See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225722412

A Model of the Dump Leaching Process that Incorporates Oxygen Balance,

Heat Balance, and Air Convection

Article  in  Metallurgical and Materials Transactions B · December 1975

DOI: 10.1007/BF02913857

CITATIONS READS

121 414

2 authors:

Lawrence Cathles John A. Apps

Cornell University University of California, Berkeley
176 PUBLICATIONS   6,475 CITATIONS    107 PUBLICATIONS   3,815 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Caves & Gypsum Formation View project

Glacial Isostatic Adjustment View project

All content following this page was uploaded by John A. Apps on 10 January 2014.

The user has requested enhancement of the downloaded file.

 .,.,,f..,...-.....«^..^..«^«^.a^yAi.-f..-^-:;iMa'H^^

^mmmmi m mm
A Model of the Dump Leaching Process
that Incorporates O x y g e n Balance,
SUBJ
Heat Balance, a n d Air Convection MNG
MDL

L. M. CATHLES .A.ND J. A. APPS

.\ one dimensional, nonsteady-state model of the copper waste dump leaching proce.ss h a s
been developed v.Oiich incorporates both c h e m i s t r y and physics. The model is based upon
three equations r e l a t i n g oxygen balance, heat balance, and a i r convection. It a s s u m e s that
the dump i s composed of an aggregate of rock p a r t i c l e s containing nonsulfide copper m i n -
e r a l s and the sulfides, chalcopjTite and pi'rite. Leaching o c c u r s through chemical and dif-
fusion controlled p r o c e s s e s in v.-hich p^Tite and chalcop\nnte a r e oxidized by f e r r i c ions in
tlie lixiviant. Oxygen, the p r i m a r y o.xidant, is transported into the dump by means of a i r
convection and o.xidizes ferrous ion through bacterial catalysis. The heat generated by the
o.-^idation of the sulfides promotes a i r convection. The model was used to simulate ihe
leaching of copper from a small test dump, and e.-'xellent agreement with field m e a s u r e -
m.ents -A'as obtained. The model p r e d i c t s that tlie most important v a r i a b l e s affecting cop-
p e r r e c o v e r y from the t e s t dump a r e dump height, pyrite concentration, copper gi-ace. and
lixiviant application r a t e .

THE-HE leaching of low-grade c o p p e r - b e a r i n g waste has heat generation, fluid flow and ether t r a n s p o r t phe-
been p r a c t i c e d eiO-ier by accident or through design nomena relating to the leacliing p r o c e s s m-jst also be
for s e v e r a l hundred y e a r s . D-uring the last fifty yeai-s, considered. A leaching system cannot be considered
i n c r e a s i n g attention lias been paid to the s y s t e m a t i c in a steady state, because all factors involved in the
leaching of low-grade v.'aste r e s u l t i n g from the open leaching p r o c e s s change p r o g r e s s i v e l y a s a function
pit mining of pcrphyry copper deposits in the western of t i m e .
United States. By novv, this activity i s yielding an i m - In this paper we have developed a one-dimensional
portant s e c o n d a r y s o u r c e of domestic copper,^ Indeed, model of the nonsteady-state dump leaching s y s t e m .
s o m e mining operations have been planned and a r e We have applied this model to a s m a l l test dump con-
operating exclusively from the production of copper structed and leached a t the Utah Copper Division of
obtained from leaching. Many of these operations a r e Kennecott Copper Corporation. To our knowledge only
e.KpIoiting- o.vide copper deposit.'; where copper is r e a d - two other attempts have been made to integi'ate the
ily leached by the application of dilute sulfuric acid. d i v e r s e aspects of dump leaching into a coherent a l l -
L o w - g r a d e waste discarded a s a resuli of open pit embracing model.^'^ Wliile we do not feel that the
mining of porph.yry copper deposits i s dumped in gul- model presented in this paper is the final answer lo
l i e s surrounding the deposit. The disposal site is d e - a clarification of the dum.p leaching p r o c e s s , v.e b e -
t e r m i n e d p r i m a r i l y by the convenience of the site lo lieve that it forms a basis upon which subsequent r e -
the mining operation, and is not usually based on con- s e a r c h in thts a r e a might be coordinated.
s i d e r a t i o n s n e c e s s a r y for optimum leacliing. In the
w e s t e r n United States s e v e r a l Vjlilion tons of v.aste has
accumulated in this m.aru-ier.
During the last decade, many people have become A MODEL O F THE DUMP LEACHING
conscious of the fact that this e n o r m o u s r e s o u r c e of PROCESS
copper is not being c-rploited effectively because in- Initial Assumptions
sufficient attention is being paid to those factors which
could lead to improved design ai-id layout of v/aste Sulfides m u s t be oxidized before their metal values
d u m p s . It i s Ijelieved that if the leaching p r o c e s s may be put into solution. Tlie conceptual basis of the
w e r e completely understood, then it would be possible irodel of dump leaching'presented h e r s is simply that
to design and leach copper from waste dumps in a far the exothermic sulfide o.^^ndation r e a c t i o n s generate
m o r e efficient m a n n e r than is c u r r e n t l y being p r a c - heat and consume o.xygen from the a i r , and by s o do-
ticed. Tlie problem is a la.rge one. Not only must the ing d r i v e a i r convection through the dump. Tnis air
c h e m i s t r y of leaching be understood, including both convection is the only significant s o u r c e of oxidant to
kinetic and thermodynamic a s p e c t s , b u l the effect of tlie dump.
-k system is envisioned in which a c o u n t e r c u r r e n t
L. M. CATHLES is a senior geophysici.st ,-ir llic Lejgemom Labora-
interlocking flow of a i r and v/ater p a s s e s through an
lory. Kennecoit Coppjr Corpora-iion, Lexington, .Massachus-eus. J. A.
APPS, rormcrly a scientist at the .Mcisl .Mining Diviiior. Research On- aggregate of rock fragments, a s shown in Fig. 1.
ler, Kennecott Copper Corporatioi-,. Sail Lake City. L"wh. ii now staff Tlie oxygen leaves the gas phase within ihe dump
scientist, Energy nno Environment Division, Lawrence Berkeley !,„-- by dissolving in the liquid phase where it o.xidizes
bonil'orj'. Universiiy of Ca!ifv-.-.T!i3, Berkeley, Calif&r.-iia. f e r r o u s to f e r r i c iron through the agency of b-acteria.
Maniiscripi subniiited FebrJsry S, 1975. The f e r r i c iron diffuses into the o r e fragments and

M E T A L L U R G I C A L T R A N . S A C T I O N S 1) VOLUME 6B, DECfiMBER 1975-617

oxidizes the sulfide minerals:* Acid, F e " and heat
* At 1/5 atTnospheic ?Q and the leirperaturcs involved in dump kachijig,
oxygen is not ver>' soluble in water (<8.6 X 10"' si I). Typical ferric iioa con-
ceniraiionsin leach dumps run ~1 g/1. These relative conccniraiionsensure Fe"*
Ml! be the oxidizing agent in the diffusion controlled processes envisioned above.
are produced along with Cu**.
For the purposes of our model, we have assumed
the formation of a leached rim which is separated
from the unreacted core by a sharp boundary, as
shown in Fig. 2. As the leached rim grows the rate
of leaching drops because of longer diffusion paths
and a shrinking reaction zone. E\'idence for a shrink- UO'JiD PHASE
SULFIDE BLEBS
ing core has been supported observationally br Braun,
Lewis and Wadsworth'' and by Madsen, Wadsworth and >
Groves.' Theoretical arguments also support the ex-
istence of a sharp boundary for the conditions of our Fig. 2—Idealization of the leaching of a single waste p a r t i c ' e .
model (see Eq. [11]). For those reasons we vsill later
employ the mathematical formulation of the so called
"shrinking core model"', as developed by Braun, Lewis gram of chalcopiTite copper leached, the following
and Wadsworth. However, we recognize tliat there are number of gi-ams of O^ will be consumed:
many conditions in which a sharp boundary between g Oj consumed = (1.25 -)• 1.75fPV) g chalcopjTite
the leached rim and the unreacted core boundary is
•not observed because of variable reaction rates of Cu leached [3]
the sulfides, acid gangue interaction, sulfide concen-
Actually the amount of oxygen consumed per gram of
tration, grain size of sulfides and gangue minerals,
chalcopyrite copper leached is somewhat greater than
and porosity of the rock. A generalized model, taking
this, if account is taken of the oxidant required to pre-
account of several of these factors has been developed
cipitate, as jarosite (KFe3(S04)2(OH)6), the iron ex-
by Bartlett.*
changed for copper during cementation (2.5 lb Fe/lb
Most low-grade waste, from which copper is leached, Cu) and the iron produced in leaching the chalcopyrite
is derived from the outer pyi-itic halo of porph^Tv cop- and pyrite. The precipitation of cemetation iron tvithin
per deposits, where the copper-bear ing sulfide is chal- the dump must clearly be taken into account even
cop^Tite.' We assume that chalcopyrite and pyrite are though the source of the iron is outside the dump.
the principal sulfide minerals and that they oxidize in Further, if the excess acid produced by the oxidation
a waste dump environment in the following manner: of pyrite is neuti-alized by reaction with gangue of bio-
O-I + I CuFeSj + I (2H* + SO;") - f (Cu'" - SO;') tite composition* additional iron is generated, oxidized,
*Biolite has been found to be more veacU've by a factor of •~ 100 than other
•t--|(Fe** + SO;-) + i H j O + | S [1] gangue minerals in a porphyry copper intrusive. Calcite. tlie only other higlily
reactive mineral likely to occur, is usually present only in minor amounts.
Qj + -I FeSj + -f H3O - -I (Fe** -t- SO^)
and precipitated. With these additions Eq. [3] becomes:
+ f ( 2 H * + S0;-). [2]
g O, consumed = (1.75 -^ 1.91 FPl') g chalcopyrite
E\-idence that tliese are the oxidation mechanisms
for the two sulfides comes from studies by Wadsworth.° Cu leached [4]
Observations by Stephens^ show that sulfur is a prod- Waste material typically contains 10 to 100 moles of
uct of the oxidation of sulfides in waste dumps. It can pyrite for every mole of sulfide copper. Thus pyrite

I?
be seen that for every mole (64 g) of chalcopyrite is by fai' the most important oxidant consumer if it is
leached, 5/2 mole (5/2 -32 g) of oxygen will be con- oxidized in proportion to its molai" ratio to sulfide
sumed. If FPY moles of pyrite are leached per mole copper.
of sulfide copper, an additional 7/2 FPY moles (7/2 Because leach solutions cannot carry significant
• 32 • FPY g) of Oj will be consumed. Thus for every oxidant with them as they move through the dump, air
FLOW Of V;ATES DOWN THROUGH is the main source of o-xidant within a dump. A liter
ir8«T^- Flow!
of air contains 0.28 g Oj. Fig. 3 shows that Eq. [4]
requires fai' more air than water to flow through a
waste dump if the effluent solutions are to contain the
copper concentrations typically observed. For the par-
ticular dump we shall consider, at least 80 times more
air passed through the dump than water. That is, for
'each liter of leaching solution leaving the dump with a
net gain of 0.25 g/1 (2 lb. Cu/1000 gal.) copper, 80
liters of air are required to supply the oxidant neces-
sai'y for the chemical reactions involved.
CONV'i:CTIVE FLOV.' OF AIR
U? THROUGH THE DUMP
Eqs. [1] and [2] not only tell us the amount of oxidant
consumed per gi'am of copper leached, but also the
Fig. 1—Couniercurrent inierlocking flow of air and water heat generated per gi-am of copper leached. (Tlie eiit-
through a ie.ich damp. The flow of-.vater is u i u a l l y i n t e r -
miitant. halpy of reaction, AW^^, of Eq. [l] is approximately

6 I S - V O L U . M E 6B.DKCL".\1BER 1975 METALLURGICAL TRANSACTIONS 13

 of Go transport into a waste dump. In this respect our
model differs from that of Harris,^ whose pseudo-par-
ticulate leaching case assumes that O3 transport into '
a dum.p occurs primarily by diffusion through the inter-
stices of the particles.
In the next section we generalize Eqs. [4] and [5]
slightly to take into account copper sulfides other than
chalcopjTite. We then develop the rate equations r e -
lating copper extraction with ox7gen upta-ke and heat
generation. We show that these rate eq-uations are gov-
erned by the chemical and diffusional processes oc-
curring during the leaching of waste particles in the
dump. Finally, we derive three equations describing
o.x-ygen balance, heat balance and convective air flow
which are the basis for our one dimensional model.

Form.ulation Of a One Dimensional Model

Because sulfide leaching is usually donsinated by the
leaching of pyrite, Eqs. [4] and [5] are quite easy to
generalize. Provided FPY is taken as the moles of
P5n"ite oxidized per mole of sulfide copper o.xidized,
and provided FPY is greater than ~ 4 , Eqs, [4] and
[5] will hold to good approximation even if sulfide min-
erals otlier than chalcopyrite are present.
We assume sulfide oxidation takes place in a dump
only where the air filled pores of the dump contain
o.-<>'gen and that the oxidation proceeds at a rate inde-
pendent of the actual oxygen concentration. Unpub-
lished studies of the bacterial oxidation of waste dump
5 10 15 20 25 30 35 40 45 50 leaching solutions with air, conducted by the second
author, have shown tliat the bacterial oxidation rate
MOLES py/MOLE SULFIDE Cu LEACHED (FPY)
of ferrous iron is essentially independent of oxygen
FPY »-
Fig. 3—Graph relating the a i r liow through a dump to the concentration in the air until the concentration falls
r a t i o of the moles of pyrite 'esched to sulfate copper leached below 1 pet. More recent studies by the second author
(FPY), .(ov a tj-pica) efliuent copper concentration. involving the uptake of oxygen by wetted mine waste
show that oxygen uptake is substantially independent
of 0-\7gen partial pressure for the same range of o.xy-
- 108.8 kcal;* AH;^ for Eo.. [2] is - 94.9 kcal). If we gen concentrations.
*l kcal = 4.1S6.SkJ. By contrast nonsulfide copper is leached with acid
again take into account the heat consumed in the pre- alone. Acid generated by pyrite oxidation anyv^here
cipitation of jarosite, require acid and iron balance, in the dump is recirculated through the dump in nor-
and assume 2.5 lb. Fe are exchanged per pound of Cu mal operation. Therefore, nonsulfide copper leaching
at the precipitation plant: sliould take place everywhere in tlie dump at a rate
independent of the presence or absence of nearby
kilocalories produced = (2.89 -f 5.4.1 FPY) g chalco- oxj'gen.
pyrite Cu leached [5] Suppose the fraction of sulfide copper remaining in
the dump after some leaching is A'^, and the fraction
Again it can be seen p\Tite oxidation -will, in all proba- of nonsulfide copper in the dump is A'^^y^. Let tlie orig-
bility, be the most significant source of heat. The rate inal sulfide copper grade be Gc and the original non-
at which a waste dump heats up is a direct measure of sulfide copper grade be G^yj. Then the rate at which
FPY. copper is leached from the dump, ^ c u : '^'^^y '^^ ex-
Eqs. [4J and [5] contain the fundamentals of a inodel pressed as: .
of the dump leaching process. Sulfide oxidation reac-
tions consume oxygen from the atr in a dump. Since dX dXi
Oj is a heavy component in air, the oxygen depleted a?Cu P;j(l-*)(^Gs-gj- +G.vs-5|-j- [6]
air is lighter. Since water vapor is a light component
of air, saturation of the air inside a dump with water Similarly the rate of oxygen consumption, (RQ,^ and the
vapor will also produce buoyant forces. Buoyant forces rate of heat generation ta^, may be expressed, from
tend to produce air convection. Furthermore, the oxi- Eqs. [4] and [5]:
dation reactions are exothermic, which also promotes
air convection. A AT of 20"C produces buoyant forces f/A'c
two or ten times larger than complete oxygen deple- fl'02 = P / ? { l - * ) G 5 - g ^ ( 1 . 7 5 + 1.91 F P y ) [7]
tion or complete water vapor saturation, respectively.
For normally observed pemaeabilities, air convec- dX,-
tion rather than diffusion is liie principal mechanism '^.•\ = P ; ? ^ l - * ^ ' ^ s " d f " ^ ^ ' ^ ^ "^ 5 . 4 1 F P y ) ' [8]

.METALLURGICAL TRANSACTiO.XS B VOLUME 6B.DnCEMBtR 1975-619

Pf^, t h e density of the rock waste, i s commonly about
2.7 g/cm^; * , the interblock porosity of the dump i s Table I. Parameters tised for tho Determination of r ^ s and r ^ j
~ 2 5 p e t ; p ; j ( l - * ) , the bulk density of the dump a s a
Parameter Description Value
whole, is equivalent to 1.7 tons/yd^.
The r a t e of leaching, 1:^X5 / d t and dXj\is/dt, may be d e - Radius of waste particle 1.5 cm
s c r i b e d also in t e r m s of leaching from a waste p a r t i c l e . "sulf .Surface aica of sulfide mineralization per ^80 cm''
unit volume of waste
Let u s suppose that the leaching of a sulfide-bearing
Diffusion constant of Fe"* in water =»2 X iO* cm*/s
p a r t i c l e i s governed by an equation of the form: FPY Mofes of pyrite leached per mole and sulfide 47
copper leached
[Ox] Gs Sulfide copper grade O.lSwipci
OA-"sulf
K O.xidEnl required to leach a unit volume of 0.444 ijcrvi^
ivarif particle
and that [O.vj^ just outside the o r e p a r t i c l e is a knowT. First Older rate constant for the oxidation ==IO'\-.-n;'s-
function of t i m e . The symbols given a r e defined in of pyrite by Fe-*"
[Ox] Concentration of Fe*** in leaching solution lO-'g/cm^
Table I.
T' Tortuosity of diffusion channels =5
For a s i m p l e one-dimensional c a s e , Eq. [9a] b e - Reaiinn skin septh (see Eq. (11)) Calculaifd-vaiue
comes: 0.142 cm
Porosity of waste throug which diffusion 4X 10"'
D'OX'^RI d'[Ox]
Rl can la.'te place
-Vv«sulf[0*] ^=0 f9b]
Th dx'
Satisfying the boundary conditions [Ox] I' [O.v]'
x=0 'DS - l^^'^ mo.'' In addition, TQ and T^J may be given
and [ O x ] f ^i;;;^ 0, the solution is
* 1 m o = -M cf.
-x/d a t e m p e r a t u r e dependence:
[0.v]f
d[Ox]f x^O
[10] jT'lOOO-E"
dx (T) = 7 ( r = 0"C) EXP
R (273) (273 + r ) . '
/?; •*p [16]
Wliere 6 = r e a c t i o n skin depth =: [11]
^oxTR(ts'Ai This introduces activation e n e r g i e s E ^ S ' ^CS.- ^OA',?:
^C\'S- Fi'oni the literature^" reasonable g u e s s e s for
The r e a c t i o n skin depth is the distance into the pai-ti- £ o 5 and ££>A"9 would be 5.0 k c a l / m o l . E ^ ^ and £cA'.?
cle where t h e oxidant concentration h a s fallen to 1 fe
might range from 14.0 k c a l / m o l to 20.0 k c a l / m o l , the
its initial value. Since by hypothesis the r a t e of o.xida-
activation e n e r g i e s r e p o r t e d for the leaching of py-
tion reaction i s proportional to oxidant concentration.
6 i s a m e a s u r e of the distance into the o r e p a r t i c l e r i t e " ' ' ^ and for chalcopyrite.^^
that significant chemical r e a c t i o n takes place. Given values for r ^ ^ r "TCS, TD.VS- '^CA'S^ Eqs. [12]
Using values given in Table I, 6 = 0.142 c m . The and [13] d e t e r m i n e the r a t e of leaching at any point
r e a c t i o n skin depth i s therefore thin, r e l a t i v e to tlie in the dump at any stage of leaching. A'^^y^ and A'^
a v e r a g e size p a r t i c l e d i a m e t e r . Leaching of a p a r t i - can be updated after each increment of model l e a c h -
ing. Model time i n c r e m e n t s m a y be taken a s short a s
cle can, tlierefore, be described in t e r m s of the s o -
desired.
called shrinking c o r e model v/hich i s of a s i m i l a r form
to that developed by Braun, Lewis and Wacswortli.^ The most s e r i o u s approximation in E q s . [12] and
[13] is probably the assumption that the dump is c o m -
rfA's - 3X1^' posed of waste p a r t i c l e s only of one s i z e . Tliis may
[12] not be as s e r i o u s an approximation a s it might at
"dT 6TosXf(l-4^')+Tcs first seem, given tlie tendency of s m a l l o r e p a r t i c l e s
to clump together and leach a s if they w e r e a l a r g e r
dX:N S ^x%l [13] aggregate, and the tendency for l a r g e o r e p a r t i c l e s
^* 6 T o , v s A ' i f c ( l - A ' t f , ) + r'^CArS to have l a r g e enough c r a c k s that they leach like s o m e -
what s m a l l e r p a r t i c l e s . ' "
w h e r e r ^ is the time riequired to leach a waste p a r t i - F u r t h e r m o r e , r e c e n t work h a s also shown that cop-
cle completely when the p r o c e s s is solely diffusion per r e c o v e r y r a t e s from operating dumps at Kenne-
controlled, and TQ the time to leach a waste particle cott's Bingham mine can be c o r r e l a t e d quite well with
completely when the p r o c e s s is controlled by the d e - laboratory studies of copper r e c o v e r y from similai'
c r e a s i n g surface a r e a of the shrinking unleached c o r e . m a t e r i a l when the mean p a r t i c l e size of the waste,
TQ and TQ can be computed theoretically from the fol- as found in the dump, i s compared with laboratory
lowing r e l a t i o n s : leaching studies.
A'a Tlie heat balance in a waste dump may be described
CS = [14] by:
^ox asulf 6 [ a v ]
T^a'K
[15]
PrCr ~ar- = -{f^iCiVi + PsCgVg)-vT + (R^
'DS
6[av]i);,.*'« + KyV ^ r [17a]

Using tlie values given in Table I, T Q ^ ~ 903 m o . -and where p and C a r e the density and lieiit capacity 01 the

6 2 0 - V O L U M E 6 B , DECEMBER 197S METALLURGICAL TRANSACTIONS B

total dump ( s u b s c r i p t T) and the liquid (subscript I) application v;as used, an approximation that has been
and gas (subscript^') phase of the dump. V; is the shown valid so long a s the leach cycle is l e s s than
Dai'cy liquid velocity (i.e. cm^ w a t e r / c m ' dump s u r - tliree months. As will be discussed later the ambient
f a c e - s passed through tlie dump). Vg is the darcy air temperatui-e was varied seasonally in a manner a p -
v e l o c i t y through the dump {i.e. cm^ a i r / c m ^ dump p r o p r i a t e to the location of the dump (temperatiore
ai-ea-s). K'f, the t h e r m a l conductivity of the dump is m e a s u r e m e n t s were available from a mine station).
taken to be 5 x 10"^ c a l / c m - ^ C • s. For calculation, The surface t e m p e r a t u r e of the dump was also varied
Pj-Cj- = 0.6, PiCi=: 1.0, and p^Cg = 1.3 x l O ' ^ (0.126 seasonally but at a higher average t e m p e r a t u r e and
+ 0.02837"), w h e r e T is the t e m p e r a t u r e of the dump. over a m o r e r e s t r i c t e d r a n g e . Air convection kept
ITiis last e.xpression takes into account the thermal the dutnp surface w a r m e r than the s u r r o u n d i n g s . Snow
effects of evaporation. It is a s s u m e d the air in the was observed to m e l t m o r e quickly on the dump than
dump i s always s a t u r a t e d with w a t e r ; account is taken in the surrounding a r e a s .
of the i n c r e a s e in water saturation values with in- Given a set of p a r a m e t e r s and operating p r o c e d u r e s
c r e a s i n g air t e m p e r a t u r e , T, and the effect, through (rate of application of water), the finite difference
the latent heat of vaporization, this would have on the model computes tlie leach history of the model dump.
heat capacity or heat c a r r y i n g ability of a i r . The dump is considered to be broken into A' l a y e r s .
F o r a one dimensional dump (i.e. a i r and water flow Tlie percent copper leached per month (or the efflu-
r e s t r i c t e d to be v e r t i c a l only), Eq. [I7b] simplifies to: ent copper heads) can be computed easily:

,C„K„
, dT
m^ + KT
fT_ fraction Cu d X r o r ( 2 . 6 8 X 10^ — ) 2.68 X 10'
^T^T - ^ = iPl^l^l mo dt \ mo/

[17b] ^VS dXl'

('^A'.S
dt dt [22]
One dimensional convective air flow thi-ough a dump E A'(Gs + C^r^)
may be d e s c r i b e d : dump
'^AVE AP [18]
v.. HEADS [g/1 Cu] = p ; 5 ( l - 4-) H
H
dXk
H is the height of the dump, \ig is the viscosity of air (GA'< + G<
dt dt - ) . , 000 [2.3]
= 1.9 X 10"' poise.* AP is the p r e s s u r e drop a c r o s s the
dump A^^;
• l poise = 0.1 Pa-S.

dump. A-P may be e x p r e s s e d a s : Tlie cumulative pet leached, 1 - XJ.Q,J. is just;

AP = Poogo^^ T j ^ M T i ) Ti + 3 (1 - [ 0 , Y ) ) [19] (G,vs-^A'S + ' ^ s ^ s )
i fraction Cu leached = 1 [24]
' N S + Gc
Here Pog i s the density of air at STP, go is the g r a v i t a -
tional constant, H^ = Hj / H , is the n o r m a l i z e d thickness The next section c o m p a r e s the r a t e of leaching and
of the ith i n c r e m e n t a l level of the dump. a{T^) is the the cumulative leaching of a test dump to the r a t e of
t e m p e r a t u r e dependent coefficient of t h e r m a l exj^an- leaching and cumulative leaching computed by the
sion which, like the heat capacity, includes the effects model through Eqs. [22] and [24].
of changing water vapor s a t u r a t i o n . ^ is a coefficient
which d e s c r i b e s the d e c r e a s e in_aLr density due to oxy-
CALIBRATION OF THE MODEL
gen depletion ((3= 2 . 8 3 x 1 0 " ' ) . [ O ^ ] ^ = [ 0 , ] S / [ O z Y
STP-
kAVE; tlie average permeability of the dump, may be Fig. 4 shovv-s a c r o s s section of the Midas test dump,
expressed: built of mine waste with normal size distribution by
1 the Utah Copper Division of Kennecott Copper Corpor-
[20] ation at Bingham Canyon, Utah. The dump is about 400
Eri,i/ki ft long, 200 ft wide. The average depth is 20 ft with a
Any distance, z^, from the b a s e of the dump, where
fresh a i r i s a s s u m e d to e n t e r , the oxygen c o n c e n t r a -
tion in the dump will be:
7160
ffi 02^i 7U0
[O,]^ = 1 - [21] 7120.
VslO^VsTP 7100

TEWPERATUflE DISTRtSUTION - S«uon C-C

E q s . [17b], [18], and [21] r e p r e s e n t a model of the
dump leaching p r o c e s s that includes both physics (air
convection) and c h e m i s t r y . The equations can be solved ;160- CO Ci CO CS C4 C3 c? Cl
using an implicit finite difference s c h e m e in which the 71iO
dump is considered to be broken into .V l a y e r s . The 7120
method used was to s t a r t the dump leaching at some 7100J
s t a r t i n g t e m p e r a t u r e and loop between Eqs. [18] and OXYGE.M D I S T R I B U I t O N - SM1I0.1 C < '

[21] until a steady s t a t e O^ profile and air velocity

L Jb_ JbO_
was attained. Then Eq. [17b] was used to d e t e r m i n e 50 - 100
H 0 B 2 , R V E R T . SCALE
the t e m p e r a t u r e of the dump at I + A.'. A/ was gener- F i g . 4—.August 1969 t e m p t s r a t u r e and oxygen d i s t r i b u t i o n in
ally taken to be one month. Tlie a v e r a g e r a t e of fluid a s e c t i o n t h r o u g h the -Midas test d u m p .

M E T A L L U R G I C A L TRANSACTIONS B VOLUME 6B. DECEMBER 1975-6-21

 /

maximiuhi depth of 40 ft. Fifty-eight ItS aching ponds; T (mo) = 10°C ^ 10 • cos ((rno, - 1)11/6)
cover the top of the dump. The waste tonnage beneath
tt-!_e..ponds is about 93_.O00 tons (assuming 1.7 tons/yd^). where mo runs from 1 to-.12 and is the number of the
The waste itself is 60 pet quartzite and 40 pet biotite •calendar inonth. Thtis the top surface- of the durnp -ifas
granite. The-average grade of the waste is dil45 pet assumed to vary seasonally betwe'eii;32°F and 68°F,
copper. 80 pet o'f the copper was sulfide, dbminantly (0°e and 20?C)i a slightly moi-e restricted and;hotter
chalcgpyT-ite; the r e s t was rionsulhde copper. range than the aiiibient terriperattire variation'Of 19°F
LeacMng of the dump^began on Apiril 9, 1969. Prior- to'es-'Ft-T'ctD n°e).
io this, there had Been some runriotf through the dump The temperature at the base-of the test duirip -^as
but very low copper extraction. Fig. & showsVthat observed to fluctuate ..some what. The assumption of a
leaching aiter water application was slb-W at first, in- constant 2Ci°C basal temperature is ^a matte if o'l con-
creased rapidly to a,maxiiiium,about five rnonths after venience and. is probably subject to-sorrie'error. B'oth •
the s t a r t of leaching, a;nd then fell steadily, with some boundary conditions are plausible.. Subset^uent work
fluctuations that appear correlated with the season has shown; these "assumptions to be .quite reasbriable.
(inaximum in summer). Fig. 4 shows.that by August The stai'ting temperature of the-model dump was
1969 the internal dump temperature had risen;to'130°F lO^C. The Midas, test dump was built In winter so the
(a4°C). There ^yas substantial oxygen depletion as the dump VJ as ihitially at least this cold.
air cori-i^'ected through the dump. It can. also be seSn Fig. 6 cpm.pares the-.talibrated model leaching-his-
that the. air eonvect'ed in alotig the, high per me ability tpry to-the leach irig history of the Midas; test duTinp
base of the-duriip iand then up through the dump—the shown in Fig. 5., The match in general is quits good.
one diii-i^nsibnal model is-appi-opriate for this case. In addition to the leachirig'hi story similarity, the model
As time went on the location of maximum dump tem- dump reached .51°C internal temperature by August
perature' sMtted frorn the fax end of the dump Jas 'shown lf69 and, then decreased'in temperature to abffat 14 X ,
iii Fig. 4) to about;the s,ame distance from the near end; as did: the far end of tlie Midas test dump. Iii August
The parairieters used ih the'model are listed in Table 1969 the, effuent oxygen concentr at-ion was 9 pet, in
good agreement with observ.atipii (see Fig, 4). The
The following thermal boundary'conditions were values bf'Tcs and TQC are quite clbse;to the values
chosen. The base of tlie dump was fixed, at 20°G. The anticipated from Eqs.[i4]<and [1.5'] (cprapare, Tables
•top surface temperature was allowed to vary: landH)'.
The. initial rise in extraction rate is d.ue to the heat-
ing up of the dump; This feature- is' hot peculiai- to the
T—1—r—r—I—^—r—1—i—r—r~~i—i r
Midas test dump, Shd can he oBs^erved in the:le.aching
histoi'y-of, many dumps. The fall in leaching rate after
the first seven mpjiths -of leaching is due to, the fact,
the more a'ccessMe cDpp'er'-.has.'bGe'n leached arid Fe*""'
must.'diffuse thro ugh. already, leached.areas to r.each
the remaining copper, The l a l l i p dump temperature
also contributes: to the decline in leachirig;.rate:

VARIATIOI^S FROM THE BASE

MODEL
MIDAS TEST DUMP tt "is of-'.iritei-est to vsry the model pai^ameters: to see.
UCO what;effect they n\ay have on the rate ;of'popper extrac-
5'Mo. Ruftning^riNveTage
of- Data

Tabla l l . Rm'ametm Used f b r the Mmiel Shown in Fig. 6

PaTa'metfii -Descripiibi) Value

H Meighrof-dunip 670 cm m S i )
FPY M oics py ri le Icachi (i / linol i:' Cu',; le ache d 47
H-'- Dum|i.pcrm£abiUly t ^ KE l,0-'cr,i*'nOO(i
•^ darcy s)
^, R;ite;of watev-!!pphcali,on 2;26.X' fO"' cm'/
cni°duiiiir-s(0.02
gal/fl=-l,);
Cs, 'DLimp Eiiiride coppe't'Ei-adc. O.i Ifv ivi pet
fi^-S Dump iionsviiriSc copperjgtails 0,D-?9 \vi pel
•rps Difj'usiunal sulfide leach lime i 2 ( f Q 1700 mo
" • « :
Lcadviime for sulfTde copost umier sutface •200 mo
aiea f;il_e: c'omrol {i6''G,'|
J L J i: 1 i L_J I I 1 1 I "ftVS DifruEJ'onal ndiisuirisk lcac!-i-.i5nK,(.30°G.) 500 mo
'^&-s "LcLicir.Uinc for noii&uirule c'Qp|>cr under .r 00.1110
iUtCace/area [aie--gonivol (.^O' C;)
p*.-. t:*. ... Aciivalian-Vrici-gies for di ff-Ji'i'on S'Olvcal/nipl
F i s . i — T h e ' r a t e of e s t r a c t i o n ajKl-.cuniiul;iti'y.e^extraci"iQn..of" ! : « ; I,-*" ., .•Vciivaijoii-eisei^ici for cl-ieraicsl leacliing .!S.01.tal,'mol
coprjer "troni t h e i l i d a s test"dui-i-i'i3. u s i n g a, tiv/e-tneVit IT r u n - tc'aciioiis;
ning';r;'cgr age.

-YOLU-ME 6&, DECEMUE-R 197:5- M'ET.ALLUR'CiC.jj.L TR.VKS.'\"C~T10NS. IS-

 V.IDAS TEST DUMP 5 MONTH RUNNING AVERAGE been included, as yet, in the model. Secondly, not all
24.0
the parai-neters listed in Table i n are mutually inde-
23.0 pendent. For example, increasing FPY at a constant
22.0 sulfide copper grade will cause TQ^ to increase sub-
- 21.0 stantially. Lastly, the combination of parameters that
20.0
successfully models the Midas test dump is not neces-
sarily a unique set or the correct set. Data from more
19.0
than one test dump is needed to resolve these uncertain-
18.0 ties. The most critical uncertainties are probably the
O
17.0 chemical activation energies and F P F (see Teble II).
O A OBSERVED 16.0 The lack cf dependence of leach rate on permeability
CALCULATED
15.0
simply indicates the dump was shallow and permeable
o enough not to be oxygen starved any\-v-here. Had the
14.0
dump been thicker (~100 ft.), a significant dependence
13.0
of leach rate on permeability would be noted.
12.0

11.0
CONCLUSIONS
O 10.0
U
9.0
From the discussion presented it can be concluded:
8.0
1) Air convection is ai-i important part of the d-amp
< leaching process and must be accounted for in any
a: 7.0
successful model of this process (Fig. 3).
6.0 2) Exothermic oxidation reactions heat up -ivaste
5.0 dumps with time (Fig. 4). Any leaching model that is
4.0 to be applicable to real dumps must account for the
3.0
temperature dependence of the leaching process.
2.0
3) A simple model that requires energy., mass and
momentum balance, and that derives rate control from
1.0
a temperature dependent shrinking core model and a
i___JI L I I I I I 1 I single "average" waste particle diameter (Eqs. [12].
4 6 8 1 0 1 2 2 4 6 8 1 0 1 2 2 4 6 8
[13], [16]) has proved remarkably successful in a c -
•69- DATE -70-
counting for the most important observed featiu-es of
Fig. 6—Comparison of observed and calculated Mitlas test
dump copper leaching behavior. the leaching history of a well studied test dump (Fig. 6).
4) Dump height, ILxiviant application rate and dump
permeability are the most important factors affecting
the rate of copper leaching that are accessible fo op-
Table I I I . The Effect of Parameter Variation on Copper Recovery erational alteration. (Table IH and discussion in text).
After 24 Months of Leaching

Increase in Copper Recovered

ACKNOWLEDGMENTS
Parameter Variation in 24 Months, Pet

// y -HO
The authors would like to thank Kennecott Copper
.5.9
Fpy y •HO 6.1
Corporation for permission to publish this paper. In
ros -10 6.2 pai'ticular, the authors would also like to thank the
Gs y •HO 4.8 Messrs. W. D. Southard, A. D. Pernichele and B. P.
Els -HO 2.7 Ream of Utah Copper Division for permission to use
•fcs -10 LS the Midas test dump data as a base case for this study.
VI V/ -10 1.3
-10 1.3
Enough recognition cannot be given to the e.xtensive,
''DNS
•HO careful work of these individuals and others that was
''-h
Gss •HO
0.9
0.9 required to collect the high quality data so easily cited
''CNS -10 0.4 in this paper.
•HO 0.4
I^D.VS
•HO
Parts of this paper were presented at the 78th Na-
Starling Temp 0.4
I^CNS •HO 0.0
tional Meeting of the American Institute of Chemical
k' •HO 0.0 Engineers in a Symposium Section on Modeling and
-10 -0.4 Analysis of Dump andln-Situ Leaching Operations,
August 19, 1974 in Salt Lake City, Utah.

tion. Table IU lists the paran-ieters of Table II and NOMENCLATURE

shows the percent increase in copper extraction after : radius of a waste particle, [cm]
twenty four months of leaching that results from a 10 sulf : surface area of sulfide mineralization
pet alteration in the listed paran-ieter. The parameters per unit volume of waste, [cm"^]
that are checked affect the rate of leaching primarily heat capacity of gas phase in dump in-
by allowing the dump to attain higher temperatures. cluding effects of variable water satu-
Tlie reader is cautioned tliat the variations in leach ration, [g/cm^]
rate shown in Table III are based only on what is in the heat capacity of mobile liquid phase in
model. Much may go on in a waste dump that has not dump, [cal/g °C]

M l i T A L L U R C I C A L TR.ANS.ACTION'S B V O L U M E 615. D E C E M B E R 1975-623

/ •

heat capacity of'dump as a. whole, ingyn-the durhp o.r a given layer 'oi the-
[cal/g "Cl dump, [-]
•ox diffusion constant of oxidant in water, : 'distance of center of ?"th layer of dump
[crnVsl from base of dum'p, [cm.'l
^DSi'^DNS activation energies describing, through
Eq. [16], the temperature dependence of GREEK LETTERS
'ps.i '"iJjV.Sr ['kcal/mole] ; coefficient M ther.mal expansion bf air
^CS' '^CA'S actiVatioh energies .describing, through irxluding effe.cts of changing watei; vapor
Eq, [16], the temperature dependerice'of satufatipn. [-]
"^'cs^ "^CNSi [kcal/mole] ; caefficient describing the, c.hainge In.air
Fpy^ mole's of pyrite. leached per'mole of sul- densitydiie tb o.xygen depletidr., (See
fide copper leached.[-]
Gs initial copper sulfide grade of dvimp, 5 • reaction skin depth, [cm ]
[wt fraction, Cu] : viscosity of the .gas phase in tha dumo.
G,NS initial copper,nonsulfide grade; of dump, M , ; • ' • • " . "
[wt- fraction Gu] Pn : density of waste;particles, ['gj/cqi'j
gravitational acceleration, [cm/s^] : dehsity'pf the,,dump as a wh'pie ("Total"
H height .of dump; fcm] ,P.T dump)" {pj' = pjf:{\ - .4]), [g/cm*].
thickness of "1th layer of durhp, [cm] : density of liquid phase of.dumD (water).
diniehsionless thickness of ith layer of Pi [g/crn^]
durapHi =H^./ffv[-l ; density of;gas,phasj in duinp including
.AHn: : ehtha.lpy: of reaction,, [kcal] effect of variable water vapor saturation,
; oxidant required to leach v.iaste parti,c,le, [g/cni'],
[g/cm^ particle], : density of air at",'stahd:aa~d_temperature
: thermal G.onductivity of dump a.s'awhole and.pressure. Ig/.'cm']
(total'durnp), [cai/crn °C • s] ^DSi^BNS : time to leach typical waste particle com-
k' .-intrinsic permeability of thejdump, [cm^] pletely of sulfide, or nonsulfide; copper
k : average permeability"of the dump. a'.ssurning rate "of- leaching is iim-ited By
{1 Darcy ii. 10"^ cm'), [cm^j 'diffusioTl of'6xidaiit..pr acid- into the par-
; Chemical rate c6iistant:'characterizitig ticle, [mo]
f ea^ctionof oxidant ahd s.ulfi'de minerals, ^QS'^-'-^m : time' to leach-typical waste particle com-
[cm/s] pletely of sulfide or nonsulfide copper
• number of layers into -which .diiinp has assuming the rate!of leaching is con-
beeTiar.bitrai-ily broken for saj^e of com- trplied by the. shrinking surface area of
putation (usually 30), [.-] tlie sulfide .or ,nen-sul'fide'copper ("cliem-
[Ox] : concentration .'of >oxidaTit, [g/cra^] ical" control), [mo]
: concehtratiqn of oxygen in air'und.er ,* : interblock porosity.df.tdump (usually
standard,conditions of temperature and - 3 5 p c t ) , [-] •
•pressure, jg/cm^] : poros'ity of •waste through which, diffusion
:• coneenti'-ation of oxygen in ga'sfphase of can;take place, [-]
du-mps, [g/cm']
: normalized: oxygen concehtration in gas
phase^f,dump tO^^ -.l0s]^/(Q,]|j,p, [-]
&P : pressure ,drcip''across (bottom to: ibp)'the REFERENCES
..). H, VV, Shcffer-anii-L. G.yyiipi-liureaii of Mines /.C.-.: 1.9.6,8, noi:S341,,'57 pp.-
dump (lO,^"dynes/cm*^ -^ 1 atmosphere), rl: J.. A. 'Harris: frocMusmliht h)S!...Miti!ngMct.. 1969:, no.-2j0;-pp, Sl;92.
[dyn/cm^] -3.«R.'J, Romanj,B.''R. Bciiner, md.CfVt'.'Bh'ckir.'S^K-Ai/lfE Trans., 'l'974.\bl.
R :"gas constant i"-i2), [caI/°C-mol] 256,-iJp.'24.7,.,ft. : " ' ' • ' i
:T^te of heat generation,, [kcai/Gm-^dump - s]" 4. R. Lj Biaun, A, E.-Lewis,'and M, E.-Wadswoi'ji: Sollitioiii'lininsSyinptjmim,
-Cliap.t; 2.1, pp.. 295-3-23, Aiiiericati Insiitu't'c ef .Nfinins, Meiaihirgical'sijid
So, : i-ate of .oxygen ccnsumptidn, I'eirokum En£inf'er5:li1,C;, New Vntk;., 1974;
[g OE/CITL; dumpvs] 5. B..W. M3"ten, 'k. I;-. Wadswonh.-ajid R. D. l3rovcs; SjyjTMtf'f Tniits,, i'975;
ISGu : rate of copper leaching [g Cu/cni^-durnp-s]
: temperat-jre' of .dump at any particular
location. Teniperature: of water, rock,
v61;.3Sg,.p'p,.'f.9-'74.:
S:_R, W. SarilciL Prcicecjiiiigsoftltc International ,Synjpopum''on'Hydro'
rniialtut-^-.-.Chipi., 1-4, pp, 331-7,4, Aiiierican Institute o'f Mmiffs.-MsiaUuigicai I
rk
and gas phases assumed.'identical, [°C]
: tortiiosityof diffu.sipn channels; [-]
and,Peii-oicuiii Eng.neers.lnc.-, M5w'Y,6i'kV'i^/3.
7. J, Di-lyiwe'll'aiia'], sil; Guilbctt: Econ. Ceo!.., !'.9'70;:v,61. 6.5'; pp- 373=4*08.
•^. M,_E._Wadswo[-Lh; Univtrsity'of-LfmIi„S!iii Lai;c CJty,.UiyL,p=rs6\ia(.cdhi-
I
I
: darcy gas yelocity through, dtimp, rriunicatjpn, 1974..
fcirijg^s/chi'dump-area •:S| '9'. i, Di Siepliens.^K'cniic-coti Copper Coiporaiion, Saji Lske CJly.^Uial!. peraoi'ial
c.onimiiiiication, 1,97.4.
: darcy yelocity of -vvatei- passing through
the'dump. Average rate oyer application.
periods arid rest cycles is used,
10: "V. T: Auck and M. E. \Vids\m'iih-,.PkKee(UiiSiof!l!i>Jiiieriii};ianiitSym-
po.iiiimoii HydftimeisUiirgy, Cliapi, 25, pp. .6;45-700, Ajii^ricai; Jnjtitiii.e of
Mining, Maialluriical'ariy.l'eirolepm-Engiheerilric.-, New Vork, .i.973.
I
[ciii^^ater/cm^dump sur'face • s] IJ.CT.Maihcwsani! k.,G.'Rribiiis'v^ysf.'pimii^Ris,,- 1972. pfpf'31-2S.
*. fraction of..initial liOnsultid.e copper "r'e- 1-2.. E. _E;-Smith';arid K...S. Shii(iia:c: WaturPoiluiipniCqniroi'i^ejearcli'.S^'rics
^MS
14"0i0FRS,()7/74.U,.5^%parime!M-of the Inier'i'Q"r,,Ffde:a! Water W'ual'lty.
-maining in dump",br given layer of dump, j^dii-iinisiraiioii.-l.S-Jd.
13, A;E...Lew;s;and R; L;-Braun; .UCRL.7'32S4.prcijrini.'Lawte:Ke Livermore'
• fract-i'on of initial sulfide.-"copper remain- •Laboiaiifiy, 'January 1,972.,

6 2 4 - V O L U . M E 6B,DEeE,MBI;R 197S .METALLU-RGIOAL T R ' A N S A C T T O N S B

Si
View publication stats