19/09/2016

Bruker Tracer III Handheld XRF

Analysis
As a tool in the
Archaeology, Conservation
and Research

All the photographs in this presentation are

taken by Dr Bruce Kaiser and they depict a
paradigm shift that has occurred in
elemental analysis in the last 3 years.

19/09/2016

Water colors
BBHC, Wyoming

Osage Museum
Pawhuska OK

CSULB in Guatemala sourcing basalt

Indianapolis
Museum

Bronze analysis Royal Ontario

Museum

Brooklyn
Museum of Art

Rochester
Museum & Science

Soil analysis
Easter Island, Chile

Ringling Museum, Sarasota Florida

Range Creek Utah

Rock Art
Museum of Art New South Wales,
Sydney Australian

Corning Museum
Of Glass, NY

Smithsonian
Natural History Museum

19/09/2016

Art conservators in Australia

Nobel Laureates from Berkeley and Harvard

Research team at CSLB

Field School on Easter Island

Teotihuacan is an enormous
archaeological site in Mexico

West Dean College in England

Gold Artifact analysis

in Hong Kong, China

National Gallery in London England

19/09/2016

All the photographs depict a paradigm shift that

has occurred in elemental analysis in the last 3
years.

Ore analysis in Mexico

The Tracer III xrf system was specifically designed

by Bruker for the xrf analysis of non uniform
materials as a tool to be used in Archaeology,
Conservation and Research.
Because of its specific design the Tracer system is
currently in use in over 500 museums and
universities world wide and in use by over 50
different archeology research groups around the
world.

Mayan Codex Pigment Analysis

In Belize

UCSD preparing for Field work in Jordan

University of Oxford in Cambodia

Why the Tracer III over any other xrf system?

The answer is simple, most non prepared materials in museums and archeology
are non uniform.
Never ever believe numbers unless you know the
physics and your sample atom by atom
Answers vary as;
the inverse square of the distance to the
element
Exponentially relative to matrix density
Exponentially relative to elemental X ray
energy emission
Exponentially relative to element location in the
sample matrix
Exponentially relative to beam filtering and
energy
X ray beam distribution
Orders of magnitude relative to sample
uniformity
IF you can not use the numbers then you must be able to control
all the x ray beam parameters and work with the raw spectra in
detail to extract meaningful information from the xrf analysis.
The Tracer III system is the only handheld xrf system that is
designed and has the software to do this.

19/09/2016

Innovative XRF Setup and Analytical

Approaches lead to Measurement Treasures
1.
2.
3.
4.
5.

Alloying in artifacts
Ppm analysis of elements in
glass
Obsidian Sourcing
Heavy metal poisons analysis
on ethnographic collections
Measurement of imaging
agents on photographs

Objects at Harvard Foog Museum

Literally the element non destructive analysis of

anything in any form anywhere anytime, a very
fundamental paradigm shift.

Analysis of Copper Artifacts in Albania

Below are the spectra of the analysis listed on the left.
This is a look at the overlay at the Rh k backscatter part of the spectra.
It is clear that the backscatter from Blaz is much more. Also the plot on the
next page clearly shows the presence of Ca and Fe. This indicates that the
Cu is covered with foreign material, including lower mass elements like O and
C. This is a non uniform material and thus the weight percent in the
analysis is wrong on the next page for Blaz. The backscatter for Maliq is
typical of a Cu metal alloy and thus the weight percent numbers are accurate.
See next page

Rh inelastic backscatter
Rh elastic backscatter
Cu sum peaks

Continuum backscatter

19/09/2016

Analysis of Copper Artifacts in Albania

Makiq4693.1

Maliq4693.2

Blaz16391.1

Blaz16391.2

Blaz16391.3

Blaz16391.4

Maliq4687.1

Maliq4687.2

Mn
Fe
Ni
Cu
ZnKa
ZnKb
AsKa
AsKb
PbLb
RhKa
Ag
Sn
Sb

0.00
0.00
0.00
99.29
0.00
0.07
0.00
0.06
0.07
0.00
0.01
0.14
0.07

0.01
0.03
0.00
97.98
0.36
0.58
0.02
0.07
0.22
0.00
0.01
0.33
0.11

Mn
Fe
Ni
Cu
ZnKa
ZnKb
AsKa
AsKb
PbLb
RhKa
Ag
Sn
Sb

0.01
0.84
0.00
95.87
0.47
0.06
0.00
0.12
0.00
0.00
0.05
0.21
0.22

0.03
1.01
0.00
94.81
0.64
0.18
0.00
0.17
0.00
0.00
0.07
0.22
0.30

0.05
1.04
0.00
94.93
0.42
0.06
0.00
0.13
0.00
0.00
0.08
0.28
0.31

0.05
1.03
0.00
94.27
0.73
0.04
0.00
0.16
0.00
0.00
0.11
0.26
0.38

Mn
Fe
Ni
Cu
ZnKa
ZnKb
AsKa
AsKb
PbLb
RhKa
Ag
Sn
Sb

0.00
0.22
0.96
94.71
0.17
0.32
0.26
0.34
0.19
0.00
0.01
0.28
0.33

0.00
0.31
0.78
94.95
0.20
0.21
0.22
0.35
0.12
0.00
0.02
0.31
0.30

SUM

99.72

99.71

SUM

97.84

97.42

97.30

97.04

SUM

97.79

97.75

Obsidian
OBSID

PPM

K Ka1
BaLa1
TiKa1
MnKa1
FeKa1
NiKa1
CuKa1
ZnKa1
GaKa1
PbLa1
ThLa1
RbKa1
SrKa1
Y Ka1
ZrKa1
NbKa1

40,486.8
364.4
1,007.8
383.5
6,807.4
5.5
3.1
53.0
16.2
33.4
17.2
161.2
68.0
19.0
100.2
15.2

Mother natures glass that humans

used to get lunch with 6000 years ago
Obsidian at the Idaho State University

19/09/2016

Results on a XRF study of the Mosaic at the house of

Neptune and Amphitrite at Herculaneum.
Pratt, RWTH, University of Aachen, and the Herculaneum
Conservation Project

Location of analysis note numbers(zoom in to see numbers):

Glasses of the Mosaic at the house of

Neptune and Amphitrite at
Herculaneum.
biege

med dark green

orange

white

yellow
l brown

Dr Bob Brill of the Corning Museum

Of Glass, NY

red
med blue

19/09/2016

Note the colour code for the

Artax plots. This is an overlay of
all plots to verify that all the
spectra are well behaved and
the samples have similar
response to x rays thus
indicating that they were of
similar materials making
elemental relative intensity
comparison possible.

Rh L x ray backscatter
for all samples is the
same indicating the
samples are the same
matrix

25000

8 white
52 med Dark
Green
21 Beige

20000

19 orange
15000

14 yellow
13 l brown
10000

5000

0
Al

Si

Ca

Mn

Fe

Ni

Cu

Sr

Zr

Rh

Rh

Sn

Sb

Ba

Pb

19/09/2016

7000

8 white
6000

5000

4000

3000

2000

1000

0
Al

Si

Ca

Mn

Fe

Ni

Cu

Sr

Zr

Rh

Rh

Sn

Sb

Ba

Pb

How deep am I measuring?

A narrow beam of mono energetic photons with an
incident intensity Io, penetrating a layer of material
with mass thickness x and density
emerges with
intensity I given by the exponential attenuation law

Note that the mass thickness is defined as

the mass per unit area, and is obtained by
multiplying the thickness t by the density
i.e., x = t

Photograph Analysis at
the Baltimore Museum of Art

19/09/2016

Thickness of Silicon required to reduce x ray beam to 1% of

initial intensity

12
10
8
6
4

x ray energy (keV)

14

web site
http://physics.nist.go
v/PhysRefData/Xray
MassCoef/cover.html

2
0

0.001

0.010
Silicon thickness (cm)

0.100

1.000
Thickness of Copper required to reduce the x ray beam intensity to 1%
of its initial intensity

Paint Analysis at
Art Gallery of New South Whales
Sydney AU

14
12
10
8
6
4
2

X ray energy (keV)

0.000

0.0000

0.0001

0.0010

0.0100

0.1000

Copper thickness (cm)

Heavy metal poisons analysis on

ethnographic collections

0.001 Cu, .001 Ti,

and .012 Al Filter
40 kV
Poison analysis in the
Tasmanian tiger at the
4 to 8 micro amps Australian
National Museum
in Canberra
No vacuum
These settings allow

Dont feed him POSION!

All the x rays from 14 keV to 40 keV to reach the sample thus
efficiently exciting the elements Hg, Pb, Br, As.
There is little or no sensitivity to elements below Ca with these
settings reducing unneeded detector activity
Eliminates any background x rays under the peaks of interest

10

19/09/2016

Measurement of Toning Agents on

Michele
Someone
Photographs

to go
home
to!

Use .006 Cu, .001 Ti, .012 Al filter

Analyze at photograph

40kV
6 micro amps( what is available)
No vacuum
5 to 10 min in
White area where there in no toner
Areas tone varies grey to black

Take the difference (toned white)

The difference will give you a very good clean spectrum of the toning
agent. And the grey to black variation will give you an estimate of the
amount of agent. The reason this works so well are
The toning materials are very thin and have very little effect on the
spectrum from the paper

The white are is just the paper and mounting materials

The filter used removes most of the backscattered x rays

Photograph Analysis at
the George Eastman House

1500

Ba L

michele dark 1970

michele white1970
Difference
Note the red difference
Sr K
spectrum clearly shows that the
imaging agent is only Ag

Ag K

1000

Pd K

500

Ag L

Do what George says!

Fe K

Cu K

Sr K

Rh K
Elastic
backscatter

Rh K
inelastic
backscatter

Pd K

Ag K
Ba k

X ray Energy (keV)

2

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

11