[PPT] - In-situ XRF analysis as a diagnostic analytical tool in the PowerPoint Presentation

SLIDE 1

A.G. Karydas, V. Kantarelou

Nuclear Science and Instrumentation Laboratory,

IAEA Laboratories, A-2444 Seibersdorf

Institute of Nuclear and Particle Physics, NCSR

“Demokritos”, Aghia Paraskevi, Athens, Greece

In-situ XRF analysis as a diagnostic analytical tool in the conservation field

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 2

Outline and the PROMET analytical campaigns across the Mediterranean

Outline

1. The PROMET project
2. The micro-XRF mobile instrumentation
2. Accuracy and pitfalls of micro-XRF analysis
3. PROMET campaigns:
Ancient Messene (2006)
Malta, Armoury Palace (2006)
Damascus National Museum (2007)
Numismatic Museum, Yarmouk University, Irbid, Jordan

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 3

PROMET FP6:2005-2008

Aim: To develop Prototype innovative and advanced analytical methods to survey large collections of metal objects in- situ, making it possible to pinpoint conservation needs without any risk of damaging the artefacts Efficient, versatile and mobile analytical methodologies: Micro-XRF and Laser Induced Breakdown Spectroscopy LIBS related tasks were carried out by Prof. D. Anglos FORTH-IESL, Crete Coordinator: Prof. V. Argyropoulos (TEI, Athens) 24 partners, including Turkey, Syria, Jordan, Morocco, Italy, France, Spain, Czech Republic

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 4

 To develop, optimize and calibrate the analytical performance of an innovative portable micro-XRF spectrometer  To develop and improve analysis procedures, protocols and the standardization of the method  To apply the micro-XRF spectrometer for systematic technological and conservation related studies of museum metal collections at the Mediterranean region:

The study of the manufacture technology of metal alloys
Non – invasive characterization of corrosion products
Contribution to the assessment of innovative protective coatings

Demokritos objectives within PROMET:

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 5

μ-XRF spectrometer: Principle of operation

X-ray tube Focusing X-ray device: Polycapillary X-Ray lensy Sample X-Ray detector Development and application of portable micro-XRF unit Customized design of ARTAX by Bruker Nano AXS

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 6

Versatility: In-situ Micro-XRF analyses

Laboratory test of TEI coupon

Numismatic Museum of Yarmouk University Irbid, Nov. 2008

X-ray Detector

Laser pointer X-ray lens Headed Eagle lapis lazuli and gold 3000 B.C. Early Bronze Age Damascus National Mus eum, Syria, October 2007

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 7

Pitfalls: Interference of XRF signal with diffraction peaks, QC/QA of micro-XRF data

6 8 10 12 14 10

1

10

2

10

3

10

4

Bragg peaks Fe

50kV, 600A, 50s

Cu Au Au #1 #2 Au

Counts Energy (keV)

 Diffraction peaks  Heterogeneity at the micro-scale  Definition of the scanning area that represents the alloy bulk composition

Alloys Filters/ Thickness (μm) Ti (23.6 ± 0.2) Co (17.7 ± 1.3) Pd (11.3 ± 0.3) Gold x x Silver x x x Copper x x

5 10 15 20 25 10 10

1

10

2

10

3

10

4

10

5

Ag Au Au Au Ag Cu Rh Au

SBL Counts Energy (keV)

SBLnorm_unfiltered_100 sec/600 A SBL_filtered_(Ti+Co)

SLIDE 8

Analytical range using He atmoshpere

1 2 3 4

200 400 600 800 1000 1200 1400

Ceramic sample Ca K Helium No Helium Si Al Counts Energy (keV)

50kV, 600mA, 100s The improvement in the intensity of Al-K and Si-K characteristic X-ray lines is significant, 22 and 7.3 times, respectively.

SLIDE 9

Analytical performance: Elemental sensitivity

600 μA 50 kV

Thin Targets ~ 50 μg/cm²

2 4 6 8 10 12 14 16 18 20 22 24 26 28 0.01 0.1 1 10

S

K-sensitivities, He atmosphere K-sensitivities, No Helium L-sensitivities, No Helium

Si Al W-Au Pt Ag-Sn Pb Co Fe Cr V Ca K Se Rb Br SrY Nb Ga Ge Ni Cl Cu

Sensitivity: cps/(g/cm

2)

Energy (keV)

Ag Sn

SLIDE 10

Analytical performance: Spatial resolution

5 10 15 20 25 30 40 50 60 70 80 90 100

FWHM (m) Energy (keV)

Filtered_Ni (25m) Unfiltered

SLIDE 11

Calibration methodology

 

1

sin 1 ) ( ) ( ) , ( ) , ( ) ( ) ( ) (              



k d k air i k i k i i k i

E E f dE F E E A E E E T E I w G E I

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 12

2 4 6 8 10 12 14 16 10

2

10

1

10

0 Thin targets

Intensity (cps/A) Energy (KeV)

Theory (FPA) (K lines) Experimental (K lines) Theory (FPA) (L lines) Experimental (L lines) (a)

Experimental/simulated pure element thick/thin elemental intensities

Andreas Karydas, ICTP, 14th of July 2015

Kantarelou et al., XRS, 2015

SLIDE 13

5 10 15 20 25

50
40
30
20
10

10 20 30 40 50

(c)

(%) Deviation Energy (KeV)

Thick targets (K lines) Thick targets (L lines) Thin targets (K lines) Thin targets (L lines)

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 0.0 0.2 0.4 0.6 0.8 1.0

(d)

Transmission (a.u.) Energy (keV)

Results of the fitting procedure

Estimated Lens transmission efficiency

Andreas Karydas, ICTP, 14th of July 2015

Kantarelou et al., XRS, 2015

SLIDE 14

4 6 8 10 12 14 16 18 20 22 24 26 28 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

(b) I J K L CNR91 CNR92 CNR152 ABQAQ ABSBL ABLLI ABKMF

Factor (Ki) Energy (keV)

2 4 6 8 10 12 14 16 18 20 0.5 1.0 1.5 2.0 2.5 3.0

620 1412 89 BAM 1831 a4 c3 d3 f3 e3

Factor (Ki) Energy (keV)

(a) 2 4 6 8 10 12 14 16 18 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4

K lines L lines

mean value of Ki Energy (keV)

Accuracy/Quantification of CH related materials

Glasses Glasses Gold/Silver alloys

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 15

Assessment of micro-XRF analysis accuracy

A. Heginbotham et al., An Evaluation of Inter-Laboratory

Reproducibility for Quantitative XRF of Historic Copper Alloys, Proceedings of the International Conference on Metal Conservation, METAL 2010, pp 178-188, Edited by Paul Mardikian, Claudia Chemello, Cristopher Watters and Peter Hull, 11-15 October 2010, Charleston, South Carolina, USA Micro-XRF (~50μm) Milli –XRF (3 mm)

Validation with respect to Cu based RMs

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 16

Semi-QA and Diagnostic Micro-XRF Analysis

Methodology:

Line and area scans to obtain in reasonable measuring time (1x1 mm2, 50 μm step, 10s/step,~1.5h) intensity maps of the detected characteristic X-ray lines Filtered excitation Analysis of corroded area vs corrosion free area Variation of the K/L or L/M elemental intensity ratios in single spot, line or area scan measurements

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 17

Semi-QA and Diagnostic Micro-XRF Analysis

Identification of the spatial coexistence of different elements,

fingerprints of certain corrosion products or of manufacture techniques.

Spatial distribution of individual elements

Results obtained:

Identification of the presence of certain minor to trace

elements that may support provenance and manufacture studies of the metal

Estimation on a semi-quantitative basis of the elements

enriched or depleted from the surface

Rough estimation of the depths that a certain element is

located, namely, on the surface, near surface (~2-10 μm) or below ~10 μm.

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 18

1 2 3 4 5 0.0 0.2 0.4 0.6 0.8 1.0 1.2

F I

#47 #9

Intensity Position (mm) Cl-K Cu-K

4 8 12 16 20 24 10 10

1

10

2

10

3

10

4

10

5

Rh Rh Pb-L Pb-L Pile-ups Sn-L+Ca-K Cu-K Sn-K CuEP Cl Cu-K

Counts Energy (keV)

# 9 # 47

Analysis of Copper coupon corrosion products

Artificially and naturally aged bronze coupon: (Cu: 91.3%, Sn: 7.5%, Pb: 1.0%) 50kV, 600μA, 30s/step,0.1mm/step, 50 measurements

#9 : green area #47: pale green area

SLIDE 19

Artificially and naturally aged silver coupon: Ag: 92% Cu: 6.5% Pb: 1.5% 50kV, 600μA, 30s/step, 0.1mm/step, 50 measurements

Silver coupon (prepared-characterized by Prof. G. M. Ingo, Polytechnico of Milano) A - Paratacamite B - Chloroargyrite C - Silver (oxide) A – Green B - White C - Black

A.G. Karydas et al, PROMET Book, 2008

Analysis of metal corrosion products

2 3 4 5 6 7 8 9 10

10 10

1

10

2

10

3

10

4

10

5

Fe + CuEP Ca Ag-L Cu-K Cl Cu-K

Counts Energy (keV)

A - Green B - White C - Black

4 8 12 16 20 24

10 10

1

10

2

10

3

10

4

10

5

Ca-K Pb-L Pb-L Pb-L Pile-ups Rh-K Ag-L Ag-K Cu-K Ag-K esc-Cu Cl-K Cu-K

Counts Energy (keV)

B A C

0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 D C B A

C u

K



A g

L



Cl-K

SLIDE 20

1 2 3 4 5 10 10

1

10

2

10

3

10

4

10

5

Cl S Ag

Black Metal

Counts Energy (keV)

Silver Bowl 1400 -1300 BC Late Bronze Age

Damascus Archaeological museum: Analysis of silver tarnishing

5 10 15 20 25 10 10

1

10

2

10

3

10

4

10

5

Ca Cl Au Rh S Ag Ag Ag Au Au Cu Cu

Black Metal

Counts Energy (keV)

Tarnish: corrosion mainly caused by the sulfur in the air

Thickness of the layer: ~ 0.5 μm

SLIDE 21

Baal Gods

 Manufacture technology (compositional analysis, raw materials)  Identification of corrosion products Late Bronze Age, 1400 B.C. Ugarit site: Issues addressed  Gilding technique

PROMET, Damascus, Syria: Gilded Bronze figurines

Kantarelou et al., JAAS, 2015

El God

SLIDE 22

Compositional analysis of bronze metal

Concentrations (wt. %) Fe: 0.26 ± 0.03 Cu: 92.5 ± 1.0 Zn: 0.60 ± 0.06 As: 0.10 ± 0.01 Sn: 6.32 ± 0.30 Pb: 0.040 ± 0.004 Cu rich corrosion Sn rich corrosion Bronze metal

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 23

1 10 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6

Ag-K/Ag-L

Theory (norm to pure Ag) Reference Alloys - Fischer El god (obj 1) Baal God (obj 4) Baal God (obj 5)

Intensity ratio Thickness / um

Compositional analysis of gold foil

Thickness of the gold foil > 10 micrometers Two (2) main compositional groups: Ag: 14-15% or 4-5% Cu: generally <2-3%

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 24

Imaging of the elements distribution on the gold foil

SLIDE 25

5 10 15 20 25 10

1

10

2

10

3

10

4

10

5

10

6

Cu Sn Fe Ca Pb

PU's K Pb Pb Cu Sn Fe Rh Ca Cu

Counts Energy (keV)

Eail God, Late Bronze Age 1400-1300 B.C.

PROMET, Damascus, Syria: Corrosion products

Malachite

5 10 15 20 25 10 10

1

10

2

10

3

10

4

10

5

Cu Sn Fe Ca As Zn Se

As Fe Zn Sn Rh Rh Se Ca Cu Cu

Counts Energy (keV)

Sn rich layer

Copper oxide

Trace elements: Zn, As, Se

5 10 15 20 25 10 10

1

10

2

10

3

10

4

10

5

10

6

Cu Sn Fe As

Fe Sn Rh Rh PU's As As Sn Cu Cu

Counts Energy (keV)

SLIDE 26

Copper-base bracelet, #216, (Ottoman Period), General + Pitting + Crevice Corrosion, Active corrosion have caused serious crack

A. Arafat et al., Journal of Cultural Heritage,

http://dx.doi.org/10.1016/j.culher.2012.07.003, 14 (3), (2013) 261-269

5 10 15 20 25 10 10

2

10

4

Rh Fe+EPCu Zn Pb+As Ni Cu Zn Pb

Cu Zn Pb Fe Ni As

Cu

Counts Energy (keV)

Cu: 76.08% Zn: 21.7% Ni: 0.80% Sn: 0.43% Pb: 0.43% Fe: 0.39% As: 0.17%

5 10 15 20 10 10

2

10

4

Cl Cu Cu Pb Pb+As PU's Fe+EPCu Zn

Cu Cl Zn Pb As

Rh

Counts Energy (keV)

PROMET at Irbid, Jordan

Umm Qais artefacts

Atakamite, Para-atakamite

5 10 15 20 25 10 10

1

10

2

10

3

10

4

10

5

Rh Pb+As Zn Zn Cu Ca Cl PU's Fe+EPCu

Cu Zn As Pb Ca Cl

Rh Cu

Counts Energy (keV)

Copper oxides

SLIDE 27

Compositional analysis of different typology copper based artifacts through time Surface characterization of high tin bronze mirrors Combined microXRF and LIBS analysis for enhancing in-depth elemental distribution

PROMET in Ancient Messene - Objectives:

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 28

Dynamic combination of micro-XRF and LIBS

LASER

Spectrograph

Fiber

Mirror

260 270 280 290 300 310 320 330 340 3x Pb Cu As As As Cu Cu Cu

}

Cu Cu

LIBS

X-ray Detector Polycapillary

5 10 15 20 25 30 10 1 10 2 10 3 10 4 10 5 10 6 10 7 Sn-K Sn-K Rh-K Cu-K Rh-K Pb-L Pb-L Pb-L Zn-K Zn-K Cu-K Cu-SUM peaks Sn-L Ni-K

Counts Energy

RC36/11 Brass 50 kV, 600 A, 150sec Unfiltered radiation

μXRF

V. Kantarelou et al., METAL-07, Vol. 2, Innovative investigation of metal artifacts, pp. 35-41, (2007)

Andreas Karydas, ICTP, 14th of July 2015

Courtesy of D. Anglos

SLIDE 29

0.77 1.0 0.55 0.33 0.11 0.84 1.10 0.60 0.36 0.12

Cu-K

Position (mm) Position (mm)

30000 60000 90000 120000 150000 180000 210000 240000 1.0 0.77 0.55 0.33 0.11 0.84 1.10 0.60 0.36 0.12

Sn-L

Position (mm) Position (mm)

500 938 1375 1813 2250 2688 3125 3563 4000 1.0 0.77 0.55 0.33 0.11 0.84 1.10 0.60 0.36 0.12

Sn-K

Position (mm) Position (mm)

200 288 375 463 550 638 725 813 900

1.0 0.77 0.55 0.33 0.11 0.84 1.10 0.60 0.36 0.12

Pb-L

Position (mm) Position (mm)

8750 17500 26250 35000 43750 52500 61250 70000

20 pulses Micro-XRF elemental mapping

f the LIBS ablated area

W2 (%): Cu: 82.1, Sn: 6.9, Pb: 10.7

PROMET at Ancient Messene

SLIDE 30

1.9 3.3 2.3 1.7 1.0 0.3 3.0 2.6 1.1 0.4

Cu-K

Relative x-position (mm) Relative y-position (mm)

1.1E5 2.2E5 3.4E5 4.5E5 5.6E5 6.8E5 7.9E5 9E5 3.0 2.3 1.7 1.0 0.3 2.6 3.3 1.9 1.1 0.4

Pb-L

Relative x position (mm) Relative y postion (mm)

4400 8800 1.3E4 1.8E4 2.2E4 2.6E4 3.1E4 3.5E4

-XRF analysis of High Tin Bronzes

PROMET at Ancient Messene

0.3 1.0 1.7 3.0 2.3 1.9 3.3 2.6 0.4 1.1

Sn-K

Relative x position (mm) Relative y position (mm)

5E2 9.4E2 1.4E3 1.8E3 2.3E3 2.7E3 3.1E3 3.6E3 4E3

SLIDE 31

Ancient Messene, Greece: Manufacture of high tin bronze mirrors

Mirror 1 (M1) (Cu: 70.6%, Sn:26.0%, Pb:3.4 % )

Mirror 2 (M2)

(Cu: 70.2%, Sn:22.9 %, Pb: 6.8 % ) Examined areas : metal, black, silverish Examined areas : metal, black, silverish, green and light grey

2nd c. BC

SLIDE 32

Clean Silverish Black 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2

M1

Relative Intensity (a.u.) CuK SnK SnL PbL Clean Silverish Black Green LightGreen 0.0 0.5 1.0 1.5 2.0 2.5 3.0

M2

Relative intensity (a.u.) CuK SnK SnL PbL

Silverish: Increase of SnLα, decrease of CuKa and PbLa : Sn surface enrichment Black : Significant increase of SnLα (Cassiterite?) Green : Increase of CuKa, Cu corrosion products (malachite?) Light green : Increase of PbLα, SnLα, decrease of Cu-K, (Lead-white? Cassiterite?)

Ancient Messene, Greece: Surface examination of high tin bronze mirrors

SLIDE 33

2 4 6 8 10 1 2 3 4 5 Black Silverish

Pure SnO2 layer SnO2+ Cu 31Sn8 Relative Intensity (a.u.) Layer Thickness (m)

Sn-Lα

Ancient Messene, Greece: Black patina on high tin bronze mirrors

Silverish < 1μm Black layer with a thickness of very few microns

SLIDE 34

 Gilding technology  Authenticity issues  Compositional analysis of armory components (rivets)  Identification of surface corrosion products Conservation related issues addressed:

PROMET at Armoury Palace, Malta

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 35

0.9 0.6 0.3 1.4 1.8 1.0 0.6 0.2

Cu-K

Relative x position (mm) Relative y position (mm)

2250 4500 6750 9000 1.125E4 1.35E4 1.575E4 1.8E4

0.9 0.6 0.3 1.4 1.8 1.0 0.6 0.2

Fe-K

Relative x position (mm) Relative y position (mm)

1E5 2E5 3E5 4E5 5E5 6E5 7E5 8E5 9E5

0.9 0.6 0.3 1.4 1.8 1.0 0.6 0.2

Hg-L

Relative x position (mm) Relative y position (mm)

1.5E3 3E3 4.5E3 6E3 7.5E3 9E3 1.05E4 1.2E4

0.9 0.6 0.3 1.4 1.8 1.0 0.6 0.2

Au-L

Relative x position (mm) Relative y position (mm)

8750 1.75E4 2.625E4 3.5E4 4.375E4 5.25E4 6.125E4 7E4

Gilded Iron Alloy Falling Buff

-XRF analysis of gilding areas

PROMET at Armoury Palace, Malta

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 36

The mean Au/Hg values were deduced among a subgroup

f

the area map spots, where the Au intensity varies between 0-10%, 10- 20%, 30-40% etc with respect to its maximum value

Fire-Gilding technique: Hg to Au ratio

C. Degrigny et al. , Metal-07, pp. 26-34, (2007)

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.1 0.2 0.3

Hg-L/Au-L Au-L relative intensity # 6 # 7 #14

Palace Armoury, Malta

SLIDE 37

Conclusions

Micro XRF analysis can offer fast elemental distribution maps, contributing thus both towards the identification

f surface corrosion products and manufacture techniques

as well The operating conditions of the micro-XRF spectrometer require careful optimization per type of samples analyzed, that it is in many cases not a trivial and straightforward procedure.

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 38

PROMET Impact

The results of the project offer a cost effective

approach in identifying the conservation problems and needs of a metals collection using portable diagnostic techniques

The achieved results can easily be applied to any

museum setting world-wide due to the portability

f the instruments developed whereas analytical

methodologies are easily transferable.

Andreas Karydas, ICTP, 14th of July 2015

SLIDE 39

Acknowledgements

TEI, Athens, Prof. V. Argyropoulos (coordinator) , M. Giannoulaki IESL-FORTH, D. Anglos, A. Giakoumaki Ancient Messene, Society of Messenian Archaeological Studies

Prof. Dr. P. Themelis, Director of Excavations

Museum of Ancient Messene, S. Polymenea, Conservator National Museum at Damascus, Syria Luda Mahfoud, Abeer Kurdab, Mayada El Saadi and Kasem Yahya Directorate General of Antiquities and Museums, Mr Maher Azar Malta Heritage, Dr. C. Degrigny, Dr. S. Golfomitsou, D. Vella Yarmouk University, Dr Manar Bani Hani, Lina Khrees, Prof. Mohammed S. Shunnaq,

Prof. Ziad Al-Saad

Royal Scientific Society of Jordan, Dr Abeer Arafat and Umm Qais Museum, Mohammad Bashabsheh