <93°C) conditions of ore deposition, and the lack of subsequent high-
temperature effects on these mineral assemblages.

Factors Controlling Copper-Ore Deposition

From the analytical treatment of the data outlined above, the inference can be
drawn that the Izluchinskaya red-bed strata was the source of copper during the
process of ore deposition. It is an acknowledged fact (Germanov 1962,
Perel'man 1968, Lur'ye 1978) that confined connate waters of red-bed forma-
tions, due to their specific geochemical properties, are capable of leaching copper
and other metallic minerals and can transport them in the form of dissolved sub-
stances. The metallic minerals precipitate in the form of sulfides upon encounter-
ing biogeochemical barriers, as a result of reactions between the metal-bearing
waters with hydrogen sulfide or pyrite produced by sulfate-reducing bacteria.
Such barriers can be marine sediments, rich in sedimentogenic OM, pyrite, and
other reducing agents (syngenetic barriers), as well as epigenetic accumulations
of OM, hydrocarbons, or H~ (epigenetic barriers) (Lur'ye and Gablina 1982).
Formation of the copper-bearing horizons A and B could have resulted from
the diffusion of copper from the pore waters of the Izluchinskaya Formation into
the organic-rich rocks of the underlying Chernorechenskaya Formation and the
bottom sediments of the Sukharikhinskaya Sea basin. The zonal distribution of
the sulfides can be accounted for by the decrease in the Cu concentrations and in-

2 Djurieite is a typomorphic mineral of copper sandstone and shale deposits and of secondary sulfide
enrichment. In metamorphogenic veins of the Alpine type of deposits, chalcocite (CU2S) is present in-
stead of this mineral. In magmatogenic ore deposits, copper sulfides are represented by chalcocite oc-
curring in combination with high-temperature tetragonal forms of chalcocite (Gablina 1984).
3 At 100 °C chalcocite and pyrite begin to react, resulting in the formation of bornite. At tempera-
tures exceeding 228°C reactions occur between pyrite and bornite that are conducive to the forma-
tion of chalcopyrite (Yund and Kullerud 1966).
536 l.F. Gablina

crease in the H~ concentrations with distance away from the Cu source, the red-
beds. This affords a feasible explanation for the distribution pattern of the cop-
per minerals and their localization in the contact zones of the marine host rocks
and the Izluchinskaya red-bed strata. The specific features of the mineral se-
quences of suifides in horizons A and B reflect the time of formation of the min-
eralization as related to that of the host rocks: syndiagenetic in horizon Band
epigenetic in horizon A.
The formation of the Graviika ore deposits can be assigned to a late-diagenet-
ic evolutionary stage of the Vendian, and seemingly, was a two-phased develop-
ment. The first phase (creation of the epigenetic barrier) can be associated with
the penetration of BOM into the reservoir rocks of the Izluchinskaya Formation,
the BOM having been contributed, presumably, from the underlying oil and gas
sources in the Riphean. Biochemical decay of the BOM probably caused the sec-
ondary alteration observed in the local area of the calcareous conglomerates:
bleaching and dolomitization, concomitant with the leaching of Si0 2• The latter
apparently was redeposited in the overlying sandstones of the Sukhatikhinskaya
Formation, thus forming the zone of silicification.
The second phase (ore deposition) can be assigned to the period when, due to
lithostatic compaction, the confined waters of the red-beds were squeezed out
from the foredeep toward the Igarka Uplift, where the conditions favored their
discharge. The inflow of these calcium sulfate waters of the red-bed formations
was a favorable factor promoting the biochemical oxidation of OM that accumu-
lated in the Izluchinskaya conglomerates. The decomposition products of bio-
chemical OM decay consisted of H~ and CO 2• Upon encountering the hydrogen
sulfide or pyrite geochemical barrier, the nonferrous metals (copper and lead),
which were brought in by the waters of the red-beds, precipitated. The direction
of flow of the metal-bearing solutions is reflected in the zonal distribution pat-
tern of copper sulfides: in the direction away from the red-beds toward the rocks
of secondarily-derived lighter colors, the mineral sequences reveal the replace-
ment in successive order of djurleite (CU1.96'5) by bornite (CusFeS4) and further
by chalcopyrite (CuFeS2), which reflect the decrease in the copper concentration
and concurrent increase in the sulfide-ion concentration in the same direction.
The spatial separation of copper and lead minerals (shifting of the galena zone
toward the barrier) is due to the difference in the geochemical mobility of these
elements and also serves to indicate their migration from the red-beds.
It is known that the concentration of sulfate ions in the solution enhance the
solubility of magnesium carbonates (reserve reaction of the Heidenberg reaction).
Possibly the calcium-sulfate waters of the red-bed formations produced the de-
dolomitization and calcitization observed in the zone of sulfide formation.
Near the steep eastern flank of the Igarka Uplift, where the Izluchinskaya
Formation consists of highly permeable rocks and the overlying marine sedi-
ments are of reduced thickness, the formation waters under pressure could have
flowed vertically upwards. This unward migration could have been controlled by
the permeable rocks of the coastal shallow facies, as well as the fractured zones
of the marginal suture (Fig. 7).
The migration of hydrocarbons from the Izluchinskaya conglomerates into
the overlying Sukharikhinskaya Formation was probably due to the upward flow
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 537

Groviiko ore-deposit

~o.
I , t.

r:ooCJ 2 E//ld 5

~ 3 6

Fig. 7. The genetic model and types of copper mineralization in the Igarka area. J Red-bed forma-
tion; 2 bleached conglomerates of the red-bed formation; 3grey marine and lagoonal deposits; 4oc-
currences of epigenetic OM; 5 lean, stratified mineralization; 6 rich sulfide ores

of artesian waters. The rocks of the Sukharikhinskaya Formation, enriched in

sedimentogenic and migratory forms of OM, served as a specific type of hydro-
gen sulfide "filter", upon which the metals, being carried upward by the up flow-
ing waters, precipitated in the form of sulfide. With increased distance from the
drainage zone, the vertical flow of the fluids changed to lateral migration within
the permeable beds. Manifest~tions of this lateral migration are reflected in the
zonal distribution pattern of the ore minerals (see Fig. 5B).

Conclusion

Several types of copper mineralization associated with the Vendian red-bed for-
mation can be recognized in the Igarka area (Fig. 7, Table 1). According to the
magnitude of the deposits they can be subdivided into two groups: (1) lean,
stratified deposits, occurring at the top and base of the Izluchinskaya red-beds,
and also in the enclosed grey layers of the flood plain and lagoonal facies that are
transitional to shallow, marine facies; (2) rich accumulations of sulfide ores, spa-
tially and genetically related to the permeable coarse-clastic rocks of the red-bed
facies. The latter are represented by the Mansfeld and Dzhezkazgan types, the
combination of both types being responsible for the Graviika ore deposits.
Table 1. Types of copper mineralization in the Igarka area
U>

Type of miner- Subtype Position of min- Host rocks Mode of ore de- Type of bio- Distribution of Magnitude of
....
co
alization eral zone with position accord- geochemical mineralization ore deposits
respect to that ing to the time barriers (or OM)
of the red-bed of formation of
formation the host rocks

Mansfeld Syngenetic At the base of Basal members Syndiagenetic Syngenetic (sedi- Areal distribu- Noncommercial
the grey marine of the Sukharih- mentogenic OM) tion
series, overlying kinskaya Forma-
Epigenetic terrestrial red- tion (horizon B) Epigenetic Mixed: sedimen- Local restriction Commercial
bed formations togenic + epige- (above perme-
netic OM able rocks of
coarse red-bed
facies)

Donbass In the zone of Grey rock lay- Syndiagenetic Syngenetic (sedi- Areal distribu- Noncommercial
lateral transition ers, enclosed in mentogenic OM) tion
from terristrial variegated la-
red-beds to la- goonal sediments
goonal-marine of the Izluchin-
facies skaya Formation

Dzheskazgan Enclosed in ter- Permeable Epigenetic Epigenetic (epi- Local restriction Commercial
restrial red-bed coarse rock of genetic OM) (to coarse-clastic
formations secondary grey facies)
colors in the Iz-
luchinskaya For-
mation

Dzhilandi At the top of the Rich in organic Epigenetic Syngenetic (sedi- Areal distribu- Noncommercial
grey marine sedi- matter, grey mentogenic OM) tion
ments, underly- marine sedi-
....
ing terrestrial ments, transi- 71
0
red-bed forma- tional from the Pl
r::r
tions Chernore- §'
chenskaya to the Pl

Izluchinskaya
Formation (hori-
zon A)
Genetic Types of Copper Mineralization in the Igarka Area, West of the Siberian Platform 539

References

Gablina IF (1984) Mineral species of the chalcocite series in copper-sandstone and shale deposits. Mi-
nerali khalkozinovovo ryada iz mestorozhdenii medistikh peschanikov i slantsev. Zapiski Vsesoyuz
Min ob-va 4:430-443
Germanov AI (1962) Hydrodynamical and hydrochemical factors controlling the formation of cer-
tain hydrothermal ore deposits. Gidrodinamicheskie i gidrokhimicheskie usloviya obrazovania ne-
kotorikh gidrotermal'nykh mestorozhdenii. Izv An SSSR Ser Geol 7:79 - 98
Lur'ye AM (1978) Factors controlling copper migration in red-bed formations. Uslovia migratsii me-
di v krasnotsevnykh formatsiyakh. Geokhimia 6:926 - 932
Lur'ye AM, Gablina IF (1982) Controlling factors of copper concentration during the exogenetic pro-
cess. In: Factors controlling the formation of rare and nonferrous metallic mineral deposits. Uslo-
via kontsetratsii medi v ekzogennom protsesse, pp 133 - 142. V kn: U slovia obrazovania mesto-
rozhdenii redkikh i tsevetnykh metallov. Nauka, Moscow, p 256
Perel'man AI (1968) Geochemistry of epigenetic processes. Geokhimia epigeneticheskikh protsessov.
Nauka, Moscow, p 331
Rzhevskii VF, Miroshnikov AE, Dushatkin AB, Shklyarik GK (1980) Copper mineralization in the
upper precambrian of the Igarka Area. In: Processes governing the accumulation of sedimentary
and volcanogenic-sedimentary deposits of nonferrous metals (Siberia and the Far East). Medenos-
nost' verkhnedokembriiskikh otlozhenii Igarskogo raiona, pp 81 - 84. V kn: Protsessi osadochno-
va i vulkanogenno-osadochnovo nakoplenia tsvetnikh metallov (Sibir' i Dal'nii Vostok). Nauka,
Novosibirsk, p 159
Yund R, Kullerud G (1966) Thermal stability of assemblages in the Cu-Fe system. J Petrol 7,
3:454-488
The Diverse Styles of Sediment-Hosted Copper Deposits
in Australia
I. B. LAMBERT, 1. KNUTSON, T. H. DONNELLY, and H. ETMINANI

Abstract

The huge sediment-hosted copper deposits at Olympic Dam and Mount Isa are of
unusual types. They formed epigenetically from hydrothermal fluids, as did sev-
eral other important Australian copper deposits in sedimentary strata. Ore within
a palaeoweathering zone at the top of a red-bed unit at Mount Gunson appears to
have formed at relatively low temperatures over a long period and to have incor-
porated biogenic sulfur. Bedded-disseminated mineralization, which is wide-
spread in South Australia, formed by reaction of cupriferous fluids with biogenic
HzS and early diagenetic iron sulfides. To date, stratiform copper mineralization
has not been of major economic significance in Australia.

Introduction

Ores in sedimentary strata have provided roughly two-thirds of Australia's cop-

per production and they contain the great bulk of the known reserves of this met-
al. The sizes, grades and ages of the most important of these copper deposits are
given in Table 1. The regions in which they occur, shown in Fig. 1, are the Ade-
laide Geosyncline-Stuart Shelf (South Australia), the Mount Isa Inlier (Queens-
land), the Tennant Creek Inlier (Northern Territory), the Bangemall Basin
(Western Australia) and the Cobar Trough (New South Wales).
This paper succinctly reviews the geology of these regions and discusses the
general features and genesis of the main deposits. The emphasis is on the results
of recent research in the two most important copper provinces, the Adelaide
Geosyncline-Stuart Shelf and the Mount Isa Inlier.

I Baas Becking Geobiological Laboratory, GPO Box 378, Canberra ACT 2601, Australia

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Table 1. Major sediment-hosted copper deposits in Australia

Region Sizea,b Status Age of Form of mineralization ;l

(1)

Deposit host strata t:I

:;:.
!.!l
Stuart Shelf >2000 m.t. @ Preparations Middle Hydrothermal mineralization in un- (1)
'"
[JJ
Olympic Dam 1.6070 Cu, 0.064070 U 3 0 8 for mining Proterozoic metamorphosed graben sequence q,
(Roxby Downs) 0.6 g t- I Au, (REE, F) dominated by sedimentary breccia Dl
units C o
.....
[JJ
Stuart Shelf 5.9 m.t. @ 1.9070 Cu, Tabular deposit in breccia along 8..
Cattle Grid (Pb, Zn) Ore nearly Middle unconformity surface at top of S·
(1)
(Mount Gunson) exhausted Proterozoic unmetamorphosed red-bed unit d ::>

Adelaide Geosyncline Lenses of hydrothermal mineraliza- :£g

Kanmantoo 12 m.t. @1070 Cu Closed 1976 Cambrian tion flattened parallel to axial plane (1)

schistosity in metasiltstone e 0.
(')
o
Mount Isa Inlier Metamorphogenic hydrothermal '0
'0
(1)
Mount Isa >232 m.t. @ 2.9070 Cu Largest mine Middle mineralization in altered, brecciated ..,
Proterozoic dolomitic siltstone f t:I
.g
Tennant Creek Inlier
Peko 3.1 m.t. @ 4070 Cu, (Bi) Closed 1976 Pipelike hydrothermal lodes with
g.
abundant quartz, magnetite, hematite, S·
Warrego 7.8 m.t. @ 2.6070 Cu, Mine Early
2gt- 1 Au, (Bi) Proterozoic chlorite, in major shear zones 8 ~
~
Gecko 5.2 m.t. @ 3.7070 Cu Care and
0.7 g t- I Au, (Bi) maintenance ~
Cobar Trough
CSA 8.6 m.t. @ 1.7070 Cu, Mine Siluro- Hydrothermal stockwork to massive
14gt- 1 Ag (Pb, Zn) Closed 1945 Devonian mineralization in silicified shear zones
Great Cobar 4.2 m.t. @ 2.8070 Cu, sub-parallel to slaty cleavage h
2.2gt- 1 Au, 11.2gc l Ag
Bangemall Basin
Nifty 1.3 km strike length Major Mid-late Stratabound in sandstone-siltstone-
secondary ore prospect Proterozoic shale sequence i
+ primary Cu
VI
t:
a Total size, i.e. ore mined plus proven and possible reverses as published in company reports and press releases. b m. t. = million tonnes; gel = grams/tonne.
CRoberts and Hudson (1983). dKnutson et al. 1983, Creelman 1983. eVerwoerd and Cleghorn 1975. fPerkins 1984. gCrohn 1975, Goulevitch 1975, Wright
1965. hBrooke 1975. iWestern Mining Corporation, press release 1983; Goode and Hall 1981
542 I. B. Lam bert et al.

200 400 600 800 1000 k'.mOlru

I

OROGENIC PROVINCES PLATFORM COVERS

E::3 EAST AUSTRAliAN r;;;--, TRANS · AUSTRAliAN

~ {Pal.eozolC;: deformelJon ) ~ t Permlln-Holoc:ene )

1::.:::. .1LATE PRECAMSRIAN

••••••• :. IO,form. tlon < 1000 m. y. - 650 m·v. )
r;;cl
L.:....J
CENTRAL AUSTRALIAN
11 400 m ..,.. TriaSSIC)
r:;.::::::] CENTRAL AUSTRAliAN rp;j1 NORTH AUSTRALIAN
b:.:::.J 10.lo,",,,,on 1900-1000 m·y·1 ~ 1'800 m.y. - e • . 1300 m.y.,

•
NORTH AUSTRAliAN r-p;I WEST AUSTRALIAN
I Deform.lion ca . 2000- 1900 m.v.) L.:..:...J 12200-2000 m·y·1 11.w.

IIIIII11 WEST AUSTRAliAN

( O.fOlmatlon> 3000 m .y. - CI . 2200 ",. y.,

~ UNASSIGNED PRECAMSRIA.."
~ METAMORPH IC COMPLEXES

Fig. 1. Map summarizing major geological features of Australia (after Warren 1972) and localities of
the main sediment-hosted copper provinces
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 543

Adelaide Geosyncline-Stuart Shelf

Regional Geology

The most recent review of the Precambrian geology of this region is by Rutland
et al. (1981). Figures 2 and 3 summarize the general geology and stratigraphy.
The schists, gneisses and granites of the crystalline basement were metamor-
phosed and deformed approximately 1.8 to 1.65 b.y. B.P. In the west of the re-
gion, post-orogenic, middle Proterozoic volcanic and sedimentary strata accu-
mulated in grabens. The youngest of the pre-late Proterozoic units is the Pandur-
ra Formation, a thick continental red-bed sandstone.
Approximately 1.1 b.y. B.P., an aulacogen developed in the eastern part of
the region, the site of accumulation of the thick, late Proterozoic (Adelaidean)
clastic-carbonate sequence of the Adelaide Geosyncline. This was separated by

D
IJ] lilt ProtltOzaic-Cambrian strati
, ,.
!ill]
........
-,.:.- Fig. 2. General geological features and
100 1m localities of main copper deposits, Ade-
laide Geosyncline-Stuart Shelf region,
South Australia
544 I. B. Lambert et al.

ERN STUART SHELF

ADELAIDE GEOSYNCLINE SOUTHERN STUART SHELF
OUNT GUNSON)

~
••••
•• ~
v ... • •••••••
••••••
'"
v

rtMU.N Ofl05un
I

Fig. 3. Generalized stratigraphy, Adelaide Geosyncline·Stuart Shelf region, and stratigraphic posi·
tions of main copper deposits. 1Wallaroo-Moonta; 2 Mount Painter; 3 Olympic Dam; 4, 5, Mount8
Gunson; 6Copper Claim, Blinman; 7Burra Burra; 9Myall Creek; 10 Kapunda; 11 Willouran Ranges

the Torrens Hinge Zone from the stable platform environment of the Stuart
Shelf, where there is a relatively thin Adelaidean sequence. The late Proterozoic
strata are characterized by a predominance of shallow water and emergent facies,
subordinate mafic volcanism and two major tillite units. There is evidence of
non-marine environments in some units.
A major erosional break preceded transgression and widespread deposition
of post-rift Cambrian strata across the region. The Adelaide Geosyncline se-
quence was regionally metamorphosed and deformed during the late Cambrian-
Ordovician, but that on the Stuart Shelf was not affected.

Mineralization

The locations of the main copper deposits are shown in Fig. 2; their stratigraphic
positions in Fig. 3.
The features of Olympic Dam (Roberts and Hudson 1983), Moonta-Wallaroo
(Johnson 1965) and Kanmantoo (Verwoerd and Cleghorn 1975) imply that they
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 545

formed from hydrothermal fluids. Because of space limitations, only the largest
of these, Olympic Dam, will be considered here.
The other two styles of copper mineralization to be discussed below, neither
of which has obvious hydrothermal alteration zones, occur in the Mount Gun-
son-Pernatty Lagoon area (Fig. 2). The most important of these occurs in a brec-
cia along the unconformity between middle Proterozoic red-bed and late Pro-
terozoic pale sandstone. The second occurs as largely bedded sulfides in late Pro-
terozoic dolosiltstone. Bedded deposits of this general style have been found at
widespread localities in the late Proterozoic strata of this province; while none
have continuity of high copper grades in primary mineralization, numerous, sec-
ondarily-enriched, near-surface ores have been mined in small-scale operations.

Olympic Dam

For a detailed description of this deposit, the interested reader is referred to the
paper by Roberts and Hudson (1983). The post-orogenic middle Proterozoic gra-
ben sequence which hosts the Olympic Dam mineralization is over 1 km thick. It
comprises mainly granitic and polymict breccia units overlain by iron formation
and tuffaceous felsic volcanics (Fig. 4). The sequence is pervasively altered, and
there is relatively intense hematite-chlorite and sericite-silica alteration associated
with the mineralization.
The copper mineralization can be divided into stratabound and transgressive
types (Figs. 4, 5) both of which contain significant uranium, rare earths, gold and
silver. The ore is quite variable in appearance (Fig. 5). Sulfides are mainly dis-
seminated through breccia matrices, but large clasts of sulfides are also common,
and there is some massive mineralization. Textures indicate a complex paragene-
sis, involving micro cavity and fracture filling, replacement and rimming of
clasts. Reconnaissance fluid inclusion and sulfur isotopic analyses have been un-
dertaken on Olympic Dam mineralization (D. Roberts, personal communication,
1984). Homogenization temperatures for inclusions in fluorite and quartz associ-
ated with sulfides range from 110° to 265°C, and freezing point depressions indi-
cate 0.5 to 14.3070 NaCI equivalent. These indicate mixing of fluids of different
salinities and temperatures. The 03~ values for ore sulfides are concentrated in
the range - 5%0 to -11 %0, and barite 0 3~ values range from + 4.6%0 to + 13%0.
In view of the evidence for hydrothermal alteration under relatively oxidizing
conditions, these preliminary isotopic data are consistent with a magmatic sulfur
component. The high contents of potassium, rare-earths, barium, fluorine and
iron in the mineralization are consistent with hydrothermal activity associated
with alkaline igneous activity.

Mount Gunson-Pandurra Formation

In the Mount Gunson-Pernatty Lagoon area of the Stuart Shelf, upfaulted

blocks of Pandurra Formation red-bed sandstone formed a depositional high
(Pernatty culmination; Figs. 3 and 6) during the Adelaidean. The relatively im-
546 I. B. Lambert et al.

wEST EAST

,
0 «..
---
-==-
?
I
?
? ?

GREENFIELD FORMATION OLYMPIC DAM FORMATION

MEMBERS MEMBERS

0 COVER EJ Volcanic 0 Upp.r G,anlt. B,.ccla

0 DOLERITE G:;] Hemallle Breccl. EJ Brook.

0 Lower Silicified ~ Whenan

Unconfo,mlty EJ Black Hemallle

a Fault CJ Lower Granite Brecc i a
.

··
·:.
L an • • • In ... . , 1..
Ric'" IIncel ••

;.

Fig. 4. a East-west cross-section, Olympic Dam area (from Roberts and Hudson 1983). b Represen-
tation of the two types of mineralization and mineral zoning, Olympic Dam deposit (from Roberts
and Hudson 1983)

permeable Tapley Hill Formation is locally absent over this structure, the Pan-
durra Formation being overlain by Whyalla Sandstone. The only economically
important mineralization known in the Pandurra Formation is in the brecciated
palaeoweathering surface at the top of this culmination (Fig. 7). The brecciation
appears to have formed partly as a result of periglacial weathering and it is local-
ly intense around faults (Tonkin and Williams 1983). The main deposit is Cattle
Grid, which has been described by Gersteling and Heape (1975), Creelman (1983)
and Knutson et al. (1983).
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 547

Fig. Sa, b. Cut sections of Olympic Dam drill core. a Stratabound mineralization, polymict breccia
clasts in chalcopyrite-rich (white) marix. b Transgressive chalcocite-bornite mineralization (Photo-
graphs kindly supplied by D. E. Roberts)
548 I. B. Lambert et al.

sw Pernatty NE Fig. 6. Schematic section, Mount Gun-

Cattle Grid Lagoon son-Pernatty Lagoon area

7j<. :-...~.: Culmination . ~- : ';,: :~

/ . '. -
CJ Tent Hill Fm.
I. -_
,
.-
./
- ,
-:. __ ; :~

D Whyalia
. Sandstone
o Pandurra Fm.
D Ta pley Hill Fm. ~ ~~~~:":Iisation
Vertica l exaggeration ce 7 to ,
181"'32

Fig. 7. Chalcocite (black) and chalcopyrite (light grey) ore in brecciated Pandurra Formation, Cattle
Grid mine, Mount Gunson

Cattle Grid is a tabular deposit, averaging roughly 4 m thick, in which the

grade of copper falls off rapidly with decreasing brecciation. The mineralization
mainly infills fractures and vugs, and it is dominated by chalcocite, chalcopyrite,
bornite and pyrite; marcasite, sphalerite and galena are minor components.
The 15 3"s values for the ore minerals are concentrated between -12%0 and
- 5%0 (Fig. 8). In the absence of evidence for hydrothermal mineralizing fluids,
these are most readily interpreted in terms of bacterial sulfate reduction; the
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 549

PANDURRA TAPLEY HILL MINERALISATION Fig. 8. Distributions of

sulfur isotopic compositions
M INERALISATION (MT . GUNSON. MYALL CREEK. of mineralization in
(MT GUNSON) KAPUNDA etc.) Pandurra and Tapley Hill
Formations

-30
,~
-20 -10 0 10 20 30 40 50

£,34 S (COT) %0
16/ A133

paucity of organic matter in the mineralized breccia, and the form and timing of
the sulfide mineralization, imply that the H~ was introduced in groundwaters.
The sulfide mineralization must have formed largely after burial of the pa-
laeoweathering zone. It was the product of protracted movement of fluids
through the permeable rocks of the culmination, resulting in a complex sequence
of deposition, dissolution, redeposition and replacement. Metal-enriched fluids
were probably generated as brines reacted with minerals in the Pandurra red-beds
and in the older volcanic and sedimentary rocks. In addition, weathering of Beda
Volcanics (Fig. 3) could have resulted in metal enrichments in waters in the un-
derlying red-beds.

Mount Gunson- Tapley Hill Formation

The Tapley Hill Formation is the only mineralized late Proterozoic unit on the
Stuart Shelf (Fig. 3). Copper, lead and zinc sulfides were widespread in minor
amounts in the basal 5 m or so of this unit, but no economic deposits have been
defined. The two main prospect areas have been at Mount Gunson and Myall
Creek (Fig. 2). The following summary of the Mount Gunson mineralization
draws information from the paper by Knutson et al. (1983).
The Tapley Hill Formation in the Mount Gunson area is up to 200 m thick
and is dominantly dolostone adjacent to the Pernatty culmination, where it con-
tains stromatolites, desiccation cracks, intraclast and edgewise conglomerates,
and possible pseudomorphs after evaporites. It grades outwards into increasingly
laminated strata with abundant carbonaceous dolo siltstone components, which
were deposited beneath wave base. The mineralization comprises fine-grained
disseminations of chalcopyrite, bornite, chalcocite-pyrite-sphalerite and galena.
The copper minerals partly replace and rim pyrite (Fig. 9). In addition, there are
discontinuous thin (millimetre-scale) sulfide veinlets and blebs which mainly fill
dewatering channels and desiccation cracks, and probably formed during diagen-
esis. It is noteworthy that the mineralization is concentrated in the relatively per-
meable dolo siltstone laminae.
There is a general vertical zoning from copper at the base of the mineralization
inwards to lead and zinc, and a similar zoning occurs laterally away from the pa-
laeohigh. Relatively minor metal anomalies occur in places at the top of the Tapley
Hill Formation, with decreasing Cu/Pb + Zn ratios downwards into the unit.
550 I. B. Lambert et at.

Fig. 9. Low-grade copper mineralization in unmetamorphosed Tapley Hill Formation, Stuart-Shelf.

Framboidal pyrite (white) rimmed and partially replaced by bornite (mid-grey) . Reflected light; ap-
prox. 200 ~ across photomicrograph

Pyrite and ore sulfides in the Tapley Hill Formation at Mount Gunson have
t5 34s values in a wide range from 0%0 to +45%0, indicating a different sulfide
source from the mineralization in the Pandurra Formation (Fig. 8). Together
with the absence of hydrothermal alteration, and the presence of carbonaceous
matter, these isotopic compositions imply biological sulfate reduction in a re-
stricted environment. Such 34s-enriched sulfides, and the high 87Sr/8tSr ratios of
dolomite from the Tapley Hill Formation (Lambert et al. 1984) are most readily
interpreted in terms of predominantly continental waters in the Tapley Hill sea.
Dolomite from both mineralized and unmineralized Tapley Hill Formation
has 15 Be values close to - 3%0 and 15 180 values close to + 26%0, indicating that
the mineralizing fluids were isotopically similar to, or dominated by, the diage-
netic fluids. However, alteration of detrital feldspar to clays is more advanced in
the mineralized strata, implying that relatively high volumes of near-neutral
fluids passed through these strata. This, and the concentration of copper in the
dolosiltstone laminae, support protracted copper addition to the strata. The late
copper appears to have been precipitated largely by reaction with iron sulfides.
The proximity and similar compositions of the deposits in the Tapley Hill and
Pandurra Formations at Mount Gunson suggest their generation from the same,
or related, metalliferous fluids.
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 551

Mount Isa Inlier

Regional Geology

The oldest rock exposed in this region comprise complexly deformed igneous and
sedimentary rocks which have metamorphic ages around 1.87 b.y. B.P. (Page
1978, Blake et al. 1983). According to Plumb et al. (1980), these basement rocks
are most abundant in a tectonic high separating the so-called eastern and western
successions of unconformably overlying middle Proterozoic sedimentary and
volcanic strata (Fig. 10), but recent studies imply that "basement" rocks are scat-
tered throughout the region (Blake et al. 1984). The largest copper deposits in the
region, Mount Isa and Mammoth, are in the western succession (Fig. 10).
The basal arenites of the western succession are overlain by up to 5 km of al-
tered metabasalts and intercalated arenites of the Eastern Creek Volcanics. Ase-
quence of quartz-rich arenites and conglomerates occurs unconformably above,

\40·

D Younger cover

• Western succession

• Eastern succession

t·: .'1 Eastern Creek Volcan ics

1. . "''''1 Basement tectonic high Fig. 10. The Mount Isa Block, showing
- - Major fault \ 00 'm - gross geological features and localities of
16/ Al30 major copper deposits
552 I. B. Lambert et al.

and this contains local felsic volcanics at the top. This sequence is unconform-
ably and disconformably overlain by Mount Isa Group strata, comprising gener-
ally shallow water to emergent, dolomitic and variably carbonaceous shales and
siltstones (Neudert and Russell 1981), with adularia- and albite-rich tuffaceous
beds. Tuffites from Mount Isa mine have been dated at 1.67 b.y. B.P. (Page
1981). Radiometric and structural studies imply periods of greenschist facies
metamorphism at around 1.61 and 1.55 b.y. B.P., and a final deformation event
at around 1.48 b.y. B.P. (Page 1978, Blake et al. 1983). Granites were emplaced
before 1.8 b.y. and approximately 1.78, 1.67 and 1.5 b.y. B.P. (Blake et al.
1983).

Copper Mineralization

The great bulk of the copper deposits in the Mount Isa Inlier are small veins in
faults and shears (Brooks et al. 1975). Roughly half of them are in metasedi-
ments, and may have a spatial association with mafic igneous rocks. Bedded de-
posits are rare.
Mount Isa is more than two orders of magnitude larger than Mammoth (ap-
proximately 1.5 million tonnes at 311,10 Cu), which is the second biggest ore body
known in the region. Both of these deposits occur in brecciated, altered, pyritic
strata adjacent to faults and altered mafic volcanics. Only the Mount Isa deposit
is reviewed here; the features of Mammoth are available from the publications by
Mitchell and Moore (1975) and Scott and Taylor (1982).

Mount Isa

Detailed information on this deposit can be obtained from the papers by Mathias
and Clark (1975) and Perkins (1984).
There are closely adjacent but separate copper and lead-zinc ore bodies at
Mount Isa (Fig. 11). The latter consist of fine-grained, laminar sulfides of synge-
netic-diagenetic origin in the carbonaceous, dolomitic and tuffaceous Urquhart
Shale of the Mount Isa Group.
In contrast, the copper ore comprises coarser-grained massive, disseminated
and veinlet mineralization, mainly chalcopyrite, pyrite and pyrrhotite (Fig. 12).
It occurs within so-called silica-dolomite (Fig. 11), a facies of the Urquhart Shale
which extends upwards from the faulted contact with the underlying greenstones
(probably Eastern Creek Volcanics). The silica-dolomite has an outer zone of re-
crystallized shale with pods of massive, recrystallized dolomite. Inside this is a
zone of irregularly brecciated and recrystallized dolomitic shale, and a central
zone of brecciated siliceous shale. The massive ore is developed within the central
zone, against the basal greenstone contact.
The timing of the copper mineralization has been a source of considerable
controversy. Several workers, most recently Finlow-Bates and Stumpfl (1979),
have argued that the copper and lead-zinc ore bodies formed from the same min-
eralizing fluids, with the copper precipitating largely in near-surface feeder chan-
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 553

Surface Fig. 11. Mount Isa mine (after Mathias

and Clark 1975). Copper ore occurs
throughout the "silica-dolomite", shown
as shaded zones above the basic volcanics

ISS] Recrystallised shale

~ Crystallrne dolomite

[J Irregularly breccIated dolomItIc shale

!ill BrecCIated siliceous shale

o Carbonaceous mylonIte

L3 Bas,c volcanics

[l] Pb· Zn. Ag orebodles rn pyritic shale

and stltstone 16 ~ 1<

nels. However, recent studies have produced evidence against co-genetic copper
and lead-zinc mineralization. In summary, structural, textural and sulfur- and
lead-isotopic data imply that chalcopyrite, and associated late-generation pyrite
and pyrrhotite, postdate the early phases of regional metamorphism (Gulson et
al. 1983, Perkins 1984).
The lead isotopic compositions of the various copper Ore bodies at Mount Isa
fall on several distinct trend lines (Gulson et al. 1983). These authors argued that
the trend lines reflect the varying importance of the Eastern Creek Volcanics and
basement rocks as metal SOurces, the latter providing a relatively high proportion
of the metals in the largest Ore bodies.
While details of ore genesis remain unresolved, the heated brines which
formed the copper Ore bodies appear to have been generated during metamor-
phism of the sequence (Perkins 1984). Sulfur-isotopic data imply that major pro-
portions of sulfide were introduced with the copper (Gulson et al. 1983). This hy-
drothermal sulfide could have come from leaching of sulfide minerals from the
underlying rocks, and/or thermochemical reduction of sulfate. The hydrother-
554 I. B. Lambert et al.

o lel1l

Fig. 12 A, B. Mount Isa copper ore. A High-grade chalcopyrite ore in silicified shale. BChalcopyrite-
pyrrhotite veinlet mineralization in brecciated dolomitic shale
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 555

mal mineralizing fluids, which probably ascended via faults and shear zones,
caused metasomatism and hydraulic fracturing of the Urquhart Shale.

Other Sediment-Hosted Copper Provinces

Tennant Creek Inlier

In this region there is a low-grade metamorphosed, early Proterozoic greywacke-

shale sequence with intercalated felsic tuffs and volcanics (Crohn 1975). Graded
bedding and slumping are common in the sedimentary strata, which have been
interpreted as accumulating in moderately deep water. The sequence has been in-
truded by massive and foliated granites, quartzo-feldspathic porphyries, diorites,
dolerites and lamprophyres. Dating by Black (1983) indicates that the felsic vol-
canics, granites and porphyries formed about 1.87 -1.85 b.y. B.P.
The main copper ores of the region, Peko and Warrego, have been described
by Wright (1965) and Goulevitch (1975), respectively. These comprise chalcopy-
rite and pyrite, with minor bismuthinite and gold, as irregular dispersions in
magnetite and quartz-magnetite ironstones. The ironstones are pipelike bodies in
fold axes and shear zones cutting the greywacks-shale sequence. They are struc-
turally-controlled hydrothermal deposits, which probably formed during the
main and waning stages of deformation, around 1.81 b.y. B.P. (Black 1983).

Bangemall Basin

The discovery of several copper prospects in the Bangemall Basin, Western Aus-
tralia (Fig. 1), was announced recently (Western Mining Corporation press re-
lease, September 1983). In. this region, a thick, dominantly shallow-water se-
quence of sandstones and interbedded siltstones, shales and carbonates accumu-
lated about 1.2-1.1 b.y. ago (Goode and Hall 1981). These authors consider
that there is a volcanic ash component in the sequence. The mineralization is in a
relatively highly deformed and weakly metamorphosed (to low greenschist facies)
belt at the northeastern margin of the basin.
The 1983 press release stated that the largest of the prospects being assessed in
this area, Nifty, contains secondary copper mineralization amenable to open-cut
mining over a strike length of 1.3 km, and incompletely assessed primary (hydro-
thermal?) mineralization with grades of up to 15.80/0 copper. No detailed infor-
mation is available on this deposit.

Cobar Trough

The geology of the Cobar region (Fig. 1) has been reviewed by Brooke (1975) and
Pogson et al. (1976). In addition, synopses of recent information from a variety
of studies in the region appear in the abstract volume of the 6th Australian Geo-
logical Convention (Geological Society of Australia, Canberra, 1983).
556 I. B. Lambert et al.

The oldest rocks known in the region are a sequence of Cambro-Ordovician

turbidites, cherts and minor basic volcanics, which were metamorphosed and de-
formed under greenschist facies conditions in the early Silurian, and intruded by
early to middle Silurian granites. The Cobar Trough formed in the latest Silurian
and Its subsidence appears to have been controlled by pre-existing faults. Rela-
tively high relief generated at the eastern margin of the trough resulted in deposi-
tion of an early fringing sequence of conglomerates and sandstones. Thin-bedded
turbidites were deposited under deeper water conditions to the west of, and over-
lying the coarse clastics, and were followed by shaley and silty beds. Finally, a
sandy facies was deposited across the shallowing trough. Felsic volcanism oc-
curred on the eastern shelf during the filling of the Cobar Trough. However,
there are no known volcanic centres within several tens of kilometres of main
mineral deposits, and the proportions of tuffaceous materials in the host strata
are not generally considered to be high.
The copper-rich deposits of this field are in steeply dipping, silicified shear
zones, sub-parallel to the regional cleavage. They comprise massive to stock-
work, variably deformed mineralization, and occur in several stratigraphic units,
mainly in fine-grained strata.
The sulfur- and lead-isotopic compositions of the Cobar ores fall in restricted
ranges which are very similar to those for Silurian to early Devonian volcanogen-
ic deposits in southeastern New South Wales (Sun 1983). While details of ore
genesis are still unresolved, there is a consensus of opinion that the copper miner-
alization precipitated epigenetically from hydrothermal fluids generated by high
thermal gradients in the trough, prior to or during regional metamorphism.

Summary and Conclusions

The most important Australian copper mine, Mount Isa, comprises stratabound,
metamorphogenic mineralization in brecciated and metasomatized middle Pro-
terozoic metasiltstone. The huge copper (gold-uranium-rare earths) deposit at
Olympic Dam, which is not yet in production, formed from hydrothermal fluids
permeating sedimentary breccias in tectonically active graben environments. Im-
portant structure-controlled hydrothermal deposits occur in metamorphosed
trough sequences in the Tennant Creek (early Proterozoic), Kanmantoo (Cam-
brian) and Cobar (Siluro-Devonian) regions.
Relatively low-temperature epigenetic deposits have been mined from the pa-
laeoweathering zone on a middle Proterozoic sandstone unit at Mount Gunson.
Bedded-disseminated deposits are widespread in South Australia, and secondari-
ly-enriched mineralization has been min,ed from these in numerous small-scale
operations. The primary bedded sulfides formed as a result of long-term influx
of low-temperature cupriferous fluids which reacted with biogenic sulfur and
iron sulfides.

Acknowledgements. The Baas Becking Laboratory is sponsored by the Bureau of Mineral Resources,
the Commonwealth Scientific and Industrial Research Organization and the Australian Minerals In-
dustry Research Association Limited.
The Diverse Styles of Sediment-Hosted Copper Deposits in Australia 557

References
Black LP (1983) Tennant Creek geochronology. BMR 83. Yearb Bur Mineral Res, Aust, pp 109-110
Blake DH, Page RW, Wyborn LAI, Etheridge MA (1983) Mount Isa base metal province. BMR 83.
Yearb Bur Mineral Res, Aust, p 50
Blake DH, Bultitude RJ, Donchack PJT, Wyborn LAI, Hone IG (1984) The geology of the Duchess
Urandangi Region, Mount Isa Inlier, Queensland. Bur Mineral Res Bull 219
Brooke WJL (1975) Cobar mining field. In: Knight CL (ed) Economic geology of Australia and Pa-
pua New Guinea, vol I: Metals. Australas Inst Min Metal, Melbourne, pp 683 - 694
Brooks JH, Wilson IH, Saurers JD (1975) Minor copper deposits in the Mount Isa/Cloncurry Dis-
trict. In: Knight CL (ed) Economic geology of Australia and Papua New Guinea, vol I: Metals.
Australas Inst Min Metal, Melbourne, pp 402 - 405
Creelman RA (1983) A petrological mineralogical and lead isotope study of copper in sandstones and
related sediments. PhD Thesis, Macquarie Univ (unpublished)
Crohn PW (1975) Tennant Creek-Davenport Proterozoic Basins, regional geology and mineraliza-
tion. In: Knight CL (ed) Economic geology of Australia and Papua New Guinea, vol I: Metals.
Australas Inst Min Metal, Melbourne, pp 421 - 423
Finlow-Bates T, Stumpf! EF (1979) The copper and lead-zinc-silver orebodies of Mt. Isa mine,
Queensland: Products of ore hydrothermal system. Soc Geol Belg Ann 102:497 - 517
Gersteling RW, Heape JM (1975) The Cattle Grid orebody, Mt. Gunson, South Australia. Aust Inst
Min Metal Conf, S Aust, pp 103 -112
Goode ADT, Hall WDM (1981) The middle Proterozoic eastern Bangemall Basin, Western Australia.
Precambr Res 16:11-29
Goulevitch J (1975) Warrego copper-gold orebody. In: Knight CL (ed) Economic geology of Austra-
lia and Papua New Guinea, vol I: Metals. Australas Inst Min Metal, Melbourne, pp 430 - 437
Gulson BL, Perkins WG, Mizon KJ (1983) Lead isotope studies bearing on the genesis of copper ore-
bodies at Mount Isa, Queensland. Econ Geol 78:1445 -1466
Johnson W (1965) Copper and lead ore deposits of South Australia. In: McAndrew J (ed) Geology of
Australia ore deposits. Australas Inst Min Metal, Melbourne, pp 285 - 297
Knutson J, Donnelly TH, Tonkin DG (1983) Geochemical constraints on the genesis of copper miner-
alization in the Mount Gunson area, South Australia. Econ Geol 78:250 - 274
Lambert IB, Knutson J, Donnelly TH, Etminan H, Mason MG (1984) Genesis of copper mineraliza-
tion. Myall Creek Prospect, South Australia. Min Deposita 19:266 - 273
Mathias BV, Clark GJ (1975) Mount Isa copper and silver-lead zinc orebodies, Isa and Hilton mines.
In: Knight CL (ed) Economic geology of Australia and Papua New Guinea, vol I: Metals. Austra-
las Inst Min Metal, Melbourne, pp 351 - 372
Mitchell JW, Moore GP (1975) Mammoth copper deposit. In: Knight CL (ed) Economic geology of
Australia and Papua New Guinea, vol I: Metals. Australas Inst Min Metal, Melbourne, pp
383 -390
Neudert MK, Russell RE (1981) Shallow water and hypersaline features from the middle Proterozoic
Mount Isa sequence. Nature (London) 293:284-286
Page RW (1978) Response of U-Pb zircon and Rb-Sr total-rock and mineral systems to low-grade re-
gional metamorphism in Proterozoic igneous rocks, Mount Isa, Australia. J Geol Soc Aust
25:141-164
Page RW (1981) Depositional ages of the stratiform base metal deposits, at Mount Isa and McArthur
River, Australia, based on U-Pb Zircon dating of concordant tuff horizons. Econ Geol
76:648-658
Perkins WG (1984) Mount Isa "silica-dolomite" and copper ore bodies: the result of a systematic hy-
drothermal alteration system. Econ Geol 79:601 - 638
Plumb KA, Derrick GM, Wilson IH (1980) Precambrian geology of the McArthur River Mount Isa
Region, Northern Australia. In: Henderson RA, Stephenson PJ (eds) The geology and geophysics
of northeastern Australia. Geol Soc Aust, Queensland Div, Brisbane, pp 71 - 88
Pogson DJ, Suppel DW, Gilligan LB, Scheibner E, Baker C, Sherwin L, Felton EA, Fail AP (1976)
Recent studies of the tectonics, stratigraphy and mineralization of the Cobar-Mineral Hill region.
Aust Soc Explor Geophys 7:31- 34
Roberts DE, Hudson GRT (1983) The Olympic Dam copper-uranium-gold deposit, Roxby Downs,
South Australia. Econ Geol 78:799- 822
558 I. B. Lambert et al.: The Diverse Styles of Sediment-Hosted Copper Deposits in Australia

Rutland RWR, Parker AJ, Pitt GM, Preiss WV, Murrell B (1981) The Precambrian of South Austra-
lia. In: Hunter DR (ed) Precambrian of the southern hemisphere. Elsevier North Holland Biomedi-
cal Press, Amsterdam New York, pp 309 - 360
Scott KM, Taylor GF (1982) Eastern Creek volcanics as the source of copper at the Mammoth mine,
north west Queensland. BMR J Aust Geol Geophys 7:93-98
Sun SS (1983) Implications of Sand Pb isotope data to genesis of massive sulphide in the Cobar area.
6th Aust Geol Convent (Abstr), pp 307 - 308
Tonkin DG, Williams GE (1983) Fossil periglacial structures in the Cattle Grid mine, Mount Gunson.
Adelaide Geosyncline Sedimentary Environments and Tectonic Setting Symposium. Geol Soc
Aust, Adelaide, (Abstr), pp 59 - 60
Verwoerd PJ, Cleghorn JH (1975) Kanmantoo copper orebody. In: Knight CL (ed) Economic geolo-
gy of Australia and Papua New Guinea, vol I: Metals. Australas Inst Min Metal, Melbourne, pp
560-567
Warren RG (1972) A commentary on the metallogenic map of Australia and Papua New Guinea. Bur
Mineral Res Bull 145, p 85
Wright K (1965) Copper ore deposits of the Peko mine, Tennant Creek. In: McAndrew J (ed) Geolo-
gy of Australian ore deposits. Australas Inst Min Metal, Melbourne, pp 183 -185
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide
Deposits - A Quantitative Kinetic Approach
E. MERIN0 1,2, C. MOORE!, P. ORTOLEVA 1,2,3, and E. RIPLEyl,2

Abstract

The zoning typical of many sediment-hosted copper-iron sulfide deposits cannot

be well accounted for by equilibrium approaches, which lead to inconsistencies
and which are inherently ill-suited to describe a spatial, nonequilibrium process
such as zoning. Much better is a kinetic approach that consists of differential
equations representing dissolution, nucleation, growth, flow, and diffusion, all
of which are clearly involved in ore genesis. In this kinetic model, time and
spatial coordinates appear explicitly, and the equations automatically
incorporate feedbacks among mechanisms. Numerical solutions of the equations
for many combinations of reasonable values of the parameters involved (such as
velocity and composition of the mineralizing water, equilibrium and reaction-
rate constants, initial amount of pyrite in the rock, diffusion and nucleation con-
stants, and others) can yield mineral zoning of several types, including the typical
chalcocite-bornite-chalcopyrite-pyrite found at the White Pine deposit.

Introduction

Many sediment-hosted copper sulfide deposits are zoned. Among them are the
Kupferschiefer, the Zambian Copperbelt, White Pine, Seal Lake, Creta, Spar
Lake, Nacimiento, and Corocoro. A summary of their typical zoning sequences
with references is given in Table 1. Excellent review articles by Gustafson and
Williams (1981) and Bjorlykke and Sangster (1981) outline the geologic and geo-
chemical features of sediment-hosted stratiform deposits. Laznicka (1985) gives a
comprehensive review of the zoning; see also Ripley et al. (1985). The host
deposits are typically intracratonic, and contain both reduced and oxidized sedi-
ments. The copper mineralization occurs in the reduced sediments.
The purpose of this communication is to suggest that the observed mineral
zones in sediment-hosted copper deposits are caused by the nonequilibrium,
kinetic interaction of dissolution, flow, diffusion, nucleation, and growth.

I Geo-Chem Research Assoc., 400 E. First St., Bloomington, IN 47401, USA

Department of Geology, Indiana University, Bloomington, IN 47405, USA
Department of Chemistry, Indiana University, Bloomington, IN 47405, USA

Geology and Metallogeny of Copper Deposits

(ed. by G. H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
560 E. Merino et al.

Table 1. Summary of the most common zoning sequences in sediment-hosted copper deposits a

Deposit or district Typical Cu - Fe sulfide zoning sequence Reference

Zambia-Zaire Pyrite-Chalcopyrite-Bornite-Chalcocite van Eden (1974)

Annels (1979)
Fleischer et al. (1976)
Bartholome (1974)
Kupferschiefer Pyrite-Chalcopyrite-Bornite-Chalcocite Rentzsch (1974)
Jung and Knitzschke (1976)
Redstone River Pyrite-Chalcopyrite-Bornite-Chalcocite Jefferson (1978)
Lustwerk and Rose (1983)
White Pine Pyrite-Chalcopyrite-Bornite-Chalcocite Brown (1971,1974, 1981)
Seal Lake Pyrite-Chalcopyrite-Bornite-Chalcocite Gandhi and Brown (1975)
Spar Lake Pyrite-Chalcopyrite-Bornite-Chalcocite Harrison (1974)
Clark (1971)
Kansas Pyrite-Chalcopyrite-Bornite-Chalcocite Ripley et al. (1980)
Creta Pyrite-Chalcocite Dingess (1976)
Hagni and Cann (1976)
Texas Pyrite-Chalcocite Smith (1976)
Kansas Pyrite-Chalcocite Ripley et al. (1980)
Nacimiento Pyrite-Chalcocite Woodward et al. (1974)
Corocoro Pyrite-Chalcocite Ljunggren and Meyer (1964)

a Note that especially for the pyrite-chalcocite deposits, the term "zoning sequence" is essentially a
paragenetic sequence. In all of these deposits pyrite is inferred to be of early diagenetic origin, with
chalcocite deposition at a later time. In the production of the pyrite-chalcopyrite-bornite-chalcocite
sequence the distinction between zoning and paragenesis may be confusing in that different Cu or
Cu - Fe sulfides may be forming in spatially distinct areas of the ore system at the same time.

Most of the deposits contain early diagenetic pyrite, which is generally recog-
nized to have been partly replaced and rimmed by copper sulfides at tempera-
tures of less than about 100°C, under the action of basinal, meteoric brines with
more than 0.1 molal CI and pH 6 - 8. The typical zoning sequences are: barren
rock-chalcocite-bornite-chalcopyrite-pyrite, and barren rock-chalcocite-pyrite.
These zones can be in some way rationalized by equilibrium activity diagrams
and by equilibrium log (oxygen fugacity) versus pH diagrams. As shown below,
however, these diagrams lead to inconsistencies. Furthermore, it is unrealistic to
expect equilibrium diagrams to portray accurately the mineralization of pyritic
sediments, which is inherently a non equilibrium phenomenon. Also, zoning
being inherently spatial (in that the grains of the minerals involved must be
located in bands, and the bands must be placed in a definite sequence), it cannot
be accounted for by equilibrium approaches, which characteristically contain no
spatial coordinates.

Equilibrium Diagrams

Equilibrium activity diagrams pertinent to copper-iron sulfide sedimentary

deposits are shown in Fig. 1. It can be seen that both the chalcocite-bornite-chal-
 Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 561

-4 -4
(A) T = 125°C (8) T = 100°C
-5 log fO
2
= -52 -5 log fO = -54
2 I

-6 -6 Chalcocite I

~ -7 i -7
0 0 Chalcopyrite
"+ "+" -8
u" -8 u
R~ction
0 0
0> 0> I
E -9 E -9 path
I
I
-10 Pyrite -10 Pyrlite

I
-11 Hematite-I -11 I-Hematite
saturation I I saturation
-12 -12 '--------'----"----'---'---'------'
o 2 4 6 8 10 0 2 4 6 8 10
log 0Fe2+/o(H+)2 log 0Fe2+/o(H+)2

Fig. 1 A, B. Stability relationships between Cu- and Cu - Fe sulfide minerals. Fluid path for reaction
between a pyritic shale and a brine (Kettleman Dome, Merino 1975) is shown in B. Thermodynamic
data from Helgeson et al. (1978)

copyrite-pyrite and the chalcocite-pyrite sequences could be produced by reaction

between pyrite and a fluid whose composition evolved from top left to bottom
right of the diagrams, either to the right of the bornite-chalcopyrite triple point
or to the left of the chalcocite-bornite-pyrite triple point. (The digrams of Fig. 1
were constructed as shown by Helgeson (1968) and Brown (1970). Mineral-
mineral reactions were balanced on sulfide, which is therefore assumed in the
diagrams to be buffered by the system - in fact, the diagrams can be contoured
for the activity product aH+aHS-. The temperature, oxygen fugacity, and water
activity are fixed arbitrarily for the two diagrams; the values chosen for T and
logf~ are intended to be roughly typical of the deposits in question.)
A logf~ versus pH diagram for the same system is shown in Fig. 2. The
temperature chosen is 100 D C, total sulfide is 10 -4 molal, and total chloride is 1
molal. The diagram suggests that chalcocite can form at fairly high oxygen
fugacities (definitely above the pyrite-hematite boundary) and for realistic and
significant amounts of dissolved copper (mostly complexed by chloride, see Rose
1976, Helgeson et al. 1978, and Crerar and Barnes 1976). The problem, however,
is that the equilibrium crystallization of chalcocite falls in the field of predomi-
nance of SO~- (aq), where the concentrations of S2- and HS - are insignificant.
Geologic evidence in most deposits suggests that deposition of copper sulfides
occurred under much lower f ~ than indicated by the diagram. Indeed, the host
sediment normally contains abundant organic matter and pyrite. A second dif-
ficulty in this diagram is that the crystallization of chalcopyrite is predicted to
happen at very small concentrations of dissolved copper, of the order of 0.1 mg
1-1, thought insufficient to form ores (Anderson 1977). These two inconsisten-
cies, together with the general considerations given at the end of the Introduc-
 562 E. Merino et at.

-45 /
/
-46 /
/
-47 /
/
-48 /
/
/ /
-49 /
/
Cu /
0 -50
"- - - -
.
'-:- - - - - 10 - - -
Cc
- -
0'
0

-53

-54

-55

-56

4 5 6 7 8
pH

Fig. 2. Logj02-pH diagram showing stability fields of aqueous sulfur species, pyrite, hematite, and
various copper minerals. Total ppm Cu is essentially that contained in CuCt' + CuCl2 + CuC~ - .
Conditions of the diagram are as follows: T= 100°C, E S = 10- 4 m, Cl- = 1 m. Abbreviations: CU
native copper; CC CU2S; BN CuSFeS4 ; CP CuFeS2; PY Fe~; HEM F~03' Thermodynamic data
from Helgeson et at. (1978)

tion, have compelled us to explore quantitatively the possibility that the crystal-
lization and zoning of copper sulfides in sediments is a nonequilibrium process
involving transport and reaction.

The Kinetic Approach

In the genesis of these sulfide deposits and their zoning aqueous solutions both
flow through and react with permeable rock. The essential feature is that
reaction, nucleation, and transport take place simultaneously, and that they
interact.
There are very powerful techniques to model the reactions between waters
and rocks (e.g., Helgeson 1968, 1979, Wolery 1979, Reed 1982), but because
these approaches are blind to space and time they are fundamentally unable to
describe transport, to predict spatial mineralization patterns, or to incorporate
reaction-transport interactions. Similarly, there are sophisticated models to
predict the flow of water through permeable rocks (e.g., Sharp 1976, Cathles
1981, Narasimhan and Witherspoon 1976, Kuyahorn and Pinder 1984), but these
models take no account of chemical reactions, or even less of their interactions
with transport. Attempts to juxtapose chemical models and flow models are
bound to require prohibitively long computing times, even in the fastest com-
puters, and, worse, such juxtaposition would still be unable to incorporate trans-
port-reaction interactions.
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 563

In contrast, our approach (also attempted by Dobrovolsky and Lyalko 1979)

is kinetic throughout. During the evolution of the system chemical equilibrium is
nowhere assumed. We write differential equations representing mass conserva-
tion (or continuity), flow, diffusion, nucleation, and growth and dissolution
rates. For simplicity, absorption and adsorption of aqueous species are ignored
here, but they could be easily included in the continuity equations. The system of
equations automatically incorporates feedbacks and interactions among the
various mechanisms. Space and time appear explicitly. Any number of aqueous
and solid species can be included; in our illustration below we have included the
ten species (six aqueous, four solid) listed in Table 3. This general approach is
common in chemical engineering and physical chemistry and has been applied to
study the dynamics of redox fronts by Ortoleva et al. (1986) and Chadam et al.
(1986a, b).
Mass Conservation. The continuity equation is simply
ac· = -V.J+R- (1)
at
_I
1 1

and states that the time variation of the concentration (Ci) of species i (aqueous or
solid) in a small volume of the system is due to the amount of i transported into
that volume per unit time plus the total rate, R i , of production or consumption of
i through reactions within the volume of interest. V is the spatial derivative
vector (a/ax, a/ay, a/az); and J is the total flux of species i. Neglecting
hydrodynamic dispersion,
J. = Jflow +J~iffusion = vc·- D·V c· (2)
I I I I I "

where v is the percolation flow velocity, Di is the effective diffusion coefficient,

and VCi is the concentration gradient for species i. The rate of production or con-
sumption, Rio for species i is related to the rates of reactions, W u , via the
stoichiometric coefficients v~:
(3)

Reaction Rates and Nucleation. In the absence (to our knowledge) of well-estab-
lished elementary reactions and kinetic data for the dissolution and growth of the
iron-copper sulfides involved here, we have adopted the stoichiometric reactions
of Table 2 as the basis for reaction-rate equations. For example, for the reaction
of dissolution or growth of pyrite
FeS2(py)+ H 20 = Fe2+ + 2HS - + t02(aq), (4)
and following our work on redox-front dynamics and deposition (Ortoleva et al.
1985, Merino et al. 1985, Chadam et al. 1985a, b), we construct the reaction-rate
expression
W = kpy(P+ p)2/3 (Kpy- {Fe 2+}{HS-f{02}112) , (5)
where Kpy and kpy are the equilibrium and reaction-rate constants, the braces
denote concentrations in moles per unit volume of bulk rock, P is the volume
percent of pyrite present at a point at a given moment, p is a "switch" that
564 E. Merino et al.

Table 2. Reactions used for the initial simulation

FeS2+H20 +:t Fe2+ +2HS- +t02(aq)

CU2S+H+ +:t 2Cu+ +HS-
CuFeS2+H+ +tH20 +:t Cu+ +Fe2+ +2HS- +-1-02(aq)
Cu,FeS4+3H+ +tHzO +:t 5Cu+ +FeZ+ +4HS- +-1-02(aq)
Fe2+ +H+ +-1-02 +:t Fe3 + +tH20

controls the nucleation of pyrite as explained below, the expression

(Kpy- {Fe + +}{HS -}2{02}112) is the chemical affinity for the pyrite rea.ction, and
W is the rate of growth (W < 0) or dissolution (W> 0) of pyrite. Let
Knuc1eation( > Kpy) be the value of {Fe + +}{HS -}2{02{112 above which pyrite nuclea-
tion starts. Equation (5) is reasonable because (a) it takes into account the surface
area dependence of the rates through the 2/3 exponent, which converts volume
into grain surface; (b) it yields the correct equilibrium condition when the rate W
vanishes (assuming the solutions are dilute); and (c) it takes account of nuclea-
tion:
1. if {Fe + +}{HS -}2{02}112 > Knuc1eation, then p is taken as a small positive
number, W becomes negative even if there is no pyrite present, and pyrite
nucleates and grows;
2. if {Fe + +}{HS -}2{02}112 < Knucleation, then p is defined as zero; now if P = 0
(i.e., there is no preexisting pyrite) W = 0, which means that there is no pyrite
growth, and
if P> 0 (i.e., there is some pyrite present already) then pyrite will grow if
{Fe + +}{HS-}2{02}1I2 > Kpyo

Thus, the formulation of Eq. (5) is able to "turn on" the precipitation of
pyrite when the nucleation threshold is exceeded; allows for the dissolution of
pyrite when the solution is undersaturated; allows for pyrite growth where there
is already some pyrite present and the solution is supersaturated; and "turn off"
growth where there is no pyrite present and the solution does not exceed the
nucleation threshold. Rate expressions of the type exemplified by Eq. (5) may be
inserted in the several expressions of the continuity equation (one for each
species). These expressions then automatically incorporate feedbacks among
reactions and between transport and reaction. More general expressions for non-
dilute solutions can be obtained by replacing concentrations by activities.
To complete the theory we have to determine the percolation flow velocity v.
This is related to the texture-dependent permeability x and the pressure p via
Darcy's law
v=-xVp. (6)

The conservation equation for water, assumed incompressible, yields

v. (f/JxVp) = af/J (7)

aT
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 565

where ¢ is the porosity and r is scaled time. The porosity is related to the grain
size and number density of the various minerals making up the rock through the
equation
¢+ r.i niLj = 1 , (8)
where Li is the local size of grains of i and ni is the number of grains of i per unit
volume of rock.

Modeling of the Cu - Fe Sulfide Zoning in Sediment-Hosted Deposits

In this attempt a modeling the zoning of sedimentary copper deposits we have

used numerical values that allow quickness in computation and do not violate the
quantitative relationships among imposed water compositions, initial rock com-
position, and equilibrium constants. We have chosen the ten-species system
shown in Table 3. Boundary concentrations refer to the concentrations of all the
elements in the system before the introduction of the inlet fluid (or fluid that
carries out the alteration to copper sulfides). We have chosen that the Fe 2 +,
Fe 3 +, 02, and H + be in equilibrium with each other in the initial water and also
that the initial water be in equilibrium with pyrite.

Scaling. It is necessary, again for computational ease, to scale concentrations and

equilibrium constants as follows:
1. Concentrations are scaled by dividing the initial and boundary concentra-
tions for each species by the larger of the two values. For example, if the initial

Table 3. Scaled parameters used for the discussed simulation (see text for details)

Species Concentrations

Boundary Initial

Cu+ 1.0 1 X to-I

FeZ + 1.0 1.0
HS- 1 X to-I 1.0
F~+ 3.2 X 10- 2 3.6 X 10- 2
°2(aq) 1.0 6.4 X to-I
H+ 9.5 X 10- 1 1.0
Pyrite 1.0
Chalcocite 0.0
Chalcopyrite 0.0
Bornite 0.0
Equilibrium constants
Kpyrite 6.5 x 10- 1
Kchalcocite 9 X to- 2
Kchalcopyrite 9.5 X to- 2
Kbornite 8.0 X 10- 4
Kferrous-ferric = 3.0 X 10- 2
566 E. Merino et aI.

copper concentration is 10 -18 and the boundary value is 10 -15, the concentra-
tions scale to 10 - 2 and 1, respectively.
2. To scale equilibrium constants, the known equilibrium constant is divided
by the corresponding ion product for the reactions, using in that ion product the
maximum unscaled value for each species.
For example, for the values Fe 2+ initial = 10- 12, Fe 2+ boundary = 10- 14,
HS- initial = 10- 3, HS- boundary = 10- 6, 0~12 initial = 10- 3°, and 0~12
boundary = 10 -28; and given that the equilibrium constant for Eq. (4) is
Kpy = 10 -45, the scaled equilibrium constant turns out be K%,sc = Kpyl(Fe 22
initial)(HS- initial)2(02 boundary) 1/2 = 10-45/10-12(10-3)210- = 10.

Numerical Simulations. We have solved these equations numerically with a com-

puter. The simulations require the following input of data: initial and inlet con-
centrations for each species (Table 3), diffusion coefficients for each mobile
species, flow velocity of the fluid, and, for each reaction: equilibrium constant,
rate constant, and supersaturation factor for nucleation. Also needed are certain
ancillary parameters for controlling the computer program and describing the
size of the system.
Of the above variables, only the equilibrium constants are well known. All
the other parameters can only be estimated within a reasonable range. These are
the parameters we have varied from computer run to computer run in order to
observe what precipitation patterns, if any, develop as the system evolves. The
lack of data, especially kinetic data, gives some freedom in choosing values for
these parameters, but one must be careful, as the evolution of the system is highly
dependent on some of them. A good example is the supersaturation factor, a
small change which can drastically change the localization of a given phase or
completely inhibit its nucleation and growth. Another example is the rate con-
stant: when the rate constant is very high for a phase and that phase nucleates,
the resulting rapid growth can effectively inhibit growth of other phases.
Because of these uncertainties we show here only that mineral zoning in sedi-
ment-hosted copper deposits can result. Much experimental work is needed to
determine values for the kinetic variables, and very detailed field work describing
the geometry and textures of the ore deposits is required. Assuming that this
model is reasonable, it is interesting to note that one could work backwards to get
order-of-magnitude figures for some of the parameters by attempting to
duplicate known ore-body geometries.
Figure 3 indicates the results of one specific computer run, generated with the
parameters of Table 3. We show here the concentrations of the various sulfide
minerals on the ordinate versus space (distance from the inlet) on the abscissa.
Water flows from left to right. Each graph represents a scaled time interval; the
one for t = 0 represents the initial conditions (i.e., all pyrite), and the one for
t = 30 represents the last calculated step of that particular simulation, which
yields precisely the kind of zoning that is commonly observed.

Discussion. The only sulfide in the initial system is pyrite. As this initial pyrite is
engulfed by the incoming water (with which it is out of equilibrium) it dissolves,
yielding waters enriched in HS -, Fe 2+, and O 2. These ions are transported with
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 567

Fig. 3. Modeled evolution of a

1=0
zoned sequence of copper-iron
sulfides in a sediment-hosted de-
posit. Symbols are explained on
the diagrams. Mineral concen-

~.. I~
trations were calculated by
numerical solution of the model
<r
described in the text, and are
here plotted vertically as func-

~ I~
tions of position and (scaled)
time t. The parameter values
u adopted in the calculations are
in Table 3
LEGEND

\~ DpYRITE
f2ZI CHALCOPYRITE
m:I BOIlNITE
_ CHALCOCITE

the fluid and, along with the Cu + in it, eventually exceed the nucleation concen-
tration threshold, precipitating chalcopyrite at t = 5. The chalcopyrite grows as
long as there is pyrite upflow or adjacent to it to provide the necessary ions.
When all the upflow and adjacent pyrite is dissolved then the chalcopyrite too be-
gins to dissolve. The concomitant release of HS -, Fe 2 +, O 2 and especially the
previously fixed copper, along with the depletion of H +, cause the solution to
exceed the concentration product required for nucleation of bornite, and bornite
precipitates. At this point an even larger amount of copper is fixed in the system.
This situation is shown at time t = 5 in Fig. 3, and continues throughout. As time
progresses all the chalcopyrite upflow of the bornite dissolves, cutting off the
stabilizing supply of ions, and bornite starts to dissolve. This abundant release
of Cu + and HS - and decrease in H + cause chalcocite nucleation and growth, as
shown in the plot for time t = 15 in Fig. 3. The concentrations of the copper-
bearing species grow with time, but appear to approach a steady-state level (note
chalcopyrite). Displacement of the mineral zones themselves is accomplished by
dissolution at the upflow end and precipitation downflow.
The width and concentrations in each zone depend on the parameters chosen,
most notably on the concentrations, degree of supersaturation needed for nuclea-
tion, and rate constant. High values of these parameters produce narrow, high-
concentration zones. We have not determined whether total copper fixed in the
system remains the same as the parameters are varied.
Preliminary investigations show that the chalcocite-bornite-chalcopyrite-
pyrite zoning may be produced by a relatively wide range of values for input and
initial water chemistries. Indeed, within a given range, it is insensitive to whether
the incoming waters have a lower or higher pH than the initial waters. In addi-
tion, other sequences of precipitation can be produced, but the same zoning
pattern eventually evolves. This may explain the occurrence at White Pine, where
chalcocite apparently replaced pyrite, and thin bornite and chalcopyrite zones
formed at the interface between chalcocite and unreplaced pyrite.
The results of three other computer runs are given in Fig. 4. Run c is the same
as in Fig. 3. Run a is for the same parameters as c except that copper concentra-
 568 E. Merino et a!.

CullVl (a) Cul l VlO (b) (c)

.I ~~ ' fC<:: 1 5~, r

.1
c
.2 '0
~
c:
Q)
1
u 15
c
1 '5 ~
o
U

20 20 20

Space
I ~
Space Space
o pyrite o bornite ~ chalcopyrite IlIIIIIm chatcocite
Fig. 4a-c. Other numerical solutions of the zoning model for iron-copper sulfides. Case (c) is
essentially that of Fig. 3, and is repeated here for reference. Case (a) was run for a mineralizing
solution ten times poorer in copper than that in (c). Case (b) was run for a solution ten times poorer in
copper and ten times faster than in (c)

tion in the mineralizing water was ten times lower than in c. Run b is for a
mineralizing solution ten times lower in copper and ten times faster than in c.
Note that mineral zonings are produced in the two new cases, a and b; that the
zones shift faster downflow in b than in a or c, and that a high degree of zoning is
reached in case a earlier than in cases band c.
The fact that these transport-reaction models succeed in duplicating well-
documented features of zoned, sediment-hosted copper deposits as well as some
aspects of roll-front uranium deposits (Moore et al. 1983, Chadam et al. 1986b)
suggests that transport-reaction coupling is an important process in the
formation of ore deposits and may playa role in other rocks showing chemical
localization. It should be stressed that so far the model only shows the possibility
that the zoning can arise in this way. Ongoing work is adding more support to
these assertions, but the true test cannot be made until much more information is
available on the kinetics of the reactions in question, the compositions of the
boundary and initial fluids, and the geohydrology of the host sediments.

Conclusions

Understanding the genesis of the copper sulfide zoning in sedimentary ore

deposits is not easy. Some of the difficulties one encounters in taking an equilib-
rium approach to account for the zoning are that the process of ore formation is
inherently a nonequilibrium one that involves reaction and transport; that in the
chalcocite field the predicted predominant sulfur species in sulfate, not sulfide;
and that in the chalcopyrite field the largest dissolved copper concentra-
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 569

tion possible is 0.1 mg 1- 1, which intuitively appears too low to give rise to an ore
deposit. In contrast to the equilibrium approach, we present above a quantitative
kinetic model of ore deposition that accounts for the zoning. The model is cast as
a system of differential equations that include dissolution, nucleation, growth,
flow, and diffusion. One strong advantage of the model is that it explicitly
contains spatial coordinates, which are necessary to describe a spatial feature
such as zoning. Time is also explicitly taken into account. The system of
equations has been solved numerically for many combinations of reasonable
values of the following parameters: velocity and composition of the mineralizing
aqueous solution, composition of the formation waters before mineralization,
equilibrium and rate constants for several reactions, diffusion coefficients of
ions in aqueous solution, nucleation constants, and initial modal abundance of
pyrite in the still-unmineralized sedimentary rock. Many of the numerical solu-
tions yield, even for very low-copper waters, patterns of precipitation that are
zoned and that evolve with time. Zonings obtained include the typical White Pine
sequence of chalcocite-bornite-chalcopyrite-pyrite, as well as chalcocite-pyrite-
chalcocite-bornite-pyrite, chalcocite-chalcopyrite-pyrite, and the striking
bornite-chalcocite-chalcopyrite-pyrite.
Much work remains to be done, but the validity, viability, and power of this
kinetic approach are, we think, clear. Interestingly, the approach also allows one
to find out easily to which parameters the mineralization and zoning are most
and least sensitive. Those most influencial parameters are the ones deserving the
most careful experimental determination.

Acknowledgments. The authors thank the U.S. Department of Energy for financial support of this
research under Grant DE-AC02-82ER12074 (Office of Basic Energy Sciences, Engineering Research
Program).

References

Anderson GM (1977) Thermodynamics and sulfide solubilities. In: Greenwood HG (ed) Application
of thermodynamics to petrology and ore deposits. Mineralog Soc Can Short Course, pp 136 -150
Annels AE (1979) Mulfulira greywackes and their associated sulfides. Inst Min Metallurg Trans Sec B
88:15-23
Bartholome P (1974) On the diagenetic formation of ores in sedimentary beds, with special reference
to Kamoto, Shaba, Zaire. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes.
Liege, Soc Geol Belg, pp 203 - 213
Bjorlykke A, Sangster DF (1981) An overview of sandstone lead deposits and their relation to red-bed
copper and carbonate-hosted lead-zinc deposits. Econ Geol 75th Anniv Vol, pp 179-213
Brown AC (1971) Zoning in the White Pine copper deposit, Ontonagon County, Michigan. Econ
Geol 66:543 - 573
Brown AC (1974) An epigenetic origin for stratiform Cd - Pb - Zn sulphides in the Lower Nonesuch
Shale, White Pine, Michigan. Econ Geol 69:271 - 274
Brown AC (1981) The timing of mineralization in stratiform copper deposits. In: Wolf KH (ed)
Handbook of strata-bound and stratiform ore deposits, chap 1, vol IX. Elsevier, New York, pp
1-24
Brown TH (1970) Theoretical predictions of equilibria and mass transfer in the system CaO - MgO-
Si02 -H20-C02 -NaCl-HCI. Ph D Thesis, Northwestern University, Evanston, Illinois
(unpublished)
570 E. Merino et al.

Cathles LM (1981) Fluid flow and genesis of hydrothermal ore deposits. Econ Geol 75th Anniv Vol,
pp 424-457
Chadam J, Hoff D, Merino E, Ortoleva P, Sen A (1986a) Porosity instability at reactive percolation.
J Inst Math Analysis (Oxford). Accepted
Chadam J, Merino E, Ortoleva P, Sen A (1986b) Geochemical self-organization, II: The reactive-in-
filtration instability. Am J Sci (submitted)
Clark AL (1971) Stratabound copper sulfides in the Precambrian Belt Supergroup, North Idaho and
Northwestern Montana. Soc Min Geol Jpn Spec Iss 3:261 - 267
Crerar DA, Barnes HL (1976) Ore solution chemistry versus solubilities of chalcopyrite and chalco-
cite assemblages in hydrothermal solution at 200°C to 350°C. Econ Geol 71 :772 -794
Dingess P (1976) Geology and mining operations at the Creta Copper deposit of Eagle-Pitcher Indus-
tries, Inc. In: Johnson K, Croy R (eds) Stratiform copper deposits of the midcontinent region, a
symposium. Oklahoma Geol Surv Circ 77:15 - 25
Dobrovolsky EV, Lyalko VI (1979) Dynamical model of fluorite mineralization of carbonate-bearing
rocks. Am J Sci 279:1022-1032
Eden JG van (1974) Depositional and diagenetic environment related to the sulfide mineralization,
Mulfulira, Zambia. Econ Geol 69:59 - 79
Fleischer VD, Garlick WG, Haldane R (1976) Geology of the Zambian copperbelt. In: Wolf KH (ed)
Handbook of strata-bound and stratiform deposits, chap 6, vol VI. Elsevier, New York, pp
223 -352
Gandhi SS, Brown AC (1975) Cupriferous shales of the Adeline Island Formation Seal Lake Group,
Labrador. Econ Geol 70:145 -163
Gustafson LB, Williams N (1981) Sediment-hosted stratiform deposits of copper, lead, and zinc.
Econ Geol75th Anniv Vol, pp 139-178
Hagni R, Gann D (1976) Microscopy of copper ore at the Creta mine, Southwestern Oklahoma. In:
Johnson K, Croy R (eds) Stratiform copper deposits of the midcontinent region, a symposium.
Oklahoma Geol Surv Circ 77:40 - 51
Harrison JE (1974) Copper mineralization in miogeosynclinal clastics of the Belt Supergroup, North-
western United States. In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes.
Liege, Soc Geol Belg, pp 353 - 366
Helgeson HC (1968) Evaluation of irreversible reactions involving minerals and aqueous solutions, I.
Thermodynamic relations. Geochim Cosmochim Acta 32:853 - 877
Helgeson HC (1979) Mass transfer among minerals and hydrothermal solutions. In: Barnes HL (ed)
Geochemistry of hydrothermal ore deposits. Wiley & Sons, New York, pp 568-610
Helgeson HC, Delany JM, Nesbitt HW, Bird DK (1978) Summary and critique of the thermodynamic
properties of rock-forming minerals. Am J Sci 278A:1-229
Jefferson CW (1978) Stratigraphy and sedimentology, Upper Proterozoic Redstone Copper Belt,
Mackenzie Mountains, N.W.T. - A preliminary report. In: Mineral industry report for 1975,
Northwest Territories, India and Northern Affairs. Econ Geol Ser, pp 1-13
Jung W, Knitzschke G (1976) Kupferschiefer in the German Democratic Republic with special refer-
ence to the Kupferschiefer deposit in the southeastern Harz Foreland. In: WolfKH (ed) Handbook
of strata-bound and stratiform deposits, chap 7, vol VI. Elsevier, New York, pp 353 - 406
Kuyahorn PS, Pinder GF (1984) Computational methods in subsurface flow. Academic Press, Lon-
don New York, 473 pp
Laznicka P (1985) Zoning and ores. In: Wolf KH (ed) Handbook of strata-bound and stratiform de-
posits, vol XI. Elsevier, New York (in press)
Ljunggren P, Meyer HC (1964) The copper mineralization in the Corocoro Basin, Bolivia. Econ Geol
59:110-125
Lustwerk RL, Rose A W (1983) Source and segregation of transition metals during diagenetic forma-
tion of the Redstone Stratiform Copper Deposit. MacKenzie Mountains, N.W.T. (abs). Geol Soc
Am (Abstracts with Programs) 15:632
Merino E (1975) Diagenesis in Tertiary sandstones from Kettleman North Dome, California-II. Inter-
stitial solutions: distribution of aqueous species at 100°C and chemical relation to the diagenetic
mineralogy. Geochim Cosmochim Acta 39:1629-1645
Moore C, Auchmuty G, Chadam J, Hettmer J, Larter R, Merino E, Ortoleva P, Ripley E, Sen A
(1983) Kinetic modeling of flow-reaction instabilities in porous rocks: Application to uranium roll
fronts. 4th Int Symp Water-Rock Interaction, Misasa, Japan, pp 352 - 355
Mineral Zoning in Sediment-Hosted Copper-Iron Sulfide Deposits 571

Narasimhan TN, Witherspoon PS (1976) An integrated finite-difference method for analyzing fluid
flow in porous media. Water Resources Res 12:57 - 64
Ortoleva P, Auchmuty G, Chadam J, Hettmer J, Merino E, Moore C, Ripley E (1986) Redox front
propagation and banding modalities. Physica D 18 (in press)
Reed MH (1982) Calculation of multicomponent chemical equilibria and reaction processes in sys-
tems involving minerals, gases, and an aqueous phase. Geochim Cosmochim Acta 46:513 - 528
Rentzsch J (1974) The Kupferschiefer in comparison with the deposits of the Zambian Copperbelts.
In: Bartholome P (ed) Gisements stratiformes et provinces cupriferes. Liege, Soc Geol Belg, pp
403-426
Ripley EM, Lambert MW, Berendsen P (1980) Mineralogy and paragenesis of redbed copper mineral-
ization in the Lower Permian of South Central Kansas. Econ Geol 75:722-729
Ripley E, Merino E, Moore C, Ortoleva P (1985) Mineral zoning in sediment-hosted copper deposits.
In: Wolf KH (ed) Handbook of strata-bound and stratiform ore deposits, vol XIII. Elsevier, New
York, pp 237 -260
Rose AW (1976) The effect of cuprous chloride complexes in the origin of red-bed copper and related
deposits. Econ Geol 71: 1036 - 1048
Sharp JM (1976) Momentum and energy balance equations for compacting sediments. J Int Assoc
Math Geol 8:305 - 322
Smith GE (1976) Sabkha and tidal-flat facies control of stratiform copper deposits in north Texas. In:
Wolf KH (ed) Handbook of strata-bound and stratiform ore deposits, chap 8, vol VI. Elsevier,
New York, pp 407 - 447
Wolery TJ (1979) Calculation of chemical equilibrium between aqueous solutions and minerals: The
EQ3/6 software package. Lawrence Livermore National Laboratory, Livermore, California. Publ
UCRL-52658
Woodward LA, Kaufman WH, Schumacher OL, Talbott LW (1974) Stratabound copper deposits in
Triassic sandstone of Sierra Nacimiento, New Mexico. Econ Geol 69:108 -120
Results of Recent Exploration for Copper-Silver Deposits
in the Kupferschiefer of West Germany
F.-P. SCHMIDT!, C. SCHUMACHER!, V. SPIETH 2, and G. FRIEDRICH 3

Abstract

In two areas of the Werra-Fulda trough (Hessia, West Germany) exploration of

Kupferschiefer-type copper-silver deposits was carried out by St. Joe Explora-
tions GmbH Western Europe.
Anomalous high copper-silver contents were detected in the Spessart-RhOn
area as well as in the Richelsdorf area. Especially in the Richelsdorf area all
features connected with the occurrence of economic mineralization are similar to
the deposits in Poland.
The relationship between grade of mineralization, spatial distribution of
metals and minerals, paleogeography and the occurrence of "Rote Faule" in two
St. Joe drill holes and two holes drilled by the BGR (Bundesanstalt fUr Geowis-
senschaften und Rohstoffe, Hannover) is discussed.

Introduction

The Central European Kupferschiefer sea covered an area which today is located
in West Germany, East Germany and Poland.
Economic copper-silver deposits occur in the Lower Zechstein strata of all
three countries.
The largest ore deposits are in the Lubin district in Poland. At Lubin 1 000
million tons of ore are located in the Fore-Sudetic Monocline at the boundary of
a Rotliegend basin to a Variscian basement high. The mineralization is concen-
trated in the Kupferschiefer itself as well as in the underlying Weissliegend and
the overlying Zechsteinkalk.
Since 1978 St. Joe Explorations GmbH has conducted an exploration
program which is focused on the discovery of a Polish-type, copper-silver deposit
in the Lower Zechstein strata of West Germany. This work is done in joint
venture with Preussag AG Metall, Goslar and is supported by the Government of
the Federal Republic of Germany (BMWi).

I St. Joe Explorations GmbH, GeorgstraBe 50, 3000 Hannover 1, FRG

2 Gewerkschaft Wilhelm, HohenzollernstraBe 2, 3000 Hannover 1, FRG
3 Institut fUr Mineralogie und Lagerstattenlehre der RWTH, WiillnerstraBe 2,5100 Aachen, FRG

Geology and Metallogeny of Copper Deposits

(ed. by G.H. Friedrich et al.)
© Springer-Verlag Berlin Heidelberg 1986
Results of Recent Exploration for Copper-Silver Deposits in the Kupferschiefer 573

Investigations of the parameters indicating a local 'epi(dia)genetic' copper-

silver enrichment in the Lubin area of Poland and application of the results to
previously untested areas in West Germany have led to the identification of
several target districts.
In two target districts of Spessart-RhOn and Richelsdorf 60 drill holes have
been sunk to depths between 150 m and 780 m. The investigations of the inter-
sected strata have improved the knowledge of the paleogeography, the Variscian
basement, the Rotliegend, and the mineralized Lower Zechstein sequence of the
Weissliegend (S 1), Kupferschiefer (T 1) and Zechsteinkalk (Ca 1) in the Werra-
Fulda trough.

----
=> Ar(lo o ,w ithou t Kupfe r ~ C h l~ f .. r

t
7"

/ Boundary of h~ Rol il egeno

bas in
(l '5 km
Torgel Area R icn e l ~oorl

Tor~et Ar pa Sppssort - Rhon

: •...

ooR
- ,
==lo.

Npuhol
Trough .....
1
Schluchl prn er Fronk e"
Mul de Trough

SR

Fig. 1. Target ares for exploration on Kupferschiefer-type deposits in West Germany

574 F.-P. Schmidt et al.

Target Areas

Two main features are characteristic for Kupferschiefer-type deposits.

(1) The geological setting at the boundary between the Rotliegend troughs
and the basement highs and (2) the occurrence of red-coloured, oxidized rocks,
called "Rote Faule" in the vicinity of the ore deposits.
The formation of copper-silver deposits related to Rote Faule is interpreted as
an 'epi(dia)genetic' enrichment according to Rentzsch (1974) and Rydzewski
(1978).
In addition, the relationship to growth faults has been presumed by Ekiert
(1960).
The Richelsdorf area and the Spessart-RhOn area which cover parts of the
Werra-Fulda trough were selected as favourable sites for exploration by St. Joe
Explorations GmbH (Fig. 1).

o 2 L 6 ~m

N~ullof

.
ISCllOfShp;m
~

•
•
•
•

Fig. 2. Distribution of copper facies and lead-zinc facies in the Spessart-Rhon area
Results of Recent Exploration for Copper-Silver Deposits in the Kupferschiefer 575

SR 13 "Kohlgrund"
6 20 m
Co 1 1 i Pb
f- ,
... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... .. .. . ... .. ""'••.a-.... _~."V'"
:
,• :
!
-i
T 1 r-:- ._ j ..t~.~. !
I ,
I-:"" 0;... "'

"5 ,~........,~ .....-.-.... ....... !

- "-~""""-I:>I- ---------------
Zn
!

-
S1
-
-0 Ag Cu
621. '"
100 200 300 l.OO 500 10ll 2Oll )(XX) t.OOO SOOOppm

Fig. 3. Distribution of copper, lead, zinc and silver in the vertical section of drill hole SR 13

Spessart-Rhon Area

In the Spessart-Rhon area only few drill data were available. A recent publication
by Schumacher et al. (1984) based on a large amount of new drill data gave rise to
new ideas about the regional geological setting in this area and focused on the
distribution of the boundary between the Rotliegend sediments and folded
basement rocks.
The main structural element is the Rhon high, where an adjacent zone of
copper predominance was discovered extending over an area of about 200 km 2
(Fig. 2).
The highest metal contents are 1.1070 copper plus 70 ppm silver calculated
over a suitable mining height of 2 m in the northern RhOn area.
Copper-arsenic sulphides and arsenides are abundant. In the economically
interesting zone the minerals bornite, chalcocite and stromeyerite occur.
The Spessart-Rhon area is characterized by anomalous silver contents, caused
by the presence of a tennantite phase rich in silver related to Saxon tectonics
(Friedrich et al. 1984). In drill hole SR 13 the distribution of copper, lead, zinc
and silver (Fig. 3) and the distribution of the ore minerals (Fig. 4) in the econom-
ical copper zone can be demonstrated.

Richelsdorf Area

In the Richelsdorf area metal enrichment is connected with the presence of Rote
Faule which represents the oxidized portion of a redox boundary. The Rote Faule
 576
F.-P. Schmidt et al.

I,keln
Boh r - St rQ II - II
gr aph"
tn str cc dg (V bn <p gn $ph py m,
O lU m

..
••
, ,
~

_or u
0
I=t ;:r:/o:. I I I I

\:tJ~
-"'-"
v _
.. 0 I .I , , I

. ,", I I
N-" I I

.
,. , , , ,
II ,
,
I
I
"
I t I
I
,
II

, " ,
•
I II

621m " " I

,• II
I I
.., , ,, "III I

••I
, ,
I
r
L
or I
..
or
.cv ,
,
~
... I I, I , , , ,
'"" , , , , I
,
I
I
c--
7 -. :-:
I

;' ,

~I
~::J .~; 622 m
o· ... ..

·,
•• •
# .. : . " .. :

, .,
.. .....
:~::.;.:~ : ~!~
'
.0 0 .:. . ~. ,
I
", ,
, I
" ,
I II

.''..."'....
'. o . • II
.. " , , "Ii
I

-.
.~~.! .-.....; ,,
II "

•· •.. •I
II , I I

-
1 ,- • '",
:-;; :;!.e.:
. ..... ,,- ~ , ,~,
......... •
623 m "I "
o : • \ '. ,.
I I
..'.
I
• e,"~ •.•• • , , , :
...........
~' • .•• " t. , ~
, "
................ I"
I
o ~ " ......,
I I I I
..'" ·· . .
~
-t ... , ... , "
..' '"c
""

........
..........
"" t

... ,
··
I

I I,
CT

3
..
c:
I I
, ,
I
I I

' ~~-""..L6 2~
~ m

Fig. 4. Distribution of ore minerals in vertical section of drill hole SR 13. Abbreviations: IllOllingite;
tn tennantite; str stromeyerite; ee chalcocite; dg digenite; bn bornite; ep chalcopyrite; gn galena; sph
sphalerite; py pyrite; me marcasite

contains hematite and iron-hydroxides. The main structural element is the Baum-
bacher high, where no Rotliegend sediments were deposited. The replacement of
sulphides by hematite indicates an 'epi(dia)genetic' nature of the Rote Faule,
which extends into the reduced facies of the basal Zechstein strata formed under
synsedimentary conditions.
Due to the changing redox conditions, a zonation of Rote Faule facies,
copper facies and lead-zinc facies occurs.
Results of Recent Exploration for Copper-Silver Deposits in the Kupferschiefer 577

Pb-Zn
•

,.~:-:-~~::' :::::. : : \.~ -:<~ ::: ~ •

::::
· · ...:RQt. e:·.Foule::::
........ ..
:'~
. . . .............
, \

:. :. :. :. :. :.:. :. :. :. :.:. :. :. :. :. ~,
\.

.....
. ............. .. . . ............ . ,
·· ..
, ' , ' , ' , ' , ' , . , ' , ' , ' , ' , ' , ' , ' , ' ,"

· . . . . . . . . . . .. ...........
.....
, ' , ' , ', ' ,',',',',' , ',',',',',',','

·. .. ... ...... .............

)
\.
~,

Fig. 5. Distribution of Rote Faule facies, copper facies and lead-zinc facies in the Richelsdorf area

In the Richelsdorf area a Rote Faule zone has been discovered in a number of
widespread drill holes. At the northern margin of the Rote Faule zone, copper
and silver is concentrated. In several drill holes the mineralization was found to
be of commercially interesting grade. The high-grade copper-silver ore zone
follows the Rote Faule margin in a sinous manner followed by a lead-zinc zone
(Fig. 5).
Several copper minerals are formed depending on the abundance of copper.
These are chalcocite, bornite, covellite and digenite, indicating a decreasing CulS
ratio in a direction away from the Rote Faule facies towards the lead-zinc facies
(Fig. 6).
The economic ore zone is composed of chalcocite, bornite, covellite and
digenite, whereas bornite and chalcopyrite are also found in the lead-zinc zone.
The thickness of the mineralization is less than 2 m and crosscuts the strata at a
low angle. This is the result of the downdipping of the redox boundary at a
shallow angle to the north.
The relationship between paleoenvironment and metal distribution is demon-
strated by the spatial distribution of the sandbars in the Kupferschiefer sea and
the position of the Rote Faule facies . Copper is concentrated in the slightly
reduced environment of the interfingering facies zone between the sandbars and
the open marine Kupferschiefer facies (Fig. 7).
578 F.-P. Schmidt et aI.

•
•
o N 111'""
=
•

",~~ -:
"
..:-: I
••
'1 •
I

"-.•:-::':::::: ::: :: '~.~. ~~ :: :A:

:. :. :. :. :. :. :. :. :. :. :. :. : . ~.-;:
~ - --'1.1
:::: ::F.:e -.h.ydr.o.xides :::: :'r' ~-~
::::
. . ::. ond·.·
. . .hemQt
.. . . . i. t~
. .::::::
.... :: ~\
'. ' . ' . ' . '. '. '. '. '. '. ' . '. '. '. '. '. ' .. ~,
abundant Cu-minerals
cc chalcosit"
bn bornite
cp chalcopyrite

Fig. 6. Distribution of iron-hydroxides, hematite and copper minerals in the Richelsdorf area

The metal distribution is demonstrated in a south-north cross-section by drill

holes Ro 16, Ro 15, Ro 18 and Ro 21 (Fig. 8).
The paleotopography is governed by prominent sandbars of the Weiss-
liegend. On the top of the sandbars the Kupferschiefer is considerably reduced as
shown by drill hole Ro 16. Here the thickness of the Kupferschiefer horizon is
0.06 m. The sandbar corresponds with the Rote Fllule zone and in the drill hole
only 0.01 % eu with 5 g t -1 Ag over 2-m-mining height is present. Rote Fllule
occurs in the Zechsteinkalk, the Kupferschiefer and the Weissliegend. In the
copper facies close to the Rote Fllule zone, the copper content increases rapidly.
In drill hole Ro 15 it is 1.3 % eu with 10 g t -1 Ag over 2-m-mining height.
The main copper mineral is chalcocite and the thickness of the Kupferschiefer
is 0.62 m.
Drill hole Ro 15 represents the marginal position between the oxidized Rote
Fllule facies and the reduced copper facies where drill hole Ro 16 is located.
In drill hole Ro 18 the copper content is of economic grade containing 2.8%
eu plus 26 g t -1 Ag over 2-m-mining height. The thickness of the Kupferschiefer
is 1.9 m and contains a high portion of sand as a result of the vicinity to the
sandbar. Abundant copper minerals are chalcocite, bornite, covellite and
digenite which show various replacement features.
Results of Recent Exploration for Copper-Silver Deposits in the Kupferschiefer 579

•
•
o N
=
•
---1J in Ql,!en mori!l..!t-
fac ies

•
•
•

Fig. 7. The paleomicrorelief and facies types of the Kupferschiefer in the Richelsdorf area

The lead-zinc content increases slowly towards the hanging wall, where it is
concentrated within the Zechsteinkalk. Drill hole Ro 21 represents the lead-zinc
facies with a predominance of lead and zinc already within the Kupferschiefer.
Thus, only 0.30,70 eu with 1 g t -1 Ag over 2-m-mining height were found. The
Kupferschiefer is 2 m in thickness, copper minerals are tennantite, bornite and
chalcopyrite. Galena and sphalerite are the most abundant ore minerals.
Generally, the lead-zinc zone is formed in a more strongly reduced environment
than the copper zone.
The same geochemical transition from oxidized to strongly reduced environ-
ment is represented by the mineral zonation (Fig. 9) which starts with hematite
and iron-hydroxides in the Rote Faule zone, followed by the dominance of
chalcocite and bornite in the slightly reduced copper zone and galena and
sphalerite in the more strongly reduced lead-zinc zone.
Detailed investigations focussing on the relation between ore grade, metal dis-
tribution and mineral association with respect to the distribution of the Rote
Faule (Schmidt, 1985) showed that only through the interplay of primary synsed-
imentary features (e.g. organic carbon content and porosity of the host rock) and
secondary 'epi(dia)genetic' features (e.g. influence of the Rote Faule-forming
solutions) economic mineralization was formed.
 580 F.-P. Schmidt et al.

s Ro1618GR) 1Io15 18(jR ) Ro 18 IS . Joe) Ro 21 IS!.J•• )

N

Ca 1

T1
(u

S1

o I 2 3 4,.. o 1 2 3 4 5 6 1 6% 0 I I I " ,.

d5~,_____
0.1 ~ __~03 __0.4~~0.5. km
0.2
PROFIL A .. 9

Fig. 8. Change of metal distribution in drill holes Ro 16, Ro 15, Ro 18 and Ro 21 in relation to the
position of Rote Faule

Conclusions

As mentioned before a post-sedimentary mobilization and deposition of metals

in the host rocks is inferred which is in agreement with the genetic interpretations
of the Kupferschiefer deposits in East Germany and in Poland.
The Rote Faule facies is formed in the vicinity of sandbars which were shoals
in the Kupferschiefer sea and consequently controlled the paleotopography. On
the slope of the sandbars the Kupferschiefer contains a high amount of quartz-
sand interlayers which cause a high thickness, whereas on the exposed centres of
the sandbars the Kupferschiefer had not been deposited.
According to Rentzsch (1981) it must be emphasized that the Rote Faule is
not the common littoral facies of the Kupferschiefer. Where the Kupferschiefer
lies directly over the folded basement, the Rote Faule facies never occurs in
coastal areas of the sea.
The Rotliegend red-bed sediments underlying the Kupferschiefer seem to be
favourable source rocks of the metals concentrated in the Kupferschiefer (Lurje
1977). The metals were probably leached by chloride brines percolating through
the Rotliegend sediments.
 Results of Recent Exploration for Copper-Silver Deposits in the Kupferschiefer 581

s Ro 16IBGRI !lo IS (BCi RI Ro 1 8 1~' Jo.1 Ro 211~' Joe l

I
N
I I
I

Cal

T1

5 1

0,2 0.3 04 0,5 km

Fig. 9. Change of mineral association in relation to the position of Rote Fliule (see also Fig. 6)

As shown by Rose (1976) the metals could have been transported as chlorides
like CuCli , CuCl ~ - , PbCli or PbCl~ -: The migration of the chloride brines was
probably caused by increased subsidence in the centres of the sedimentary basins
which gave rise to the accumulation of huge piles of sediments. The pressure of
the sedimentary overburden moved the solutions enriched in metals towards the
marginal parts of the basins. In the sandbar areas, where the Kupferschiefer is
not present, or replaced by permeable sand-rich interlayers, the solutions spread
out. On the outer flanks of the sandbars, where the Kupferschiefer was formed
as an impermeable and reducing black shale, the metals were trapped and
concentrated (Fig. 10).
The geological model used for exploration of Kupferschiefer-type copper-
silver deposits was confirmed. Studies of controlling parameters, such as paleo-
geography, metal zonation, mineral zonation and the occurrence of Rote Faule
are important exploration tools on Kupferschiefer-hosted copper-silver min-
eralization.
Based on this tools, several holes drilled in the Spessart-RhOn and Richelsdorf
area have encountered economic ore grades.
582 F.-P. Schmidt et al. : Results of Recent Exploration for Copper-Silver Deposits

MINERALISATiON MOOEL OF
KUPFERSCHIEFER TYPE DEPOSITS W COPPER SILVER ORE

•
ct!1fIP LEAD ZINC ORE

ROTE FAULE

FLOW DIREC TI ON OF METAL

./ ENRICHED SOLU TlONS

/\ /\ /\ /\ /\ /\ /\ /\ /\ /\ /\
/\/\/\/\/\/\/\/\/\ /\ /\ /\ /\ /\ /\ /\ /\ /\ 1\ 1\
/\/\/\/\/\/\/\/\/\ /\ /\ /\ /I 1\ /\ /\

Fig. 10. Model of the formation of Rote Fiiule controlled and epi(dia)genetically-formed Kupfer-
schiefer-type deposits

Acknowledgements. The authors wish to thank Mr. T. N. Walthier, Vice President of Exploration,
St. Joe Minerals Corp., USA, for his permission to publish the first results of the ongoing exploration
campaign in West Germany. Our thanks are also due to Prof. F. Bender, Bundesanstalt fiir Geowis-
senschaften und Rohstoffe (BGR), Hannover, for making data available from drill holes Ro 15 and
Ro16.

References
Bundesanstalt fiir Geowissenschaften und Rohstoffe (BGR) (1983) Bundesbohrprogramm, Teil-
projekt Kupferschiefer. Forschungsvorhaben NTS 30043 Abschlu13bericht, Hannover
Ekiert F (1960) Neue Anschauungen iiber die Herkunft des in den Sedimenten des unteren Zechsteins
auftretenden Kupfers. Freiberger Forschungsh C 79:120-201
Friedrich G, Diedel R, Schmidt F-P, Schumacher C (1984) Untersuchungen an Cu - As-Sulfiden und
Arseniden des basalen Zechsteins der Gebiete Spessart/RhOn und Richelsdorf. Fortschr Miner 62,
Beih 1:63 - 65
Lurje AM (1977) Zur Herkunft des Kupfers in den Basisschichten des Zechsteins und der Kasanstufe.
Zeitschr Angew Geol 23:270 - 274
Rentzsch J (1974) The Kupferschiefer in comparison with the deposits of the Zambian copper belt.
Centrenaire de la societe geologique de Belgique gisements stratiformes et provinces de cupriferes.
Liege, pp 395 -418
Rentzsch J (1981) Mineralogical-geochemical prospection methods in the Central-European Copper
Belt. Erzmetall 34, 9:492 - 495
Rose A W (1976) The effect of cuprous chloride complexes in the origin of red-bed and related depos-
its. Econ GeoI71:1036-1048
Rydzewski A (1978) Oxidated facies of the copper-bearing Zechstein shales in the Fore-Sudetic
monocline. Prz Geol 26: 102 - 108
Schmidt F-P (1985) "Rote Fiiule" kontrollierte Buntmetallvererzungen im Kupferschiefer Osthessens,
BRD. Diss TH Aachen
Schumacher C, Kaidies E, Schmidt F-P (1984) Ergebnisse iiber den basalen Zechstein der Spessartl
Rhon-Schwelle. Z d g G 135/2
Subject Index

Abitibi Belt (Canada) 47,57 - K-feldspar 295

Accretionary tectonics (Alaska) 423,439 - K-metasomatism 274,295
Activity-activity diagrams 202 - metamorphism 435 f
Adiabatic ascent 43, 53 f - metasomatic wnation 266f
Mrica - pyritization, metasomatic 354
- Kambove W, (Shaba, Zaire) 390 - quartz-sericite 295
- - Cu-Co, Roan 499 (see there) - secondary 530f,536
- Kamoto 390 - silicification 506
- Shaba 390,444,454-457 Amphibole (Baltic shield) 151,153
- Zaire 390 Anhydrite 419
- Zambia 444-454 Anorthosite (Duluth, Minnesota) 8, 10,
- - Copper belt 479, 483 12-16,22
Mro-Arabian craton 332 - chromite rich -gabbro 120
Agnew (Canada) 45,48 f Antraxolite 532f
Alaska Arsenic 415
- Binocular prospect 425 Arsenide (W-Germany) 575
- Chitistone (see there) 431-433 Assimilation 31,43f,55-58
- Copper Attapulgit
- - mineralization 433-435 - argillite 455 f
- - prospects 425 Australia
- Geological setting 423 - Adelaide geosyncline-Stuart
- Lithostratigraphic relations 426, 428 Shelf 540-550
- Mc Carthy Quadrangle 423-425 - Bangemall Basin 540-542, 555
- - faults, fault trends 427, 439 - CSA deposit 541
- - geological setting 423 - Gecko deposit 541
- - location 423 f - Great Cobar deposit 541
- - Mountain Grill prospect 425 - Kambalda (see there) 63-82
- Nelson thrust fault 425, 439 - Kanmantoo deposit 541, 543 f
- Nikolai greenstone (see there) 428-431 - Mammoth deposit 551 f
- Peavine Prospect 425 - Mt. Gunson
- Radovan Prospect 425 - - Cattle Grid deposit 540f, 543,
- structual relations 427,439 546-548,556
- Tectonostratigraphic terranes 427 - - Pandurra Formation 545-549
- Wrangell lava 426 - - Tapley Hill Formation 549-550
- Wrangell terrane 426 f - Mt. Isa (see there)
- Zebra beds 433 - Nifty deposit 541,555
Aljustrel (Spain) 315 - Olympic Dam (Roxby Downs) de-
Alteration 283-285 posit 540f, 543-545, 556
- albitization 506 - Peko deposit 541-555
- chloritization 457 - sediment-hosted copper 540-548
- diagenetic 444,454,458 - Tennant Creek Inlier 540-542,555
- feldspathization 454 - Warrego deposit 541,555
- hydrothermal 199f, 206,444,454 Aznalcollar 315
584 Subject Index

Baltic Shield 151,153, 171 - - Igarka (USSR) 525, 532, 534, 536
- greenstone belt 171 - - Sibiria 517-519
- Kola Peninsula (see there) 152 Breccia
- Komatiites (see there) 171-173 - explosive 298
- Kuhmo 152 - fluid explosive 297
- layered intrusion 154 Brine
- minerals - anaerobic pools 359
- - amphibole 159 - bottom 369
- - clinopyroxene 158 - chloride 580, 581
- - mica 159 - evaporitic 458
- - olivine 156 - hydrothermal 355
- Nickel (bearing provinces) 176 f - Mg- and Ca-rich 457
- Pechenga 151, 153 - sediment interface 362 f
- picrite 164 Buoyant plume 360f
- rare earth elements 163 - hydrodynamics 359
- sills 154 - hydrothermal stream 361
- ultramafic rocks 155
Barite
- Hakuroko District, Japan 370,375 Calc-alkalic rocks 198
- - chemistry 375, 378 Canada
- - fluid inclusion 380 f - Abitibi Belt 47, 57
- - isotopic composition 379,384 - Agnew 45,48f
- - mineralogy 372 f - Circum Superior Belt 47
- - model offormation 384 - Dumont 45
- - Sr-content 378 f - Redstone, NW Territories 390
- - - in solution 382 - Sudbury (see there)
- - volcanogenic massive sulfides 370f Carbon, organic 579
- Hungary 335 f Caucasus
Basement - alpine
- Germany 580 - - base metals 332, 334
- S-America, Precambrian 24lf - - tectono-magmatic cycle 326
Basin - Caucasian segment 325
- Bangemall (Australia) 540-548,555 - Miskhan Zangerur (S) 332
- Range Disturbance 218 - Sevan-Akerin zone 332f
- Range Province 221 f, 225, 228 - Somkheti-Kafan Zone 332
- Rotliegend rift b. 474 - volcano-sedimentary rocks 325
Batholite 226,229-231 - volcano-plutonic association 333 f
- buoyancy, gravitational instability 216, Cenozoic 218 f, 221
226f,231 - late 221
- Chitina Valley 439 - meso 240f
- coding rate 220-222,225 f, 229, 231 - middle Tertiary 218 f, 220
- composition 221 - Neogene 229
- cupolas 216,225-227,229,231 Chalcocite 415,559 ff
- density 216,226f - USSR 495-498, 505, 577 ff
- depth 221 f, 231 - - Igarka 535
- dome in dome structures 216,225-227, - - Sibiria 517-519
229-231 Chalcopyrite 559 ff
- models 216,226-228 - W-Germany 575 f
- Perm 226 - Minnesota, Duluth 8,10,18-21
- Premineralization 221,223,225-232 - USSR 495-498, 505
- Sonora-Sinola b. 220-225,231 - - Igarka 525,532,534,536
- strato-volcano coupling 216,219, - - Sibiria 517-519
227-237 Chemical interaction 26
Biochemical barriers 535 f China 180
Bornite 415,559f - Ailaoshan 190
- W-Germany 575 - Chromium (podiform Chromite) 181
- USSR 495-498, 505 - Dal 189
Subject Index 585

- linchuan 189 - porphyry 197,216,219, 225f, 228, 230,

- Luobusa 181 235-250,251-255,257-259,280,286,
- Nickel 187 288f,335
- stratiform 186 - pre-minerals
Chitina Valley Batholith (Alaska) 439 - - epigenetic 497
Chitistone Limestone (Alaska) 431-433 - - silver content 433-435, 572 f
- chemical composition 432 - sandstone deposit 492f
- Copper mineralization 434 - solubility 416
- petrography 430,432f - source 522
- stratigraphic relations 426, 428 - stratiform
Chloride complexes (Cu) 416 - - diagenesis 390, 392-394
Chlorine activity 295 - - metal wning 390,391-393
Chlorite 451 f, 455-457 - - pene-exhalative model 392, 395 f
Chromite - - pyrite, metasomatic 335f
- anorthosite gabbro 120 - - red beds 390, 392
- China 181 - - sandstone-type 477,504
Circum Superior Belt 47 - - sediment hosted 390-397
Clays, (Mg) 448-455, 456 - - shale deposit 477
Clear-Porcupine prospect 425 - - slate type 504
Climax-type molybdenum 198 - - transport 390
Clinopyroxene (Baltic Shield) 158 - supergene enrichment 244
Coates Lake stratiform Copper deposit 412, - USSR, Sibira 235-250
420 Cordillera, south 217
Concentric structure 339f, 345-347 Country wall rock 216,222,225-227,
Conformable, disconformable 280, 289 229-231
Contamination 37 Covellite
Continental - Alaska 433 f
- crust 26 - W-Germany 577 ff
- uplift 217 Crescumulate 46f
Continuity equation 563 Cretaceous 217-219,221
Copper 572 f, 304 f Cross Province (South) 47
- adsorption on Fe-hydroxides 416 Crude Pyrite 323
- Mrica (belt) 444-454, 479, 483 Cryptic
- - Kambove, Mine group 400f - layering 117f
- - - Roan 398f - sulfide layering 116
- S-America 235-250 Cubanite
- banded ores 312 - Duluth, (Minnesota) 8, 10, 18-20
- bearing wne, USSR 492-503 Cu-Co
- chloride complexes 416 - Roan (Kambove) 398f
- complex compounds 212f Cu-Mo 291-302
- CulMo ratios, S-America 242-244 Cupreous stockwork 304
- CulS-ratio 577
- depletion from red-beds 417 f, 420 Damba-Silwane 45f,48
- epigenetic 497 Darcy's law 564
- epochs of accumulation 499f Diagenesis 362,412,418
- W-Germany 572ff - diagenetic fluid 550
- Harbor Formation 476, 483 - epi(dia)genetic 572,574,579
- Kupferschiefer 444, 454, 572 f - paragenesis 405 f, 407
- mineralization 433, 440 - sediments 363
- - controls 433 f, 439 Differentiation
- - Kennecott-type (Alaska) 433-435 - crystallization 116
- - model 437-440 - interchamber 115
- - stratabound 422f Diffusion 563
- - timing 439 Digenite 415
- nickel ones Ill, 124-134 - W-Germany 577f
- - sulfides (Duluth, Minnesota) 8, 10, 92, - USSR, Igarka 525, 532, 535
102, 109, Ill, 124-134, 155 Discharge wne 353
586 Subject Index

Dispersion, effect of 132 - metalliferous 550

Dissolution 559 - meteoric 199,204
- reaction rate 563 - phase of melt 120
Djurleite - pore 412,416,419
- USSR, Igarka 525, 532, 535 Fluorine activity 295
Duluth (Minnesota, USA) 8-10,16, 23f, 33 Flux 563
- Anorthosite 8,10,12-16,22 Forrestina 45-49, 55
- Babbit, Cu-Ni 35-37,39 Fractional crystallization
- Drillholes 9,12, 14f, 17 - flow through 46,49,51,56
- Dunka Road, Cu-Ni 33f
- faults 8, 12, 17,22 Gabbro
- geochemical trends 14 - chromite-rich anorthosite gabbro 120
- Partridge River troctolite 9 f, 23 f - upper series 122
- South Kawishiwi Intrusion 8-10, 13, Gai deposit 340, 344f, 347f
21-24 Galena
- Sulfide - W-Germany 579
- - bearing wne 8,12-15,22 - USSR, Igarki 532, 534
- - compositions 20 Geochemistry 96, 159
- - silicate intergrowth 19 - Isotopes (see there) 103,545,548-550,553
- whole rock, nickel content 14-16 - Major elements 96,443-500
Dumont (Quebec, Canada) 45 - Rare earth elements 101,457
Dunite - - Baltic Shield 163
- intrusive 45 - Trace elements 101
- komatiitic 43,45-49, 54-58 Germany
- Basement 580
Embayments 48f - Baumbacher High 576
Epigenetic - Fore Sudetic Monocline 572
- barrier 477 f, 488-490, 505 - Fulda 572
- exogenic-hypothesis 477 - Kupferschiefer 572 f
- ores 135, 141 - Lubin 572
Equilibrium/non- - Mansfeld 478,481, 484f
- activity diagram 560 f - Oder trough 48
- mineral association 530f, 535 f - Rhon 572
- modelling 562, 566 - Richelsdorf 481,484,486,572
- wning 569 - Saale trough 481, 486
Erosion 217,224 - Saar trough 481, 486
- rates 222, 224, 231 - Sangerhausen 481, 484 f
Evaporites - Spessart 572
- argillites, marls 444,447-449 - Sudeten (North) trough 481
- environment 444-448, 456, 458 - Werra 572
- meta-evaporites, identification 443-460 - Zechstein-Kalk 572
- sediments (USSR) 505 Gold
- subcycle (Poland) 464 - S-America 244, 248
- Spain 304
Facies changes Gossons 304
- symmetrical/asymmetrical 508 Granites, pegmatites
Fault 8,12,17,22,217,427 - two mica 217 f
- trend 427,439 Granitoid tpyes I, S 196f, 206
Fluid ~ Greenstone 44, 47, 57 f
- diagenetic 550 Greenstone belt 152, 112
- flow, computer modelling 203f,206 - Baltic Shield 171
- hydrothermal 199-206,545,553, 555f - Basalt composition 175, 177
- - salinity 198,200-202,205 - Komatiite (see there) 171-173
- inclusion 200,203, 205f, 210, 380, 457, - Mineralization 171,174f
545 - Nikolai (Alaska) 428-431
- magmatic 199,204 - Platform, phase gr. 57 f
- - convection 122 - Rift phase gr. 57 f
Subject Index 587

Gusljakovskoe deposit 340, 346-348 Iron

Gypsum 419 - formation 44, 47 f, 55, 57
- hydroxides 576
Hanawa deposit 340, 345, 347 Island arc type magmatism 280f, 283
Hematite 561 Isotopes 203 f, 206, 545, 548-550, 553
Hercynian Orogeny 306 - light element 470
Honeymoon Well 45-49 - radiometric ages 239
- - K-Ar 221-223,225, 230f
Hungary 280
- - lead 247
- Contact metasomatism 285
- - strontium (initial) 243, 246-248
- Darno megastructural zone 281
- Genetic aspects 280 - stable isotopes 25,40
- Island arc-type magmatism 280f,283 - - carbon 32
- magnetite deposit 285 - - hydrogen 32
- massive sulfide 287 - - oxygen 30 f, 34, 36
- morphogenetic aspects 289 - - sulfur 29f, 34, 38,56,415
- Ore grade contours 288 f
- porphyry J-M Reef 122
- - copper 280,286,288f Jurassic 211£
- - diorite intrusion 280, 282, 285 f, 288
- Priabonian, Upper Eocene 281 Kambalda (Australia) 45-50,55-59
- Recsk - Contact ore 67
- - andesite formation 281 - Hangingwall ore 67
- - mineralized complex 280 - highllow tenor ore 71
- - space-time model 285 - Interflow sediments 68
- Remobilization 280, 286 - Juan fault 77
- Replacement deposit 286 - Kambalda Dome
- Skarn zone 284-286 - Lunnan Shoot 75
- Stratovolcanic stage 281, 287 f - Norseman-Wiluna Belt 74
- Troughs 70
- Triassic host rock 283
- Windarra 67
- volcano-sedimentary ore 287
Hydration-dehydration reaction 435, - Yilgarn 63
437-439 Katininq 45 f, 48 f
Hydrodynamics of buoyant plums 359 Kennecott (Alaska)
Hydrofracting 200, 206 - formation 426
Hydrogen sulfide 504 - K. type Copper deposit 422, 442
Hydrothermal 435 - Modelofmineralization 437-440
- activity 281,283,286,288 - prospect 425
- alteration 199 f, 206, 545, 552 K-feldspar rich schist 295
- brine 355 Kinetic interaction 559 f
- discharge 359 Kola Peninsula 152
- fluid 545,553,555f - Deposits
- hypothesis 477 - - Allarechka 137, 139, 142, 145
- metasomatic association 209,330 - - Eastern Ore Cluster 138, 142, 145
- meteoric h. system 300 - - Flangovoye 138, 142, 145
- stratified sedimentary 352, 355, 358 - - Lovnoozero 139, 142, 145
- stream (Buoyant plume) 360f - - Nickel provinces 135-137, 146
- submarine exhalation 330 - - NKT 140-142, 145
- systems and vents 216,225,228-232 - - Sopcha 140-142, 145
- - Vostock 137,139,142,145
- vents, pumps in volcanoes 216
- Ores
- - chemical composition 142 146-148
Iberian Pyrite Belt 304 - - mineral composition 143--'146
Illite 451 f - - types 131£,140, 147f
Indicator mineral 339 f - pyrite-pyrrhotite mineralization 137,149
- variables 340 - Runnijoki 137
Infiltration 563 - Severnoe 137
Injection structure 329 - Sueinlagash 142, 145, 147
588 Subject Index

Komatiite 43-58,171-173 Melting

- cumulate 43-48,50, 53 f, 56-58 - partial 43, 53, 58
- komatiitic dunite 43,45-49, 54-58 - sequential 43, 58
Kupferschiefer 444, 454, 572 f Metallogenetic provinces 217-219,304f
- deposition 580 Metamorphic
- exploration 572, 581 - conditions 446
- facies - Cu bearing shales 444-454
- - copper 576f Metamorphism 435f
- - lead 576f - stages 135, 143-146
- - littoral 580 Metasomatism
- - open marine 577 - contact, magnetite 285
- - reduced 576 - potassium 274,295
- - Rote Faule 576f - zonation 209,266f
- - zinc 576f Mexico 216-220
- hematite 576 - Bancarit formation 221,224
- migration 581 - Cabo corrientes 218
- mobilization 580 - Chiapas 217
Kuroko deposit 346,348, 350f, 354 - Consejo de Recursos Minerales 217,232
- parallel ranges 222
- Sierra Madre occidental 217,221 f, 225
La Zarza 315 - Sonora-Sinaloa 216-221,225,231
Laminar flow regime 52 Middle American trench 218
Langmuir 45-47 Mines Group (Shaba) 455
Laramide orogeny 217,219-221, 224f Mo 210-212
Lava conduit 43-46, 48, 54, 56, 58 Moho boundary 129 f, 133
Leaching zone 505 Molasses, red colored, USSR
Lead 415 Mongolia
- West Germany 576f - Erdenetyn-Ovoo ore field 271
Limestone - explosion breccia 275
- Chitistone 431-433
- high grade ore 277
- marine 221, 225
- Okhotsk lineament 272
Lithium 448-455
- Ore concentrating structure 273 f
Lousal (portugal) 315
Morpho-structural modeling 263
Mt. Isa (Australia)
Magma - deposit 540f,551-556
- chamber 43-46,51, 54f, 58, 293f - Inlier 540-542,551-555
- composition 217, 226f Mt. Keith-Betheno 45
- conduit 52f
- density 132, 227 Neves-Corvo (Portugal) 319
- mafic, ultramafic 125 New Guinea 229
- - contamination 27f Ni-Cu 44-58
- Ni-bearing magmatism 128 - Baltic Shield 155
- ore bearing m. 120f, 129 Nickel
- ore system 292, 294 - China 187
- parental 163 - Kola Peninsula 135-137,146
- primary 165 Nickel-bearing provinces 176f
Magmatic - classification 176
- fluid convection 122 - ore potential 177
- group 124, 126f, 135f, 140 - magmatism 128
- origin 82 - nickel-free rift 129
- series 124, 126f - structures 124
Malachite 418 - tectonic environment 177
Manganese ores 309 Nikolaevskoe deposit 340, 346, 348
Mantle 43, 53, 58 Nikolai Greenstone (Alaska) 428,431
- high temperature melts 128 - Alteration-metamorphism 435 f
- melting 128, 134 - Chemical composition 431,436
Margin, convergent 217 - Copper mineralization 433 f
Subject Index 589

- Petrography 429-431 Picrite (Baltic Shield) 164

- Stratigraphic relations 426,428 Plagioclase (olivine) cummulates
Norilsk (USSR) 114f - Duluth, Minnesota 11 f
Norseman Wiluna Belt 46f,57 Plate tectonics 197 f
North America 8, 23 - Rotliegend rift basin 474
- Duluth Complex (see there) 33 Platinum 122
- Minnesota 8 f, 23 f Pluton 216f, 225, 228-231
- Nevada 217 - rhythmically layered lll, 115
- White Pine, Michigan 390,479,483, 559 Poland
Nucleation 564 - Evaporites 464
- Fore-sudetic monocline 462
Oder trough (Germany) 481 - Lagoon
Oil forming components 506 - - barrier 467
Oktjabrskoe deposit 340, 345, 348 - - relief 465
Olivine 44,46,49-51,70 - Paleogeography 463
- Baltic shield 156 - Provenance of metals 474
- intra telluric 44, 50, 53 - Rudna Lagoon 464
Ore - Zechstein Sea ingression 462
- association 291 f Porosity 579
- bearing bodies 124 Porphyry
- bearing formations 124-133,251-255 - bodies 274 f, 293
- bearing magmas 129 - Copper 197,216,219, 225f, 228, 230,
- bearing sags 129 235-250,251-255,257-259,335
- belts 78 - - Hungary 280, 286, 288 f
- deposits - Copper-Molybdenum 251,254-257
- - enrichment and preservation 217 - dikes 276f
- - factors of ore deposits 213 f - diorite intrusion 280, 282
- - formation 135, 149 - - Hungary 285 f, 288
- - fracturing 199 f - Molybdenum 197,251,254-259
- - occurrence 217-219 - Molybdenum-Copper 251,254-258
- - relation to complex (Sudbury) 104 - style 251, 258f
- - relation to magma composition Precambrium 221
- forming solution (parameter) 210 Pyrite 559f
- host rock system 435 - Kola Peninsula 137,148
- magmatic system 229,294 - massive 304
- recrystallization 331 - metasomatic 354
- stockwork 262 - Poland 466
- tenor, high grade ore 57,71,261,277 - polymetallic 309
- zonation 268 - Spain 309
Organic matter 506f, 515, 519 - USSR
Osennee deposit 340f,347 - - Igarka 532,535
- - Sibiria 518 f
Paleo Pyrrhotite
- environment 577 - Duluth, Minnesota 8, 10, 18-21
- geography 572,581 - Kola Peninsula 137, 148
- topography 578, 580
Paleozoic 240
Parandovskoe deposit 339f, 347 RAT (Shaba) 454-457
Pechenga-Imandra Varzuga 112 Reaction and Transport, feedback 562
Pegmatoids 12, 14, 16f Red-beds
Pentlandite (Duluth, Minnesota) 8, 10, - Copper deposit 412
18-20 - - stratiform 390, 392, 477
Permian Redox
- Rotliegend 572 - boundary 575
- - Saxon 575 - condition 576
- Zechstein 462, 572 f - reaction 364
"Phyllitic overprint" 199 Redstone area, Canada 412,419
590 Subject Index

Replacement 576 - model 364f

- feature 578 - neutralization 366
Reynolds number - pore solution 362, 367
Rhythmic structure 339f, 342f, 347 - physical parameter, ore forming 210
Rhythmo-stratigraphic complexes 327 - superhigh concentrated 363
Rifting 8,10,16,22 - water dissipative state 363
Rio Tinto 313 Sotiel Coronada 315
Roan (Shaba, Zaire) 479,483,486 South America
- CuoCo, Roan 399 - Argentinia 239 f
- diagenesis, paragenesis 405-407 - Central Andes 237
- Kambove, Mine group 400f - Chile 239f
- model - crustal
- - for deposition 408 - - setting 241
- - for origin 409-411 - - thickness 242
- sulfide distribution 402-404 - gold 244, 248
Rock permeability 228 f - metal sources 235 f
Rotliegend 572 - metallogenesis 245
- metallogenetic subbelts 238-241
Sabkha 419 - Paleocoic 240
Sandoar 577, 580 f - Peru-Chile trench 239,241
Sapropelic model 472 - porphyry copper 235-250
Scaling 565 - Precambrian basement 241 f
Scotia 45 f, 48 f - subduction 236,247
Sediment Sphalerite (Germany) 579
- assimilation 83 Stockwork 312
- brine interface 362 Stokes, flow regime 51 f
- chemistry 471 Stratabound 308
- diagenesis 363 - massive sulfide deposits 308
- d~tribution 467 Stromeyerite (Germany) 575
- evaporite 505 Subaerial
- hosted copper 390-397,540-558,559 - cementation 473
- interflow 68 - exposed shales 466
- pelagic 246 f Subduction 217f
- sedimentary hypothesis 477 - Paleo s. zone 217
- sequence 470 Sudbury (Canada)
- sulfide 43 f, 47 f, 55 f - Geochemistry %
- synsedimentary pyrite 359 - - Isotopes 103
Sedimentation 217,220 - - major elements 96
- evolution 469 - - rare elements 101
- kinetics 362 - - trace elements 101
Settling, velocity 51 f - Geological setting 92f
Shakanai deposit 340, 343 f, 346 f - Igneous Complex 91
Shale - - petrology 94
- cupriferous 514f - Main mass 95
- Nonesuch Shale 479 - Model offormation 108
- ore shale (Zambia) 444-454 - Nickel-Copper 92, 102, 109
- potassic shale 447-454 - On aping formation 94
Shangai 45 f, 48 - SIC
Shinokawa deposit 340f,347 - - distinguishing features 107
Siberian Platform 513-523 - - indudies 96
Silver - - sublayer 95
- Alaska 433-435 - - - relation between inclusion s.1. and
- Germany 572 f Main Mass 105
- Spain 304 Sudeten trough (North) 481
Six Mile Well 45-47 Sulfides 575f
Solution - accumulation 132
- experimental solubility data 211 f - bearing
Subject Index 591

- - magma series 124, 128 Syngenetic

- - zone 8,12-15,22 - barriers 477 f, 490, 504, 507
- composition 20 - ores 135,141,309
- Cu-Fe s. zoning 566f Synsedimentary pyrite formation 309, 359
- Cu-Ni s. deposit 8, 10,44-58
- dissiminated s. 14,43,45, 49f, 54-56, Taphrogenesis 218 f, 225
58 Temperature
- distribution 402-404 - barrier 300
- droplets 43,51-54 - blocking 221-223,225
- free zone, 1-15,22 - gradient 226, 228 f, 294
- globules 117 - homogenization 297
- hydrogen s. barrier 504 Tennantite (Germany) 575
- - allochthonous 477,488 Tharsis 313
- - autochthonous 477,488f Thermal erosion 83
- immiscible liquid 115, 119 Thermo-osmotic transfer 300
- included 18 f TIshinskoe deposit 340,342-344, 346-348
- interstitial 17 if Troctolites (Duluth, Minnesota) 8-17
- magma 112 Trojan 45, 48
- magmatic 43 f, 55 . Turbulent flow regime 52
- massive sulfide 43,45,48, 54 f,
112, 338 Uchali deposit 340, 342f, 347
- - Altaity 340, 342, 345 Ultramafic
- - base metal 358 - lava 115
- - Cyprus type 339-341 - mafics and ultramafics (magma) 125,
- - deposit 350, 352 135,137-140,144,149
- - Hungary 287 - - - massif 135,137-140,144,149
- - Karelian type 339 f - rocks (Baltic Shield) 155
- - Kuroko 340, 342 f, 345 Upper gabbro series 122
- - pyritic 358 Uranium (Canada) 415,417f
- - Ural type 339 f, 342,345 USSR
- - volcanogenic, barite 370 - Alrnalyk 262
- metasomatic ore 358 - Bashkirian 481
- mono solid solution 115 - Copper bearing zone 492-503
- Ni-sulfide - - geosyncline 398, 493
- - complex 128 - - orogenic 493, 494-497
- - bearing formation 124f, 133 - - platform 493-497 f
- saturation 50, 53-56, 58 - Dzhezkazgan deposit 488 f, 496, 505-511
- segregation 43,53-55,58,91,105,107, - Igarka
109 - - Chernorechenskaya ore occur-
- separation 43 f, 54, 56, 58 rence 534
- silicate - - Copper area 493,524, 526f, 535,538
- - interaction 119 f - - Graviika ore deposit 530-533, 537
- - intergrowth 19 - Kizil Dere deposit 498
- solubility 28 f - Udokan deposit 498,516
- texture 18
- transport 128 Vanadium (Canada) 415
- vein 20 Variscian 572
Sulfur 304 volcanic
- biogenic 496 - arcs 217-219,280, 283f
- fugacity 116 - belt
- isotopes 29f, 34, 38,43 f, 55-58,415, - - North, Central Mongolia 272
495-497 - - South Mongolia 273
- solubility 53 f - central 43,46, 55
- source 356 f - distal 43, 46, 55
Sulphate - peridotites 45
- bacterial reduction 548-550 - proximal 43, 53, 58
- biogenic reduction 355 f, 507 - rocks 218-221, 224f
592 Subject Index

volcanic Widgiemooltha 46
- sub volcanic 43-46, 55, 57 Windarra 45, 48, 55
- volcano-sedimentary 287, 457
Volcanism
- acid 308 Zimbabwe 47
- eruptive history 229 Zhairem deposit 350f,354f
- explosive 308 Zhilandinskaya group deposit 489
- migration 217-219 Zinc (Germany) 576f
Volcano 216f,225-231 - (Spain) 304f
- andesitic 251-256 Zonation 576,581
- as hydrothermal vents, pumps 216, - eu-Fe 566f
228-231 - facial 352
- basaltic 251-256 - hypogen 363
- basic 308 - lateral z. 350, 354, 356
- epi-eugeosynclinal 251,254-256,259 - metal 243
- epi-miogeosynclinal 251-256,259 - metallogenetic z. 198,251,254-256,259
- eugeosynclinal 251-256 - metasomatic z. 266, 518 f
- metallogenic zoning 251,254-256,259 - mineral z. 559ff
- period of emplacement 254 f - - metallic minerals 525,527,532,
- terrigenous arc 327 534-536,537
- volcano-plutonic belt (VPB) 251-259, - mineralogical, geochemical 338-348
292,333 f - monoascending 363
- non-equilibrium 569
Wannaway 45,48 f - ore z. 268
WeiBliegend 572 - vertical z. 350, 352, 356