STUDY OF SOME ROMAN BROOCHES DISCOVERED AT TOMIS CONSTANÞA, BY X-RAY FLUORESCENCE TECHNIQUE

APPLIED NUCLEAR PHYSICS STUDY OF SOME ROMAN BROOCHES DISCOVERED AT TOMIS CONSTANÞA, BY X-RAY FLUORESCENCE TECHNIQUE V. GHIªA 1, I. V. POPESCU 2, 3, M. BELC 1, ANTOANETA ENE 4 1 Physics Department, Ovidius University of Constanþa, Constanþa, 8700, Romania 2 Institute for Physics and Nuclear Engineering, Horia Hulubei Bucharest Mãgurele, Romania 3 Physics Department, Valahia University of Târgoviºte, 0200, Târgoviºte, Romania 4 Physics Department, Dunarea de Jos University of Galati, 800201, Galati, Romania Received May 9, 2008 XRF method was used in order to determine the concentration of the main compounds in the raw materials and on the other hand to make a classification of the different types of brooches discovered at Tomis, an archaeological site in Romania. Key words: XRF, archaeological samples, Roman brooches. 1. INTRODUCTION The main point of this paper is to study a number of 27 samples of different brooches founded in Dobrudja, more exactly in the antic metropolis of Tomis, by XRF method [1, 2], a high sensitive elemental analysis technique and nondestructive, which is used intensively in archaeological studies. This technique proves to be one of the best because of the good results and the rapidity of utilization. The brooches are some small accessories for clothes extremely used in antiquity. These are a kind of mixture between today s safety pin and on art object manufactured in order to symbolize a religious imagine [3 17]. The brooches which were discovered in Dobrudja zone, the antique Pontic-Dacian site, are part from archaeological camps, where there were discovered a great diversity of dailyuse objects, clothes, decorative ceramics, metal mirrors and great variety of tools. The opinion of the specialists from the Museum of Archaeology and History from Constanta, was that we deal with small sources of artefacts, this fact being certified by finding some local workshops specialised in producing them in small series. The conditions of a lot in the search, was running wild because of the corrosion process. In same time some of the objects asked improvement because of the high level of dismentlement. It takes us a short period of time to realise that we deal with some Pre-Roman sources, from the Greek colony-tomis, and after the Roman conquest. The very point of this research is the attemption of classifying these brooches related to the each sample, and observing from the start that we deal of four or five representative category of those artefacts, observing the shape of closing device and the level of ornamentation. Rom. Journ. Phys., Vol. 53, Nos. 3 4, P. 557 562, Bucharest, 2008

558 V. Ghiºa, I. V. Popescu, M. Belc, A. Ene 2 2. EXPERIMENTAL In our study we try to present the result of same samples measurement (from five different type of brooches), in order to investigate differences and analogies between local and abroad ancient techniques. The XRF method was used to measure the elemental composition of materials. The radioactive source, 241 Am of a ring shape was used for the excitation of characteristic X-ray radiation. The more complex character of excitation radiation had a negative impact on the growth of background in the measured spectra and the radiation s interference with analytical peaks somewhat lowered detection limit of some elements. On the other hand, this had an advantage of comparable sensitivity of determination of both metals with low energy characteristic radiation (Fe, Ni, Cu, Zn, As, Pb) and heavier metals (Ag, In, Sn, Sb) in the course of single measurement event. Because XRF-analysis is a surface method, the influence of inner heterogeneity of the artefact itself and in many cases, its corrosion of it is very serious. In order to have an accurate result, the samples were cleaned before analyse. This operation was used for the elimination of corrosive elements. It was necessary to have a smooth surface for avoiding the reflection of incidental radiation. The samples were put in order in irradiation geometry test. The 2 geometry was used here, more explicit the sample was placed in the front of the source, the radiation being spread in all directions. The samples were placed about 5 cm by the source. The radioisotope 241 Am, the excitation source, after the disintegration emits photons gamma heaving the energy of 59.5 kev. The annular source of 241 Am has an activity of 10 mci is by the best efficiency providing the energy which has the values near the biggest absorption point in the target. The element which is excited by this method is supposed, after that to be desexcited and to emit the characteristic X-ray spectrum. The X-rays spectra were detected with a thin Si with the energy resolution of 180 ev at the K line of Fe, 6.4 kev. The 1000 sec acquisition time of one spectrum was used in experiment. This will show a large spectrum of elements from the content of the sample. The Si (Li) detector used here will receive the characteristic X-radiation and can determine the elements, if the radiation energy requests it. The attached soft to the detection chain processes the spectrum and calculates the energy element concentration directly through the method of interstandardisation. The standard is composed by Cu 99.7%, Pb 0.005%, Fe 0.005%, Zn 0.005%, Sn 0.005% (MERK products). The concentrations were determined by the detection of the characteristic X-rays for the elements, using formula: RR CF( RL) CF( AT), (1) K

3 Roman brooches discovered at Tomis Constanta 559 where: is the concentration of the element, RR is the perturbation rate in X-ray detection, CF(RL) is the loss correction factor, CF(AT) is the correction factor for attenuation and K is the calibration constant. 3. RESULTS AND DISCUSSION The concentrations, in percents, of some elements determined by the XRF method for different type of brooches discovered in archaeological site, Tomis Constanta, are given in Tables 1 5, where: Type A: brooches with the shape of spur; Type B: brooches with hasp or articulated; Type BSP: plane simple brooches; Type BSC: brooches with simple circular ornaments; Type BCf: cross-form brooches; SD: Standard deviations. Table 1 Concentrations for brooches type A Sample Zn% SD Pb% SD S1 21.314 2.909 2.117 0.026 S2 15.076 1.671 1.943 0.008 S3 14.822 1.417 1.835 0.006 S4 14.658 1.253 1.692 0.004 S5 14.133 1.272 2.024 0.017 S6 14.028 1.377 3.976 0.002 S7 12.462 0.943 1.797 0.005 S8 8.747 0.658 1.029 0.102 Table 2 Concentrations for brooches type B Sample Zn% SD Pb% SD S1 0.313 1.155 1.122 0.066 S2 0.227 1.114 1.435 1.005 S3 0.109 1.102 2.621 0.012 S4 0.421 1.182 3.774 0.003 S5 1.556 1.901 5.616 0.003 S6 1.076 1.711 6.135 0.002 S7 0.193 1.112 6.382 0.002 S8 1.817 2.238 8.135 0.001

560 V. Ghiºa, I. V. Popescu, M. Belc, A. Ene 4 Table 3 Concentrations for brooches type BSP Sample Zn% SD Sn% SD S1 9.18 3.622 1.33 1.265 S2 9.07 3.608 1.48 1.289 S3 4.93 1.523 1.77 1.475 S4 7.28 2.556 2.09 1.584 Table 4 Concentrations for brooches type BSC Sample Zn% SD Sn% SD S1 4.05 1.472 12.45 2.796 S2 3.78 1.388 8.23 2.148 S3 3.47 1.354 5.76 2.008 S4 2.62 1.108 4.47 1.817 Table 5 Concentrations for brooches type BCf Sample Zn% SD Sn% SD S1 2.15 1.022 4.12 1.587 S2 3.8 1.388 9.64 2.377 S3 2.81 1.140 9.53 2.346 S4 4.46 1.613 11.02 2.606 After the determination of the concentration through XRF and also through a metallographic interpretation of the structure, we observe that the brooches can be divided in two main types according to the localization in time. The simple brooches, manufactured from a single part, are from the Pre-Roman civilization and those manufactured in a complex way, rich ornamented, are from the Roman civilization. We also observe that it was a superior technology in roman period, because of functional quality. After researches, we observe that all Iron Age samples were manufactured by bronze and Sn alloys. This type of alloy was often used in that period. The Latene civilization brooches had their origins in the Pre-Roman period but some types continued to be produced after the Roman conquest. Those are divided in two different categories; the first one contains the samples with more than 10% of Zn, and the others made by Sn, with insignificant traces of Zn. In all the analysed samples Pb is present in small concentrations, especially in the type A. Other types of brooches are those, which have a lot of ornaments on the main rod. The standard brooches, with no ornaments have arched needle, extremely flexible, being manufactured from

5 Roman brooches discovered at Tomis Constanta 561 relatively reached content of bronze. Types of brooches with oblique preeminences on the rod, were manufactured from bronze with a low level of Zn. A lot of the samples with fixed needle are made from bronze with a content of Pb (6 7%) in mean. The samples with mobile needle are manufactured by pliant composition: brass or bronze with a low level of Pb. These can be divided in two small groups based on the methods of manufacture. The plane ones are from brass with Zn (4 10%) and don t have a significant level of Pb (1%). Those, which have a special ornamentation, contain Pb (4.6%) in mean. The brooches with the shape of spur (type A) can be divided in two groups. The first group is those with discs on the bow, they are manufactured from brass with (10 20%) of Zn and with a low level of Pb. The second group is formed by same samples which are not very different by the others but which have their surfaces better made in the curved way. The brooches with hasp, or articulated (type B), need to have a rich level of Pb and the bronze presents a low concentration of Zn. The rich ornamented brooches appear in the same time with the other, but they keep on being used to the end of the 3rd century B.Ch. The simplest classification of these brooches would be: plane brooches (BSP), with simple circular ornaments (BSC) and cross- form brooches (BCf). The first type of these samples has a high contain of Zn (4 10%) and of Sn (1 2%) but the last two types have a relatively small content of Zn (2 4%) respectively (2 5%) and more contain of Sn (7.5%) respectively (8.5%), in mean. CONCLUSIONS As a result of XRF analyse it was established that there were differences of Zn and Pb concentrations for the first two types of Pre-Roman brooches. In these conditions, for the type (A), the average of concentrations was (14.4%) for Zn and (2%) for Pb. On the other hand, for type (B), it was about (0.7%) for Zn and (4.4%) for Pb, this being a demonstration of that the antiques discovered in their experience that the lower of Zn concentration and increase of Pb concentration simultaneously, leads to superior properties of the alloy, as follows: the increase of plasticity and the mechanical breaking resistance, the decrease of fragility and especially the increase of malleability degree. These properties offered a better possibility of making the hasps and catching pin. In the same time, the increase of ornamental level is an occasion to observe the tendency of Sn rising in the alloy, simultaneous with improvement of moulding techniques. It was proved in this way that the first two types are very old (Greek period), so the others some to be more recent (Roman period). The tendency of using small quantities of Zn in the alloys rich in Sn can indicate graduated increase of technologies, in touch with Roman s knowledge. Because of the raw material quality, and also the moulding and plastic distortion techniques, we can

562 V. Ghiºa, I. V. Popescu, M. Belc, A. Ene 6 conclude that all these features were due to the ascendancy of commercial and technological relationships between the Roman Empire and the Greek colony from Tomis. We can discuss the possibility that the different high-quality objects which were discovered to be brought from abroad. If it is so, they were imported from Roman s occupied zones, they representing models for local handicraftsmen. REFERENCES 1. D. Camp, W. Ruther, Nondestructive energy dispersive X-ray fluorescence. International Atomic Energy Agency Report IAEA-R-1557R, 1977. 2. P. Bertin, Principles and practice of X-ray spectrometric analysis. Plenum Press, New York, 1975. 3. Haselgrove 1997, C. Haselgrove, Iron Age brooch deposition and chronology. In : A. Gwilt, C. Haselgrove (dir.), Reconstructing Iron Age societies (Oxbow Mon. 71), Oxford 1997, 51 72. 4. B. Webster Smith, Sixty Centuries of Copper, http://60centuries. copper.org/right.html 5. R. F. Tylecote, A History of Metallurgy, Second Edition. London: Institute of Materials, 1992. 6. Allason-Jones 1995: L. Allason-Jones, Trumpet brooch from Healey. Arch. in Northumberland 1994-95, 24. 7. Bayley 1995: J. Bayley, The composition of Roman brooches found in Britain. In : S.T.A.M. Mols et al. (dir.), Acta of the 12th International Congress on Ancient Bronzes (Nijmegen 1992), Provincial Museum G.M. Kam, Amersfoort-Nijmegen 1995, 113 119. 8. Bayley 1997: J. Bayley, S. Butcher, The composition and decoration of Roman brooches. In : A. Sinclair et al. (dir.), Archaeological Sciences 1995. Proceedings of a conference on the application of scientific techniques to the study of archaeology, (Oxbow Monogr. 64), Oxford 1997, 101 106. 9. Bishop 1996: M. C. Bishop, Finds from Roman Aldborough. A Catalogue of Small Finds from the Romano-British Town of Isurium Brigantum (Oxbow Monographs 65), Oxford 1996, 116 p. 10. Bitner-Wróblewska 1996: A. Bitner-Wróblewska, Unpublished star-footed brooches from Scandinavia, Studia Gothica, I, 1996, 73 83. 11. Cool 1998a: H. E. M. Cool et al., The copper-alloy artefacts. In: H. E. M. Cool, C. Philo (dir.), Roman Castleford Vol. I. The small finds (Yorkshire Archaeol. 4), Wakefield 1998, 27 120. 12. Dent 1995: J. Dent, A distinctive form of inlaid brooch from Iron Age Britain. In: B. Raftery et al. (dir.), Sites and sights of the Iron Age. Essays on Fieldwork and Museum Research presented to I. M. Stead (Oxbow Monogr. 56), Oxford 1995, 41 48. 13. Dubin 1995: L. S. Dubin, The history of beads, from 30.000 BC to the present (concise edition), London 1995. 14. Edwards 1994: D. R. Edwards, The social, religious, and political aspects of costume in Josephus. In: J. Lynn Sebesta, L. Bonfante (dir.), The world of Roman costume, Madison 1994, 153 159. 15. Graham-Campbell 1975: J. Graham-Campbell, Bossed penannular brooches: a review of recent research, Med. Arch. 19, 1975, 15 pp. 16. Goldman 1994: B. Goldman, Graeco-Roman dress in Syro-Mesopotamia. In: J. Lynn Sebesta, L. Bonfante (dir.), The world of Roman costume, Madison 1994, 163 181. 17. Goldman 1994: N. Goldman, Reconstructing Roman clothing. In: J. Lynn Sebesta, L. Bonfante (dir.), The world of Roman costume, Madison 1994, 213 237.