Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986

Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. "Accelerator Based Analyses and the Provenance of Ancient Marbles." Presented at the 25th International Symposium on Archaeometry, Athens, Greece, May 19-24, 1986.

Saved in:
Bibliographic Details
Main Author: Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D.
Format: Online
Language:eng
Created: Saint Louis University Libraries Digitization Center 1986
Online Access:http://cdm17321.contentdm.oclc.org/cdm/ref/collection/speccoll/id/1426
Tags: Add Tag
No Tags, Be the first to tag this record!
id sluoai_speccoll-1426
record_format ojs
institution Saint Louis University
collection OJS
language eng
format Online
author Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D.
spellingShingle Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D.
Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
author_facet Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D.
author_sort Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D.
title Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
title_short Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
title_full Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
title_fullStr Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
title_full_unstemmed Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986
title_sort folder 20: kusko, b.h., s. howard, t.a. cahill, c. higgins, and d. evans. accelerator based analyses and the provenance of ancient marbles, 1986
description Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. "Accelerator Based Analyses and the Provenance of Ancient Marbles." Presented at the 25th International Symposium on Archaeometry, Athens, Greece, May 19-24, 1986.
publisher Saint Louis University Libraries Digitization Center
publishDate 1986
url http://cdm17321.contentdm.oclc.org/cdm/ref/collection/speccoll/id/1426
_version_ 1798403657717252096
spelling sluoai_speccoll-1426 Folder 20: Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. Accelerator Based Analyses and the Provenance of Ancient Marbles, 1986 Cahill, Thomas A., 1937- ; Kusko, B. H.; Howard, S.; Higgins, C.; Evans, D. Proton-induced X-ray emission; Imaging systems in archaeology; Archaeometry; Antiquities -- Analysis Kusko, B.H., S. Howard, T.A. Cahill, C. Higgins, and D. Evans. "Accelerator Based Analyses and the Provenance of Ancient Marbles." Presented at the 25th International Symposium on Archaeometry, Athens, Greece, May 19-24, 1986. 1986 2011 image/pdf 67 1 20_Item 0001.pdf Thomas A. Cahill Papers--Crocker Historical and Archaeological Project, 1981-2009 The items in this folder are part of the Thomas A. Cahill Papers--Crocker Historical and Archaeological Project, 1981-2009. They are from Series 1: Thomas A. Cahill Research Papers, 1981-1994. This series consists of various research papers and published articles based upon Dr. Cahill's research using Particle Induced X-ray Emission (PIXE) techniques in analyzing inks and papers. 67 1 20 Permission to copy or publish must be obtained from the Saint Louis University, Pius XII Memorial Library, Special Collections Department Saint Louis University Libraries Digitization Center text/image eng Saint Louis University Libraries Special Collections, Archives & Manuscripts FILE COpy ACCELERATOR BASED ANALYSES AND THE PROVENANCE OF ANCIENT MARBLES B.H. KUSKO, S. HOWARD, T.A. CAHILL, C.G. HIGGINS, and D. EVANS Crocker Historical ~nd Archaeological Project, Crocker Nuclear and Departments of Art, Physics, and Geology, University of California, Davis, CA 95616 (USA) Laboratory, PRELIMINARY ABSTRACT Kusko, B.H., Howard, S., Cahill, T.A., Higgins, C.G., and Evans, D., 1986. Accelerator based analyses and the provenance of ancient marbles, J. Title. A team of investigators affiliated with the Crocker Nuclear Laboratory (CNL) at the University of California, Davis, are examining marble samples from ancient quarries to establish a data base for use in determining the provenance of marble artifacts. The purpose of this project is to determine whether use of the cyclotron and other analytic equipment at CNL, coupled with computer-based statistical analyses, can indeed assist in "fingerprinting" ancient marble specimens and identifying the quarries of their origin. The facility being developed for neutron activation analysis (NAA) with mixed-energy neutrons is described, and the results of particle induced x-ray emission (PIXE) analyses of samples from Bassae, Carrara, and Thasos are presented. 1 INTRODUC TION We are in the preliminary stages of using the facilities of the Crocker l-luclearlaboratory (CNU to study the major, minor, and trace elements composing ancient marbles. The ultimate goals of this project are (i) to determine whether a battery of non-destructive analytical tests can distinguish the products of different quarries and thus help to determine the provenance of specimens and (2) to establish a data bank containing the results of these analyses, with which additional samples of ancient marble may be compared. We are aware of the great heterogeneity of marble, and of the failure of past attempts at fingerprinting quarries, but we propose that by combining the results of complementary analytical techniques and using computer-based statistical analysis a high level of confidence concerning sample provenance can be achieved in a significant number of cases. The fact that the totally non-destructive tests we propose here can be made on not only marble chips but entire statues opens up new opportunities in the study of ancient marble artifacts. It has been postulated that the proportions of carbon and oxygen isotopes in marble are unique to a given quarry (Craig and Craig 1972). Though some degree of success in provenance studies has been achieved, it has become clear that in many cases such isotopic ratio data must be supplemented with data obtained by other methods, in order to get definitive results. It is equally clear that stud;~s of the trace elements in a marble sample, using standard neutron activation analysis techniques, cannot in most cases uniquely define its source. Yet, again, some su~cess has been achieved in a few cases (Herz et al. 1986). Both of these methods, however, are destructive of at least small portions of the specimens, which limits their use. The results of cyclotron determination of the major and minor elements with particle induced x-ray emission (PIXE) are completely harmless and can be used to complement isotopic ratio and activation analysis measurements. The cyclotron can also provide an external neutron beam, which we are using to explore ways to make neutron activation analysis of large statues truly non-destructive by eliminating the need to remove samples from the artifacts. We are presenting some preliminary results of our marble analyses here because they look promising. Our team of physicists, geologists, and archaeologists represent the areas of special knowledge necessary for a viable program in the difficult investigations concerning marble provenance. 2 FACILITIES OF THE CROCKER NUCLEAR LABORATORY (CNL) At the heart of the Crocker Nuclear Laboratory is the 76" (193 cm) isochronous cyclotron, which is used to produce ion beams with energies ranging from 4.5 to 88 Mev. Presently, beam time is used more for applied work than for basic physics research, although what looks like applied atomic and nuclear research sometimes turns out to be basic research in biology, medicine, or geochemistry. 2.1 Particle Induced X-ray Emission (PIXE) Particle induced x-ray emission (PIXE) is an analytical technique in which high energy ions (usually protons in the Mev range) are used to ionize atomic inner-shell electrons in a sample being tested, which leads to the emission of characteristic x-rays. Every element from sodium to the heaviest element on the periodic table can be detected by its x-ray signature, whether it is present in major, minor, or trace amounts. Since 1972, the Air Quality Group at CNL has been using PIXE with 4.5 Mev protons to analyze trace elements of atmospheric aerosol samples. In 1977 Professors T. Cahill in Physics and R. Schwab in History decided to try PIXE to investigate the chemical composition of papers and inks of old books and manuscripts. They had to modify the target area in order to bring the proton beam out of the evacuated beam line and into the air, so that the analysis of lnrge documents could be carried out safely, without the need for removal of a sample (Cahill et al. 1981). External beam PIXE--what we call the proton milliprobe--turned out to be ideal for the elemental analysis of rare and fragile manuscripts and printed works (Kusko and Schwab 1986). The new information gained by PIXE analysis of, for example, the Gutenberg Bib1e has given incunabula sc,holars insight into Gutenberg's printing shop they never thought possible (Needham 1985). Over 3500 analyses of the papers, parchments, inks and pigments of old books and manuscripts are now on file in Davis. Preliminary attempts using non-destructive PIXE techniques were made to analyze other objects of interest to historians and archaeologists--clay, glass, and other ceramic objects; coins, jewelery, and other metallic artifacts; and obsidian, marble, and other mineral specimens. These materials present special problems, since the proton beam is stopped after penetrating less than one hundred micrometers of the surface of the object. The change in x-ray production with decreasing proton energy, as well as x-ray absorption-enhancement effects, make accurate quantitative results difficult to obtain. Although it is doubtful that PIXE will come to play the dominant role in any of these areas that it plays in analyzing aerosols and documents, the potential is present and it is worth exploring these uses. 2.2 Neutron Activation Analysis (NAA) When the cyclotron is used for making radioisotopes for medical purposes, a relatively intense beam of neutrons is generated from the target. This unusual neutron source is being put to good use at CNL. For example, Iodine-123 is produced by sending 18 microamperes of 68 Hev protons through a molten pool of NaI in a stainless steel container. The reaction of interest is a (p, 5n) reaction, and results in an intense flux of neutrons peaked in the forward direction, with energies ranging from thermal to 65 Mev. Since this beam is presently run for 10 hours a day, four days a week, we decided to take advantage of this inexpensive source of neutrons and develop a facility for activation analysis (Verosub et al. 1986). Figure 1 shows the target area. A rabbit tube is located directly behind the NaI target, allowing small samples to be inserted and removed from the neutron beam without disturbing the radioisotope production. Behind this rabbit assembly is an adjustable neutron collimator, which is capable of providing a neutron beam from ten centimeters to less than one centimeter in diameter. The facility was designed so that even very large objects can be tested. We are in the initial stages of determining the neutron spectrum and calibrating the system with standard samples. The mixture of low and medium energy neutrons, while making it difficult to detect the trace rare-earth elements, is sensitive to light elements otherwise not detectable with the use of reactor-produced neutrons. External beam HAA remains a good quantitative, bulk method, and it avoids the patina problems of PIXE. 7~if',)' t· "r Figure 1. The mixed-energy neutron activation analysis facility. 3 PROCEDURES AND RESULTS OF ANALYSES Two sets of key values must be determined for each parameter in any technique that is used to establish provenance of an unknown sample: (1) the mean values and standard deviations of the parameters in the unknown sample and (2) the mean values and standard deviations of the parameters in the known possible sources. By comparing these two sets of values using discriminant analysis techniques one obtains an estimate of the confidence with which the unknown sample can be assigned to a certain source. While generating the information required for the unknown specimen is relatively easy for homogeneous samples, generating the data required for the known possible sources is extremely difficult. Many areas must be tested around each ancient quarry. Layers of marble may each be somewhat different in age and composition, and the ancient site may, in fact, have been mined so heavily that little of the original rock remains (e.g. Paros). Faced with such difficulties, it waul d require a minor miracle to obtai n a clear and unique "quarry signature". However, by examining all our information and using discriminant analysis techniques, we may achieve reasonable confidence of sample provenance in a useful number of situations. 3.1 Bassae temple roof tiles The marble used for the Temple of Apollo at Bassae has been attributed by various scholars to different quarries in Greece, including Paros, Penteli, and Doliana. In 1981, N. Herz and colleagues used isotopic analyses of carbon and oxygen to rule out Paros and Penteli as sources of the Bassae marble. Two Doliana quaries were also tested; one had isotopic ratios similar to those of the Bassae samples, and one had very different isotopic ratios. The authors in fact proposed a recently discovered classical quarry on the tip of the lower Mani peninsula as the source of the Bassae marble, on the basis of its close similarity in isotopic ratios and petrographic attributes (Herz et al. 1982). Professor Herz sent us several small unpolished samples of the Bassae roof tiles and four suspected quarries, and we tested them with the external beam PIXE system. The results are given in Table 1. The minor elements potassium, manganese, strontium, and iron sometimes show surprisingly small deviations within a quarry, yet the levels vary from quarry to quarry. Of course, with such a small number of samples one must be cautious, but it appears that the source of the Bassae roof tiles was not one of the four quarries tested. TABLE 1. PIXE results of Sassae Temple roof tiles and possible quarry sources. Values are relative and uncorrected for x-ray absorption or enhancement effects. Description ~1g K Ca V Cr ~1n Fe Sr Bassai temple BS-1 <9 34 10300 <7 40 <7 40 290 roof tiles 8S-2 <9 80 27900 10 <7 <7 26 160 8S-3 <11 34 18500 11 <7 <7 51 190 Pentelicon, Pe-1 11 85 25600 <7 <7 40 100 190 Spilia Daveli Pe-2 <9 71 23200 <8 11 44 60 180 quarry Pe-3 15 132 27800 <8 <7 41 170 170 Mani Penninsula FC-8 <13 98 28500 8 <7 <7 210 140 quarry FC-9 <10 92 34600 <7 13 <7 . 170 160 FC-10 <10 67 24900 <7 <7 28 190 100 Paros, Pl-1b <11 61 25200 13 19 21 170 <20 lynchnites Pl-1e <10 56 21000 12 <7 18 36 190 quarry and Pl-2 <12 57 22000 <7 <7 <7 30 <20 mine Pl-3b <11 57 18100 8 17 28 250 Ooliana, 00-2 <12 44 21300 10 <7 <7 23 160 Peloponnesos 00-3 <11 42 16600 <7 <7 <7 13 160 00-4 <12 57 22200 <7 <7 <7 25 120 Thus, while we, like Professor Herz, cannot claim definitive association of a particular sample with a single source, we can effectively rule out some quarries and favor others. The minor elements appear to show less variability than trace rare-earth elements in a given quarry. We interpret this as a reflection on the general geochemical conditions under which the creatures built their shells and the original limestone accumulated. These minor elements are in many cases essential to life, and thus the biological processes tend to produce relatively constant concentrations linked to survival of the animal. In this regard, the minor elements behave more like isotopic ratios of carbon and oxygen than like tracers such as rare-earth elements. 3.2 Carrara quarry samples To further explore variations within a single quarry, we obtained samples from the Fantiscritti quarry in Carrara. These samples were taken systematically across a quarry face with the assistance of Professors c::, 1O~'''-------------------;I II I!I \ \ Q.) > • "T ~_. ': [.) " , ~. ~C" ),,)' A o lc iurn .•... o C1J 2 cr: 10 • o a 0• ABCDEFGH Figure 2. PIXE results of Carrara quarry samples, horizontal scan (A-H) and random locations (X1-X3)· III ii I: \1 ~~C1aS~iurt, I' I:: I V" 1rIron II' I" \:;! ~'.:, "b " ISh 0 n t iuIT' II 1 ~II I .•... o .••...... "- OJ U ~ 103 U! .y-o 0:; 2 0: 10 12345678 Figure 3. PIXE results of Carrara quarry samples, vertical scan. H. Franz; ni and L. Leoni of the University of Pisa. Eight samples were taken horizontally along a stratum at intervals of approximately 70 centimeters, 8 were taken vertically across the strata at intervals of approximately 35 centimeters, and 3 samples were taken in random locations. The results of external beam PIXE analyses of the rough chips are shown in Figure 2 for the horizontal scan and three random locations and in Figure 3 for the vertical scan. The results are presented on a relative scale--i.e., no absorption-enhancement effects are included. Although potassium and strontium are fairly constant, iron varies by a factor of more than 25. Repeated measurements of different faces of samples 7 and G show excellent reproducibility for all elements except iron. It is very interesting that in the vertical profile the iron values appear to be periodic, and anti-correlated with strontium. We need to explore the significance of this more fully. The samples are now being polished for electron microprobe analysis at the UCD geology department, and they will be returned for PIXE re-analysis to determine what effect surface roughness and patina may have on the analytical results. 3.3 Thasos quarry samples A similar intraquarry comparison was done with samples from Thasos. C. Lahanier of the Louvre ~'useum brought these samples, in powder form, to Davis for analysis. Twenty-one samples were taken vertically across the strata. They were prepared for analysis by applying a thin layer of powder onto the sticky coating of a sheet of adhesive-backed mYlar, and they were analyzed by PIXE in vacuum, external beam PIXE, and x-ray fluorescence (XRF). Figure 4 shows the results of PIXE analysis in vacuum. The feature that stands out is the sharp jump in magnesium values--by a factor of 100--as the scan evidently goes from calcite to dolomite and back to calcite. Potassium values vary by a factor of 3 and manganese by a factor of 2; iron varies by a factor of 10, but again, the values appear to be periodic. Strontium was below the minimum detectable limit for these PIXE runs; the XRF results show a weak anti-correlation with the iron values. 4 CONCLUSION To date we have done 257 PIXE analyses, 85 XRF analyses, and 2 NAA analyses of a variety of marble specimens. We have concentrated our efforts on analyzing large unpolished chips, but we have also analyzed powdered samples in air and under vacuum. As noted, some of the results are encouraging and some are disappointing; we plan to continue our modest level of research as time and money permit. Eventually we want to do only non-destructive analyses, but since PIXE is essentially a surface method, patina and texture effects may require 10G.-----------------~--------_. l( x )l. )( l( x.x Pc,t=.isSiun, 1 5 10 15 Figure 4. PIXE results of Thasos quarry samples. small samples of the core to be removed. On the basis of early results, we see that analysis of patinas with PIXE can be valuable in studying marble; for now, patina is a handicap to bulk characterization. We are as yet far from being able to determine the provenance of ancient marbles, and it may never be possible to determine their origins with complete certainty. Nevertheless, the results shown here are promising. These accelerator based analyses, when combined with other analytic and statistical techniques, will provide useful information so that we may make progress in these invlstigations. 5 ACKNOWLEDGMENTS We would like to thank S. Margolis and P. Schiffman of the UCD Geology Department for useful discussions. Funds for cyclotron beam time were obtained from the Crocker Nuclear Laboratory. 6 REFERENCES Cahill, T.A., Kusko, B.,and Schwab, R.N., 1981. Analysis of inks and papers in historical documents through external beam PIXE techniques. Nuc. Instrum. and Methods, 181: 205-208. Craig, H. and Craig, V., 1972. Greek marbles: Determination of provenance by isotopic analysis. Science, 176: 401-403. . Herz, N., Cooper, F.A., and Wenner, D.B., 1982. The Mani quarries: Marble source for the Bassae Temple in the Peloponnesos. Am. J. Archaeology, 86, 270. Herz, N., Wenner, D.B., Robinson, H.S., Grimanis, A.P., and Vassilaki-Grimani, M. 1986. Geochemical tests for the association of limestone statuary: the Pan Head of the Cleveland Museum and the Miltiades statue of the Acropolis Museum. (These proceedings). Kusko, B.H., and Schwab, R.N., 1986. Historical analyses by PIXE. Invited talk, Fourth International Conference on PIXE and its Applications. June 1986, Tallahassee, Florida (USA). Proceedings to be published by Nuc. Instrum. and Methods. Needham, P. 1985. The paper supply of the Gutenberg Bible. The Papers of the Biblio. Soc. Am. 79:3 303-374. Verosub, K.L., Casteneda, C., Cahill, T.A., Young, J. and McGie, M., 1986. External beam activation analysis with mixed neutron energies. Presented at the 1986 Meeting of the Geophysical Society of America, San Francisco (USA). http://cdm17321.contentdm.oclc.org/cdm/ref/collection/speccoll/id/1426

Similar Items