The new external microbeam facility of the Oxford nuclear microprobe and its application to problems in archaeological science

436 I!; ' I ' : i| i ' k /^aUPTMANM ^9] Keesmann. I.. 1989, Chemischc und m • - tation eisenreicher Schiacken. In R p. nciailuntcrsucluiniien zur 1987. Symposium Lihlice Í9H7. Praguc of ifO'^' Í20] Craddock,PT., 1995.£fír/v/V/ív«/A/,»,,,„ , 1211 r^i vrif n I - * " l*rn(im ti(>n. fuJinhuruh '■''i Gale, N.H., Papastamataki. A.. Stos-Gale y . " rce- and Copper Metallurgy in the Acgean Bro— ^ CoppC w Antiquity, fed. PT. Craddock and ""<1 Snwltinfi Technology 81-101. ^22] Levy, T., Adams, R.A. and Na|j „ Hughcs). Brir. .Musi:um C)cc. ' Jordán: The Jaba! Hamrat Fidan (^krdí!Tu Homepage. Arduu-nlo^y Projcc. 1999. Internet- Kólschbach,S., 1999. Expcrimcnic zur .Simni. . . . . ■y.i.- Verfahrenstechnik von Winddfcn. Priili'sK.rischer Kuprcrgcwinnung. Wissenschaften, Rhcin.-Westral. T-cchn M ""u «-''"ghau. Hiittenwesen und [241 D , "• ^'<»thschulc Aachcn. ' oachmann, H.G. and Rothcnbcrg. B.. 19x0 In ^ri/cesFupfer/,n77nwa-ra/. (cds.H.G Conr von S-te heft J, Bochum, 215-236. ^ Rolhcnbcrg). Dcr Anschnitt. Be»- Hauptmann, A., Lutz, J., Pernicka. f£. yp.i v. 1 • , , -y der fr iihesten Kupferverhüttung im Vordcrcn Oricnt ' í'"u ii Teclinolog» Mountains-Archaeologica Anatolica et M " ""iTn ver! ciío r o . r, / I tr '-^''poiarnica Alba Painiicn Dedicata Univer Sita La Sapienza, Rom, ícd. M. Frangipane H w. \4 1 ;, 'tau - « M. Mellink^ 541-572. Hauptmann, M. I.i^crant, P. Mntlhiae & New ¡=™r' A . B^'^h-nann H.G, ttnd Maddin. R,. |yy„, c halcl.thic Cpper Smefing; AM r ?T h . Archnc.mciry (ej s. Demirci. Ozer and G.D. Summcr.s),Tubitak. Ankara 3 jo Haup'^nn.A.andRoden.C.. '''««■ Archa«mciailurg¡sche Untcrsuchungcn zur Kupfcr- N.H., Bachmann, HG.. Rchenberg. B., Stns-Cútlc. Z.A. and Tylccttte R F.. 1990, ne Adventitiou.s Produclion of Iron in ihc .Smeliing al Coppcr. In Resmn hes in 'hn ^abah 1959.1984. V„l. II.-The Aneen, Me,.,lu„.^, „, c- In.,-,. A,-cl,.,eo-McU,ll■ S'"' (ed. B. Rothenberg), London, 182-191 . Roilieriberg, B. and Shaw, C.T., 1990. The Discovery <,f a Oippor Mine and Sme"®'' m Ihe End of Ihe Early Bronze Age (EB IV) in the Timna Vallcy. In.st. Archaeo-Me- • Studies, Newsletter 15/16, 1-8. ^^"Ptmann. A., 1989, The Earlie.st Period.s of Copper Me.allurgy in Feinan/Jordan. I" « ^orld Archaeometallurgy, íed: A. Hauptmann. E Pernicka and G.A. Waener). C>er ^nschn¡tt,Beiheft7 .Bochum, 119-135. [27] [28] [29] [30] [3l] p. VCU.y, I7/0, -y [32j Israel Explor. Soc., Jeru.salem. ®=''-Arieh, I., 1983, Cenlral-Southem Sinai in the Early Bronze Age II and ,,, Reía ^"■ran, R. (ed,), 1978, Earty Arad: The Chalc„u,hu Seniemen, and Early Bronze Age ^"y- Israel Explor. Soc., Jerusalem. ^'•-Arieh, I., 1983, Cenlral-Southem Sin 'dnship with Palestino. Levan! XV, .39-48. THE NEW EXTERNAL MICROBEAM FACILITY OF THE OXFORD NUCLEAR MICROPROBE AND ITS APPLICATION TO PROBLEMS IN ARCHAEOLOGICAL SCIENCE G.W. Grime*'^ Abstract Recent developments of the extemal beam facility of the Oxford Nuclear Microprobe have led to an enhancement of the capabilities of the instrument for analysing large or sensitive objects in air with a spatial resolution of 50 to 100 (xm. Used in conjunction with the 1 pm resolution in-vacuo facility this provides a unique elemental analysis facility which is being applied to a number of archaeological problems. This paper describes briefly the capabilities of the faci lity and outlines the range of applications. 1. INTRODUCTION Ion Beam Analysis (IBA) has a well-established role in the characterisation of archaeological and historical objects. The combination of trace element sensitivity and quantitative accuracy offered by the techniques of Proton Induced X-ray Emission (PIXE) or Rutherford Backscattering (RBS) presents a solution to many problems involving the study of artefacts and raw materials. Using a suitable micro-focusing system, IBA can be carried out with a spatial resolution of the order of 1 pm [1], but this is often inappropriate for archaeological objects. This is due to the nature of the ma terials, which are often heterogeneous, containing particulate inclusions with diameters of less than 100 micrometres. The long penetration depth of MeV light ions averages the signal over the range of the ions (tens of micrometres) and so unless thin samples are used, it is not possible to analyse small inclusions separately from the matrix and the advantage of using micron or sub-micron beams is lost. Another practical disadvantage of high resolution microbeam facilities for archaeological applications is the requirement to place the sample in an evacuated chamber, which for most archaeo- University of Oxford. Deparlment of Materials. Parks Road. Oxford. OXl 3PH. UK 438 íi i ilüf 11 ! '<. 1^1 N G.W. logical or historical objects requires sampling, which may be highiy undesirable for precious, delicate objects. 2. THE OXFORD EXTERNAD MICROBEAM FACILITY The long range of MeV ions in air presents the possibility of analysing large or fragüe objects in air with no sampling required. The beam diameter is degraded by collisions with air molecule (by approximately 10 pm per mm of air path), but as discussed above, this is not a major problem for archaeological samples and by careful design, a spatiai resolution of less than 50 pm can be achieved. For the majority of sample ion beam analysis in air is non-destructive; the presence of air or helium around the sample disperses any localised heating cause by the beam, and so prevents any thermal damage. In organic materials, decomposition of large organic molecules can III Congreso Nacional de Arqueometría 439 x-ray detector (light elements) Beamiine \ (High vacuum) Partióle detector (below) Magnetic deflector ;^^Sample Exit nozzle Gamma ray detector absorber x-ra/detector (trace eie/wddtsj j . ihe V ^^hematic plan view of the external beam analysis facüity al Oxford. The beam emerges frotn Q "^of the beamiine through a 300 pm diameter hale in a 2mm diameter copper nozzle sealed Qf^ . ^ hm) plástic foil. X-rays emitted from the sample are detected by two detectors mounted aí ray /• ^he beam. One detector isfitted with an x-ray ab.sorber to kiü the intense low energy has ensure good detection limits for the trace elements while the other detector the °^^t)rber and is fitted with a magnetic deflector to ensure that high energy protons do not reach for detector for gamma rays is mounted at 90" to the beam. Below the beamiine. a detector ple'^^^^'^'tS protons records the RBS spectrum and allows the total amount of charge falling on the sarn- ^^mtored. Not shown in this diagram are a video microscope which uses a mirrar to view the samni"'!'"^^ o/rte sample during analysis and a low power alignment láser to a.ssist in po.sitioning the ^ for analysis. Figure 2. A general view of the new external beam facility being used for the characterisation of surface layers on an IB"" century brome statue of Buddha. cause discoloration, but in many cases (e.g. papers) this fades over a period of days and in any case is confíned to the small area of the beam. An external beam system has been developed for the Oxford Scanning Proton Microprobe facility using design concepts described first by the AGLAE group in París [2] (see fi gures 1 and 2). The key to high performance of this type of system is to minimise the distance between the exit of the beam from the vacuum and the sample. This reduces beam scattering and also reduces the absorption of the emitted signáis, but in order to allow a free path from the sampling point to the detector. In this design the exit nozzle consists of a hollow copper cylinder with the diameter reducing in steps from 12 mm to approximately 2 mm. Copper was chosen for the nozzle material in order to reduce the gamma ray background which might result if light elements such as Al were used and the beam strikes the inner surface. The only signáis induced from copper are the Cu x-rays, which do not escape from the walls of the nozzle. The beam emerges into air through a 300 (jim diameter aperture covered with an 8 (jim Kapton 440 (y \y. GRME III CoNGRi-so Nacional db Arqueombtría 441 $ ,. r| lli i ■ 'i : í' i I : 1 , Figure 3. View through the video "ticroscope of the externa! beam focility, This shows the copper exit ttozzle projecting fr om the bottom the image and the ^tirface of a sampie of J 8"" century islamic lustreware. Fhe outline of the lustre giazed decoration can be seen as the dark regions of the sampie. ^indow. Two quadrupole lenses focus the beam through the no/.zlc, so that the amount beam striking the aperture is minimised. Note that this aiso means that the spatial '"esolution is not determined by the size of the aperture holc. The sampie is mounted 3pproximately 4 mm from the window, allowing a rcsolution of 40-50 pm to be ^chieved on the sampie. Detectors for x-rays, recoiling protons and gamma rays allow ®"3lysis using PIXE, RBS and also Proton índuced Gamma-ray Emission (PIGE) analyfor the detection of light elements such as Li and F. This allows virtually the en- Periodic tabie to be covered in a single measurement. A video microscope and a Power láser are used to align the desired región of the sampie in the beam (sea gure 3). pyjj technical details of the new extemal beamiinc will be presentad alsawhere [3]. ^ separata development, the external beam facility has been fitted with a high ^®^^sed pulsad láser to enable the surface of the sampie to be ablatad at axactly spe ^Pot as the analysis is taking place (4). This exciting development is aimed rem^' corrosión layers on ancient metáis, but the ability to dia ^"^3ce films in a well controlled manner over an area of less than 100 pm have implications for many aspect of the analysis ot ancient materials, ^ desirable to characterise separately both the corrosión film or patina and •^nderlying bulk material. 3. ^PPLICATIONS OF THE OXFORD EXTERNAL BEAM FACILITY The teriai/- applications of the Oxford external beam to archaeological masummarised below: Panel t (fun) Hg% Top 5,08 11,30 Top 5,08 11,60 Centre 5,06 15,00 Bottom 6,16 14,50 Mean 5,35 13,10 s.d. 0,54 1,92 3.1. CORROSION FILMS, SURFACE LAYERS AND BULK COMPOSITION OF METALS In addition to using PIXE to charac terise the composition of bulk metáis, thin (< 10 pm) surface layers can be characteri sed accurately and non-destructively using RBS. This provides information complementary to that which can be obtained for thick layers using the láser ablation facil ity. The Oxford extemal beam has been used to investígate a wide range of metal types, and alloys based on copper, silver, gold and iron have been investigated. In a study of a medieval altar cross [3], the thickness and composition of the gilding layers was de termined from the relative intensities of the gold, mercury and copper x-rays induced by the beam. The ratio of the gold and mercury L-lines provided the Au:Hg ratio in the amalgam gilded fi lm, and since the copper x-rays from the substrate are absorbed by the overlying gilding layer, the ratio of the copper K-line to the gold L-line pro vides a very accurate measurement of the thickness of the fi lm, as shown in table 1, in which presents the thickness and percentage of mercury recorded on the three back panels. Another example, shown in fi gure 2, involved the investigation of a I?"* century Buddha from Thailand. Although this object was bronze, the black surface had an unusual appearance. Using RBS, it was possible to determine that the surface was coated with a thin layer of organic material rich in carbón and oxygen which was diffusing into the copper. One explanation for this could be the interaction of the metal with the smoke of devotional lamps and candles over a long period. 3.2. POTTERV AND GLAZES A major advantage of the extemal beam facility in the study of pottery is the abili ty to analyse the whole object without sampling, which can be difficult and undesired on rare fragile objects. A range of analyses have been carried out both on whole objects (e.g. the characterisation of blue and white glazes on Chínese porcelain) and on fragments (e.g. the analysis of IS"* century Islamic lustre glazes). In the case of the lustre glazes, the composition of the metal surface layer can be determined using PIXE and the thickness of the metal layer can also be determined using RBS. Figure 3 shows the surface of a lustre giazed sherd viewed through the video microscope of the ex temal beam facility. The copper beam exit nozzle can be seen at the bottom of the picture and the decoration of the sampie can be seen. The dark regions are lustre giazed. Figure 4(a) shows the RBS spectmm acquired from the lustre giazed región using a primary beam energy of 3 MeV. The small high energy peak in the spectmm indicates the metal lie copper layer resulting from the lustre glaze. The total area of this peak is a measure of the amount of copper present in the film, while the shape gives an indication of the depth distribution. In this case, although the statistics were not con- 442 Q.W. Griue III Congreso Nacional de Arqueometría 443 ÍH5 195 205 2i5 225 2^^ 2-3** 2^^ 275 Kca.ojÍ ^MeVr "f instreware ^lazc rernrdcd in air nm/ií; proton.s \\ ir/¡ an iniíUil en^^'Sy iiXCfl TÍ/fZ*^/ I í* (*/t y I li f u y/ í* y/y 1 f v yi»9y/ (/«•■• /■'>•.. «7... *í/ / S'/ //// i Iíi t ¡ í/// //V / // t? ^ ^ 'neasured data is shown as dots and thc thin ¡huis a thc(fretín profiig , "'"•^"rea aai ^ndthe appeurs as the snudl peak in thc hiith eners^'v re pión of the trun¡ ig J Ofrom the glaze appear as steps at lower enerpy. The ¡arpe peak in the centre of a bg recoils from nitropen and nxypen nuelei in the air. The dala nv/.v reeorded in 90 ^Pectr^ of JOOpA. (h): The iheoretiea! sample depth distrihution used to ealeidate the "^figure 4 (a). ^lus feasibility study, there is a strong indication of a diffusion profile in the glaze, as shown by the simuiated curves, resulting in the profile shown "gure 4(b). the 3.3. *3lass and enamel beajp ^ considerations of sampling aiso appiy to glass objects and the external h3s also been applied to the characterisation not only of ancient jectjy í^rge fragments but also miniature mediaeval enamelled panels. Thc obsuch analyses has largely been to determine the nature of added constituents ^olorants. Jewp,. ■^LLERY and gemstones Th Provepa^ increasing interest in the analysis of ancient gemstones as an aid studies [5 and 6]. Many gemstone minerals do not have a fixed comp^sition, and may comprise a mixture of end-members in varying concentrations. Examples of these include gamets which have three end members and tourmalines with as many as twelve. Determining the end-member fraction allows the comparison of stones (e.g. to determine whether all stones in an object were from the same source), while the trace element content may be characteristic of the mineral deposit, although there is still a only a limited amount of comparison data available for this type of application. The extemal beam facility is ideal for the characterisation of mounted gemstones, since they do not have to be removed from the object in which they are mounted and the technique is non-destructive. The Oxford facility has been used in a number of feasibility studies, including the characterisation for the fí rst time of the tourmaline intaglio of Alexander the Great from the Ashmolean Museum [7]. 4. CONCLUSION AND ACKNOWLEDGEMENTS This brief description of the Oxford extemal beam facility has demonstrated the capabilities of in-air micro-analysis using focused beams of MeV ions. This technique offers unique advantages to the archaeological scientist for the characterisation of ancient materials and artefacts, and a brief summary of applications has been presented. This work has naturally involved the collaboration of co-workers from many different institutions, and the author especially wishes to acknowledge the contributions of the following people: Ms M.H. Abraham and Dr J.P Northover of the University of Oxford Department of Materials, Dr R. Camber of Sotheby's, London, Dr M. Vickers of the Ashmolean Museum, Oxford, Prof. M. Tite of the Research Laboratory for Archaeology and the History of Art, Oxford and Dr L. Thoreson of the Getty Museum, Los Angeles. 5. REFERENCES [11 Grime, G.W., Dawson, M., Marsh, M., McArthur, I.C. and Watt, F., 1991, The Oxford sub-micron nuclear microscopy facility. Nucí. Instr. and Meths. B54, 52-63. [2] Calligaro, T, Macarthur, J.D., and Salomen, J., 1996, An improved experimental setup for the simultaneous PIXE analysis of heavy-elements and light-elements with a 3 MeV proton extemal-beam. Nucí. Instr. and Meths. B109, 125-128. [3] Grime, G.W., Abraham, M.H., and Northover, J.P., The new extemal microbeam for the Oxford Nuclear Microscopy facility, to be presented at the 7"' Intemational Conference on Nuclear Microbeam Technology and Applications, Bordeaux, September 2000, and published in Nucí. Instr. and Meths B, 2001. [41 Abraham, M.H., Grime, G.W., and Northover, J.P., The combination of extemal ion beam analysis with controlled láser ablation for the study of thick corrosión fi lms on ancient metáis, to be presented at the 7"' Intemational Conference on Nuclear Microbeam Tech nology and Applications, Bordeaux, September 2000, and published in Nucí. Instr. and Meths B, 2001. [51 Querre, G., Bouquillon, A., Calligaro, T., Dubus, M., and Salomón, J., 1996, PIXE analy sis of jewels from an Achaemenid tomb (4"'-century-BC), Nucí. Instr. and Meths. B109, 686-689. 444 G.W. Gk/aíe [6] Calligaro, T., Mossmann, A., Poirot, J.P., and Querré. G., 1998, Provenance study of ru bíes from a Parthian statuette by PIXE analysis. Nucí. Instr. ami Mah.s . B138, 846-850. [6] Grime, G.W., and Thoresen. L.. Ion Beam Analysis of Gemstones - Investigation of a Tourmalíne Intaglio of Alexander the Great in the Ashmolcan Museum. in Archaeogenimology, (ed. L. Thoresen), Getty Museum Publications. Los Angeles, (in prcss). THE DEVELOPMENT OF SPANISH METALLURGY AND COPPER CIRCULATION IN PREHISTORIC SOUTHERN SPAIN Sophie Stos-Gale'" 1. INTRODUCTION There is no doubt that in the Spanish Copper Age the earliest technological developments of copper smelting, and the extraction of silver and tin, were based on the mineral resources of the Iberian Península. Much work has been conducted recently in Spain by scientists trying to link the trace elemental composition of prehistoric metal artefacts with the known ore sources, in particular a large amount of data on copper based metáis and silver artefacts from all districts of Spain have been published re cently by Rovira Llorens [1]. However, the archaeometallurgical survey of mineral re sources, that needs to accompany such research, is largely limited to southem Spain, in particular the south west [2]. In the eastem Mediterranean in the last twenty years there were several projects combining archaeometallurgical surveys and chemical and lead isotope analyses of ores and Bronze Age artefacts [3 and 4]. The lead isotope compositions of ores and arte facts are used as their characteristic 'fmgerprints' [5], The analytical method used currently for lead isotope characterisation of archaeological materials and ore minerals is Thermal lonisation Mass Spectrometry (TIMS). This technique provides highly accurate data for lead isotope (LI) compositions of lead present in ancient metáis, glasses, glazes and pigments, even if lead is present only at the level of a few parts per million [6]. The LI provenance studies are based on the diversity of proportions of lead isotope atoms in ore minerals related to their geological history. This feature allows to characterise geographical sources of minerals according to the pattem of the lead isotope ratios measured in them. The lead isotope characteristics of ores do not change during their smelting, carrying the LI fingerprint of the ore into the metal. The methodology of lead isotope provenance research is based on accumulating isotopic, geochemical and archaeometallurgical information about known ore deposits ox Isotrace Laboratory of the Research Laboratory for Archaeology. University of Oxford. Keble Rd. .^RH.