Using Portable XRF for Elemental Analysis of Obsidian, Metals,

Ceramics, and Other Archaeological Materials

Robert H. Tykot, University of South Florida

email: [email protected]
 Corsica

Palmarola
Monte Arci
Lipari
Sardinia Antiparos
Pantelleria Melos Gia
Mediterranean Sources, and
Archaeological Sites
with Obsidian Artifacts
(Sardinian, Aegean,
Carpathian sites not shown)
Typical neolithic obsidian tools
Most sites also have flakes and tools less formal
 Monte Arci (Sardinia) Obsidian

Obsidian from
Primary Deposits at
Conca Cannas (SA)
easily collectible

Four main subsource groups:

SA, SB1, SB2, SC
with further subgroups identifiable
SB1a, SB1b, SB1c, SC1, SC2
Visual and Physical Differences

Density Lipari
Palmarola
Measurements Pantelleria

Useful Sardinia A
Sardinia B1
Sardinia B2
Sardinia C

2.3 2.35 2.4 2.45 2.5 2.55

Density
Electron Microprobe

Major and minor elements only

1-inch disks with 14-16 samples

Electron Microprobe Discrimination of Monte Arci Subsources

14.5

14
Al2O3 (%)

SA
SB1
13.5 SB2
SC

13

12.5
5 5.5 6
K 2O (%)

multivariate statistical analysis provides even clearer distinctions

Obsidian from Filiestru Cave (Sardinia) over 4 Neolithic Periods
% 27 20 29 10
100

80

60 SC
SB2
SB1
40 SA

20

0
Cardiale Filiestru Bonu Ighinu Ozieri

published in Journal of Mediterranean Archaeology (1996)

 Missouri University Research Reactor

Instrumental Neutron Activation Analysis (INAA)

 Laser Ablation ICP-Mass Spectrometry
(with Michael Glascock & Jeff Speakman)

Laser ablation
sample holder
and instrument
Laser ablation trace
200 wide

Laser Ablation Inductively

Coupled Plasma Mass
Spectrometer (LA-ICP-MS)
at U Missouri
microscopic signature left
 X-Ray Fluorescence Spectrometer

non-destructive, if sample
fits in the sample holder

modern XRF instrument: needs liquid nitrogen to

maintain vacuum and analyze lower Z elements

automated sample system

X-Ray Fluorescence

An electron in the K shell is ejected

from the atom by an external primary
excitation x-ray, creating a vacancy

An electron from the L or M

shell "jumps in" to fill the
vacancy, and emits a
characteristic x-ray unique
to this element

When a vacancy is created in the L shell, an electron from

the M or N shell "jumps in" to occupy the vacancy. In this
process, it emits a characteristic x-ray unique to this element
Multiple X-rays are produced, with characteristic energies,
for each element in the sample

Overlaps are minimized by using filters, and different beam

settings (voltage and current)
 Portable XRF Systems

ElvaX Niton

Bruker, with vacuum Bruker, on tripod

 Portable X-Ray Fluorescence Spectrometer (pXRF)

Advantages
 Non-destructive
 Any size sample
 Quantitative analysis
 Low Z elements (w/ vacuum)

Settings for pottery analysis

 no filter
 40 kV, 1.5 μA, 180 seconds
 results calibrated against standards
 Isolino di Varese
Zagreb area

Les Terres Blanches

Gaione
Arene San Polo Istria sites
Combes Candide
Fornace Cappuccini

Ste. Catherine Grotta all’Onda Pokrovnic

Giribaldi
Danilo
La Cabre Casa Querciolaia
Lumaca Paduletto di Castagneto Obsidian analyses by:
Pietracorbara
Strette Colle Cera
A Petra Pianosa Orbetello Catignano Susac
Monte Ortu, A Fuata, Carcu Monte Grosso Poggio Olivastro • Acquafredda & Muntoni, in press
Sarra Cinescu
Monte Lazzu Castellare Giglio La Marmotta • Ammerman et al. 1990
St. Pancrace-Tiggianese I Calanchi Santo Stefano
Basi Cardiccia Casale del Dolce Tavoliere
• Ammerman & Polglase 1997
Passo di Corvo
Liscia Pilastru • Barca et al., in press
Monte d'Accoddi
Grotta Sa Corona
Contraguda
Palmarola
Pulo di Molfetta • Berton et al. 2004
Grotta Filiestru
Sa Ucca de su Tintirriolu • Crisci et al. 1994
Duos Nuraghes Molia
S. Caterina di Pittinurri
Mt. Arci
• Crummett & Warren 1985
Cuccuru s'Arriu
Domu Beccia Ortu Comidu • de Francesco & Crisci 2000
San Gemiliano Serra Cannigas
Buon Cammino Cuccuru Craboni • Francaviglia & Piperno 1987
Barbusi Cuccuru Ibba
Su Carroppu Antigori Acconia • Lugliè et al. 2007a; 2007b
Is Solinas Tracasi Ustica
Lipari • Mello 1983
Grotta dell'Uzzo • Michels et al. 1984
Bova Marina • Pessina & Radi 2006
• Petrassi & Zarattini 1997
Zembra
Mursia • Tykot 1995…2009
5 sites
Pantelleria

Hergla Brochtorff Skorba

Tas-Silg

250 Km
 Monte Arci Subsource Distribution
Early Neolithic Sites in Sardinia and Corsica
%
100 21 10 10 42 26 20 27 35 96

80
Sardinia C
Sardinia B
60
Sardinia A
40

20

0
Strette
Lumaca

Filiestru - EN

Filiestru - Filiestru
A Petra

Santa Caterina

Su Carroppu
Sa Corona - Filiestru
Pietracorbara
 Monte Arci Subsource Distribution
Late Neolithic Sites in Sardinia & Corsica
%
100 10 181 93 11 62 17 83 22 29 24 11 10

80 Sardinia C
60 Sardinia B
Sardinia A
40

20

S. Gemiliano

Is Solinas
Liscia Pilastru

Filiestru
Contraguda

Sa Ucca

Molia

Cuccuru Ibba
Li Muri

Monte d'Accoddi

Barbusi

Tracasi
%
12 20 10 125 59 20 17 21 5
100

80 Sardinia C
Sardinia B
60
Sardinia A
40

20

0
Monte Lazzu
Monte Grosso

Saint Pancrace
Strette

Lumaca

A Fuata
Basi

I Calanchi

Castellari
Open-air, Late Neolithic
site of Contraguda

Areas excavated at Contraguda (above);

Structural feature A-B (left)
 Contraguda Results (246 analyses)
% %
100 47 152 9 28 10 100 45 22 81 5

90 90

80 80
SC SC
70 SB2 70 SB2
60
SA SA
60

50 50

40 40

30 30

20 20

10 10

0 0
unlisted Area 3 Area 4 Area 19 Area 20 Area 3 Area 3 Str. A Area 3 Str. B Area 3 Tumulo B

Comparison of subsources present in Comparison of subsources present

different areas of the Contraguda site within Area 3 at Contraguda
(combined LA-ICP-MS and pXRF (combined LA-ICP-MS and pXRF
data) data)
 Sardinia (Italy)

Previous analytical studies mostly Contraguda

in northern (e.g. Filiestru Cave)

Obsidian from 15 sites in the Museo Grotta Filiestru

Pigorini collections tested here, plus
from open-air site of Contraguda
Frumini
Serra de Castius Triarzu
Nghe. Loddu
Is Arrius Su Casteddu Becciu
Palas de Casteddu Simaxis
Mes’e Arrius
Nghe. Nieddu Nghe. Tiria
La Gumarense
Palmas Arborea
Monte Arci

Pigorini Collection
S. Bartolomeo & S. Elia cave artifacts acquired in
1880s by Francesco Orsoni; Oristano area surface finds San Bartolomeo
collected in 1899 by T. Zanardelli
Sant’Elia
Examples of Obsidian Artifacts Tested from Museo Pigorini

Note some visual differences in transparency, luster, inclusions

Archaeological Sites in Sardinia from Pigorini Museum collection

SiteTime VisualpXRF
Su Cuccuru de Is Arrius (Cabras) Middle Neolithic 13420
Grotta San Bartolomeo (Cagliari) Late Neolithic 15440
Grotta Sant'Elia (Cagliari) Neolithic? 2110
Simaxis Neolithic 1010
Palas de Casteddu (Cabras) Late Neolithic 1010
Mes'e Arrius (Cabras) Late Neolithic 5910
Stazione La Gumarense (Santa Giusta) Late Neolithic-Chalcolithic 99
Domus de janas di Triarzu (Paulilatino) Neolithic-EBA 1010
Cantoniera Frumini (Sili) Chalcolithic 1010
Palmas Arborea (Oristano) Chalcolithic 44
Serra de Castius (Sili) Chalcolithic 1010
Su Casteddu Becciu (Fordongianus) Chalcolithic-EBA? 88
Nuraghe Tiria (Villaurbana) Nuragic 1010
Nuraghe Loddu (Fordongianus) Nuragic 98
Nuraghe Nieddu (Oristano) Nuragic 109

total 468178
Using Rb, Sr, Y, Zr, Nb to
identify obsidian sources

High Rb relative to low Sr

Roughly equal Rb and Sr

pXRF Discrimination of Monte Arci Subsources
and Assignment of Artifacts

SC

SB
SA

multivariate statistical analysis provides even clearer distinctions

 Map of Sardinia

shows obsidian source area (Monte

Arci) and the area of Nuragic sites
with obsidian tested
 Central Marghine Region in Sardinia

large cluster of Nuragic sites occupies about 40 sq km in Borore and adjacent comuni
Nuraghi in the Marghine Area
 Excavations near Tower A at Duos Nuraghes

In the mid-1980s led by Penn State faculty Joseph Michels & Gary Webster
 Excavations done first in the village area outside of the nuraghi…

A
B
…followed by excavations within the nuraghi
 Distinctions visibly clearer using multiple elements

13

10
SA
SB1
SB2
Rb/Nb

7 SC

1
0 3 6 9
Sr/Nb
Nearly 300 artifacts tested for these 5 Marghine sites

%
100
228 71

80
SA
SB1
60 SB2
SC

40

20

0
Duos Nuraghes

Other Sites

Only about 90 artifacts

tested in total for six other
Nuragic sites
Over 200 tested from specific contexts within Duos Nuraghes
%
100 39 15 168

80 SA
SB1
SB2
60
SC

40

20

0
Nuraghe B

Village
Nuraghe A
 %

0
20
40
60
80
100
10

Tiri
8

Loddu
9

Nieddu

Duos Nuraghes
228
71

Marghine
10

Domu Beccia
Other Nuragic Sites Tested

145

Ortu Comidu
13

Antiguri
SC
SB
SA
 The Chaîne Opératoire
Reduction

Production Modification

Use Re-use

Deposition
Acquisition
(extraction)

Transport

Exchange
Greenstone and Copper Artifacts from Florida tested

Pure copper, consistent with native (natural)

copper, rather than smelted from copper ore.
From the Blueberry Site
Soil Samples
Use in lab, field, museums…
…on stone, metals, ceramics, etc.
Orlando Museum of Art – South American Collections
More Examples of Testing at the Orlando Museum of Art
Tampa Museum of Art – Classical Period Metal Artifacts
More Examples of Metal Artifacts Tested at Tampa Museum of Art
Silver object in Tampa Museum of Art
Brass object in Tampa Museum of Art
 Malta National Museum of Archaeology, Valletta

tested obsidian from residential site of Skorba

and mortuary complex of Brochtorff Circle
 Malta National Museum of Archaeology, Valletta

Tested wooden sculpture to see if Tested statue to see if modern

there was gold paint on surface... or ancient gold alloy...

...and it appears to be truly ancient

...and there was!
Analysis of Copper-Based Artifacts from Morocco, at University of Siena, Italy

Specifically testing to see whether objects were

brass (containing zinc), rather than pure copper
or bronze (containing tin)
At USF - testing human bone samples to look at diet
(differences between terrestrial plants/animals and seafood)

need to use vacuum to detect phosphorus (P) –

important for comparison with calcium (Ca),
strontium (Sr), and barium (Ba) values
Human bone sample from coastal site of Is Arrutas, Sardinia

note Sr peak
 Human bone sample from inland site of Scab'e Arriu, Sardinia

note lower Sr peak

remember, this is just the visual concentration peaks; calibrated numeric values are what is significant
Yon Mound (Apalachicola, Florida)

excavations by Nancy White & Jeff DuVernay (USF)

Fort Walton Incised

Lamar Ceramic Types

 For homogeneity, powder samples removed by drilling...

...while analyses of clean surfaces also shown to be consistent and reliable

pXRF of Yon Mound Ceramics
 Florida Ceramics
(tested samples for Tom Pluckhahn)

Differences in ceramic
composition for Sr, Zr
 pXRF of inner surface of Yon Mound ceramics (2009)

1.6
CheckStamped
1.4 Fort WaltonIncised
LakeJacksonIncise
1.2 LakeJacksonPlain
Zr/Sr

Lamar ComplicatedS
1 MarshIslandIncise
SandTemperedPlain
0.8

0.6
0.9 1 1.1 1.2 1.3 1.4 1.5
Rb/Sr
note cluster of Lamar Complicated Ware samples, while probably multiple
sources of Fort Walton and Marsh Island incised wares
 Kolomoki (Georgia)
(excavations by Tom Pluckhahn)
 Ceramic Samples from Kolomoki (Georgia)

180
13
170 93

160
Zirconium (ppm)

150
140 24
238
8 31
130 5 46
29
120
110
203
100
83
90
80
100 110 120 130 140 150
Strontium (ppm)

note: one or two central groups (local?), plus several outliers (from other sources?)
 Poverty Point & St. Johns Pottery from Florida and Louisiana
2 C h e c k S ta m p e d
F o r t W a lto n I n c is e d
L a k e J a c k s o n I n c is e d
L a k e J a c k s o n P la in
L a m a r C o m p lic a te d S ta m p e d
M a r s h I s la n d I n c is e d
S a n d T e m p e r e d P la in
1 .5 C la ib o r n e - P P O
C la ib o r n e - S t. J o h n s
P o v e r ty P o in t - P P O
P o v e r ty P o in t - S t. J o h n s
A p a la c h ic o la - P P O
T ic k I s la n d - P P O
T ic k I s la n d - S t. J o h n s
Zr/Sr

T ic k I s la n d - W a lle r
1

0 .5

0
0 0 .5 1 1 .5
R b /S r
Many possible applications, but need to ...
... have specific purpose for the testing;
... know what elements should be tested;
... have other data to compare with
... remember that analysis is of surface (if sample not powdered)

Florida possibilities include:

... sourcing ceramics;
... sourcing recently found obsidian;
... identifying native vs. historic metals;
... composition of metal alloys (ancient and modern tools, jewelry);
... soil testing (ancient activities, modern contamination);
... seafood consumption (by testing bones)

Samples can be solid, powder, or liquid

Portable and precise...
but not inexpensive!

And does require use by

someone trained in X-
ray analysis and data
processing
 Thank you!

If interested in collaborative research projects, I can be contacted at:

813 974-7279 (office phone)

[email protected] (email)
http://shell.cas.usf.edu/~rtykot (website)

or just google “Tykot”