LASER ABLATION COuPLED TO MASS

YOCOCU: Contribute and role of youth in conservation of cultural heritage

© ISBN: 978-88-97484-01-1

Laser Ablation coupled to Mass Spectrometry (LAMS) applied to the Cultural Heritage Alfio Torrisi*, Lorenzo Giuffrida**,***, Francesco Caridi****, Tiziana Serafino*****, Eligio Daniele Castrizio******, Guglielmo Mondio*****, Lorenzo Torrisi** *

Dipartimento di Fisica, University of Catania, Italy ([email protected])

**

Dipartimento di Fisica, University of Messina, Italy

***

INFN-LNS of Catania, Italy

****

Facoltà di Scienze M.F.N., University of Messina, Italy

*****

Dipartimento di Fisica della Materia e Ing. dei Mat., University of Messina, Italy

****** Dipartimento Scienze dell’Antichità, University of Messina, Italy

Abstract

A Nd:Yag pulsed laser (3 ns pulse duration, 532 nm wavelength, 15-150 mJ pulse energy, 1-10 Hz repetition rate) is employed to irradiate ancient bronze coins placed in vacuum (10-6 mbar). During Laser Ablation a Mass Quadrupole Spectrometry (MQS) permits to detect the ablated masses from the sample surfaces with less than 1 amu mass resolution, in a range of 1-300 amu, with a sensitivity of the order of 1 ppm. Coins from IV – X century A.D. coming from different Mediterranean sites (Egypt, Greece and Italy) are ablated, in a controllable manner to remove about 1 mm per laser shot from 0.5 mm2 spot size. During laser ablation the removed atomic and molecular species are analyzed with MQS relatively to the elemental composition, chemical compounds and isotopic species. The elements of interest are those of the surface and bulk composition, preliminary checked out through XRF analysis. The chemical compounds concerns the main species of the patina and of the first bulk layers. Pb isotope ratios measurements on the lead contents in the coin patinas have been compared with literature. The used technique, the obtained measurements and the result correlations are presented and discussed. Keywords: Laser, Ablation, Mass-spectrometry, Coins

Introduction Ancient bronze coins coming from Mediterranean basin are of special interest in numismatic and archeology in dependence of the historic period of their coinage and use. The archeological bronze generally is based on an alloy containing: copper, lead, iron, tin and zinc. These coins are the best indicator for tracing the ore sources and the trading of the huge amount of copper used to realize the bronze alloy, thus their investigations permit to improve

223


© ISBN: 978-88-97484-01-1

the archeological knowledge. Recently the use of the physical technique so called LAMS (Laser Ablation coupled to Mass Spectroscopy) can be employed to analyze the bronze elemental composition both in the bulk and in the superficial patina, to detect special chemical compounds vs. depth and to measure the isotopic ratios of some peculiar elements, such as that of bulk elements content in significant amount (>1%) [1]. Particularly interesting is referred to the Pb content of the bronze alloy, that can be different depending on the geographic area and on the historical period of coinage and to the lead isotopic ratios that, also minimal differences, permit to characterize the specific mineral used and sometimes to go back to the geographic area of the origin mine [2, 3]. This article want to report mainly about the powerful of the LAMS technique used for the physical analysis and secondarily about the possible archeological interest of the obtainable results for the useful comparison between the analyzed samples and the correlation between the experimental data and the literature. The analysis is applied to some coins from IV – X century A.D. coming from different Mediterranean sites: Egypt, Greece and Italy.

Material and Methods The set of analyzed coins are the reported in the photos of Fig. 1. Catania, Sicily, Italy (VI-VII sec

Reggio

Syracuse,

Syracuse,

Syracuse,

Egypt

Calabria, Italy

Sicily, Italy

Sicily, Italy

Sicily, Italy

(IV sec AD)

(X sec AD)

(VIII sec AD)

(VII sec AD)

(VII sec AD)

Side B

Side A

AD)

Alexandria,

Fig. 1: Catalogue of the six analyzed samples.

Each coin, after having been cleaned in an ultrasounds bath, was directly placed in a high vacuum chamber (10-6 mbar) without any other special preparation of its surface. Characteristic-X-ray fluorescence (XRF) technique, induced by 20 keV electrons, was employed to identify the elemental composition of the sample surface. A Nd:Yag laser, operating at 532 nm, with 3 ns pulse width, 1-150 mJ settable pulse energy, single pulse or 1-10 Hz repetition rate, was employed. The laser was focussed on the target

224


© ISBN: 978-88-97484-01-1

(sample) up to a spot size of 0.5 mm2. A preventive calibration on a bronze alloy permitted to evaluate the ablation yield in terms of removed thickness per laser shot. The calibration uses a surface profile meter to measure the ablation depth with a resolution of 10 nm [2]. Generally analysis are performed at the pulse energy of 30 mJ at which the ablation yield is of about 10 nm per laser shot. A mass quadrupole spectrometer (MQS), Pfeiffer Vacuum-Prisma 300, with 1-300 amu mass range, mass resolution less than 1 amu and sensitivity less than 1 ppm was employed to measure the mass of the particles emitted from the laser ablation process. Depth profiles can be measured operating with a laser ablation at 10 Hz repetition rate and analyzing, contemporary, the mass yields of the element of interest. A special attention is given to the detection of the masses of the four Pb isotopes: Pb204, Pb206, Pb207 and Pb208.

Results Fig. 2 shows a typical X-ray spectrum of the analysis relative to the sample #2. Tab. I shows a synthesis of the XRF results about the elemental composition (percentage of atomic composition) of the surface patina of the investigated coins.

#2

Fig. 2: Typical XRF spectrum of one analyzed bronze coin (#2).

225


© ISBN: 978-88-97484-01-1

Table 1: Comparison of the XRF analysis of the six analyzed samples

Element

#1

#2

#3

#4

#5

#6

Al

-

0.94

5.57

0.09

-

1.40

C

52.25

39.59

21.92

36.18

44.33

37.20

Ca

0.87

1.93

0.22

0.43

0.11

4.69

Cl

0.39

2.74

1.44

0.39

1.35

0.10

Cu

4.78

7.49

5.09

4.05

13.48

1.45

Fe

0.20

0.51

7.63

0.06

-

0.21

K

-

0.70

1.31

-

-

0.14

Mg

-

0.70

0.11

-

-

0.21

Na

-

-

0.39

-

-

-

O

35.84

37.26

49.35

52.71

37.64

49.83

P

1.81

0.14

-

1.21

0.67

0.14

Pb

2.52

0.08

-

0.66

-

0.05

S

-

1.92

-

-

-

-

-

-

-

1.48

4.52

Se Si

0.81

5.99

6.64

1.40

0.94

Sn

0.56

-

-

2.85

-

Ti

-

-

0.08

-

-

0.06

Zn

-

-

0.29

-

-

-

100%

100%

100%

100%

100%

100%

The elemental composition of the patina contains both elements of the bronze bulk (Cu, Pb, Fe, Sn and Zn).and many elements of contamination, with high concentration of C, O and Si that mask the real bulk composition. Fig. 3 shows a typical LAMS spectrum, in the mass range 100-300 amu, of the analysis relative to the sample #1. The spectrum is complex and shows many chemical compounds based on copper, iron, tin, lead, oxygen, sulphur, chlorine and nitrogen. A special attention was given to the lead compounds detected in the patina of the six investigated coins as reported in Tab. 2. The higher lead concentrations is found in the samples of Syracuse #5 and #6.

226


Ion current (x10-10 A)

4 3 2

© ISBN: 978-88-97484-01-1

Catania, VI-VII sec., 30 mJ, 10 Hz, focus, φdet=90° FeOCl Cu2S SnSO4 PbSO2 Cu(CN)2 Cu2SO4 SnCl4 Cu(N3)2 C7O5H5 Cu4 SnCl2 SnO Cu4Si

1

Sn

Cu3N

0 100 120 140 160 180 200 220 240 260 280 300

mass (amu) Fig. 3: Typical LAMS spectrum, relative to the analysis of the sample #1. Table 2: Relative amount of lead compounds MQS detected in the patina of the six bronze coins

Compound PbO PbSO2 PbCl2 Pb

Mass (amu) 224 272 278 208

#1 0.06 0.009 0.10 1.5

#2 0.08 2.9 2.1 0.05

Relative Yield (x 10-10 A) #3 #4 #5 0.03 0.05 0.15 0.05 0.05 0.04 0.08 0.02 0.02 0.04 0.04 0.5

#6 0.03 0.06 0.02 0.06

The ranges of natural variation of the four lead stable isotopes are 1.65-1.04, 27.48-20.84, 23.65-17.62 and 56.21-51.28, for Pb204, Pb206, Pb207 and Pb208, respectively [4]. Thus the isotope ratios Pb208/Pb206 and Pb207/Pb206 can vary in the range 1.86-2.70 and 0.64-1.13, respectively. The lead isotope quantitative analyses are difficult, due to the possible lead mass interferences with different compounds, such as the Pb-H (molecular weight 205-209 amu) and the Cu3NH (molecular weight 204-210 amu) presents in the patina. However an accurate analysis using the maximum MQS mass resolution is possible. Fig. 4a shows an example of mass spectrum in the range 203-210 amu obtained reporting the yield as a function of the ablated mass. Fig. 4b shows an example of mass spectrum obtained by fixing the masses of interest (206, 207 and 208 amu) and reporting the yields as a function of the laser ablating time. In order to not damage the coin surfaces low times, below 60 s, are used.

227


© ISBN: 978-88-97484-01-1

Reggio, X sec., 30 mJ, 10 Hz, focus, φdet=90°

2,0

Pb/206Pb = 1.9 Pb/206Pb = 1.1

Ion current (x10-12 A)

208

1,5

207

1,0 0,5 0,0 205

206

207

208

209

mass (amu)

Fig. 4: LAMS spectra relative to the Pb yield vs. mass (a) and vs. time (b).

The evaluation of the lead isotopic ratios in the six coins was compared with the literature data concerning other bronze coins coming from different geographic areas of the Mediterranean basin relative to different historic periods. Fig. 5 shows this comparison obtained with different literature data [5, 6, 7].

Fig. 5: Comparison of the lead isotope ratios with literature data.

This comparison shows that the lead isotopic ratio measurements, although within the range of lead isotope variations, have higher values with respect to other measurements on bronze coins coming from the different areas of the Mediterranean basin. Further investigations are

228


© ISBN: 978-88-97484-01-1

in progress in order to investigate about these differences that can be attributed or to the lead employed for the coinage of the analyzed samples coming form different mines or to a not good evaluation of the isotope ratios that can be due to the contaminant compounds interference with the isotopic masses.

DiscussionS and ConclusionS The LAMS analysis can be efficiently employed to analyze the elemental composition, the chemical compounds and the isotopic ratios of ancient bronze coins. The characterization of the samples can be employed to compare the composition of the coins coming from the same or different historical periods, from different geographic areas, from the mineral compositions of different mines of the Mediterranean basin, etc.. Obtained results, mainly relative to the coin patina compositions, demonstrated that the major differences in the elemental composition is found for the sample #5 from Syracuse (VII sec A.D.) that have higher Cu content and for #1 from Catania (VI-VII sec A.D.) that has the higher Pb content with respect to the others. From the point of view of the lead chemical compounds the major differences are found for the sample #2 from Alexandria (IV sec A.D.), that has higher concentrations of PbSO2 and PbCl2. It is important to underline that some times the results of the elemental composition obtained by XRF and that of the chemical compound obtained by LAMS are not in agreement; this result is justifiable on the base of the different depth analyzed by the two techniques. The XRF analyse only the first micron of the patina coin surface (corresponding to the range of 20 keV electrons) while the LAMS analyses about 10 ns per laser shot, thus, operating at 10 Hz repetition rate, the yield is depending on the laser irradiation times. After 60 s irradiation the composition is relative to a depth of about 6 microns. From the point of view of the lead isotope ratios the major difference are found for the coin #3 from Reggio Calabria (X sec A.D.), having higher Pb208/Pb206 and Pb207/Pb206 ratios, and for the coin #6 from Syracuse (VII sec A.D.), having the lower Pb207/Pb206 ratio. Although the above discussed differences, the six analyzed coins have not strong differences between them, an observation this that can be attributed to mines of minerals from the Mediterranean basin commune to the all analyzed samples and from different populations operating in this geographic area.

References 1. 2.

L. Torrisi, G. Mondio, A.M. Mezzasalma, D. Margarone, F. Caridi, T. Serafino, A. Torrisi; Eur. Phys. J. D 54, 225–232 (2009). L. Torrisi, F. Caridi, L. Giuffrida, A. Torrisi, G. Mondio, T. Serafino, M. Caltabiano, E. D. Castrizio,

229


3. 4. 5. 6. 7.

© ISBN: 978-88-97484-01-1

E. Paniz, A. Salici, Nucl. Instr. And methods B, in press, 2010. F. Caridi, L. Torrisi, A. Borrielli, G. Mondio, Rad. Effects & Defects in Solids: Incorporating Plasma Science & Plasma Technology Vol. 165, No. 6, June 2010, 1–12. Brettscaife.net, 2010, Lead isotope analysis in Archeology, http://www.brettscaife.net/lead/data/index.html. S. Klein, Y. Lahaye, G. P. Brey, H. M. Von Kanhel, Archeometry 46, 3(2004) 469-480. M. Resano, E. Garcia-Ruiz, F. Vanhaecke, Mass Spectrometry Reviews, 2010, 29, 55-78. C. G. Eiseman, “Greek Lead”, PhD Thesis, University of Pennsylvania, 1980 http://www.penn. museum/documents/publications/expedition/PDFs/22-2/Eiseman.pdf.

230