The analysis of lead, which is used to make slingshot projectiles mainly from random collections in Dobrudja (Romania), was carried out. Such shells were presumably used in the Macedonian and Roman armies. Mass spectrometric analysis with inductively coupled plasma was used to determine trace elements, and the same analysis with a multicollector was used to determine the isotope ratio. Comparison of the results with the literature data on known lead mines in the Balkans and Greece allows us to determine the sources of lead raw materials.

Keywords: lead, projectiles, Roman Empire era, Dobrudja, mass spectrometry.

Introduction

The article presents the results of the analysis of lead, which was used to make throwing projectiles for slings, accidentally discovered in the territory of Dobrudja (Romania). This territory has been part of the province of Inner Moesia of the Roman Empire since the first century BC.

Located between the Danube and the Black Sea, Dobrudja has been the scene of a clash of different civilizations at various stages of its history. Tribes from the East penetrated here, and this area fell under the control and influence of the great civilizations of antiquity - Greek, Macedonian, and Roman. The material remains of these civilizations contain evidence of a turbulent historical past. Determining the sources of raw materials from which metal products were made is of interest for reconstructing the history of this strategically important area.

This problem can be solved by compositional and isotopic analyses of lead products. Currently, it is customary to analyze the same type of artifacts, the origin of which is established with sufficient probability, and compare the results with literature data on possible sources of raw materials.

Analysis of trace elements, although it does not give definite indications of the source of raw materials (during the smelting of metal from ore and its subsequent purification, the chemical composition may change slightly [Tylecote. Ghaznavi and Boydell, 1977, p. 305]), but it is important for selecting and grouping finds for isotopic analysis. Lead isotope analysis is a reliable method that helps to determine the origin of raw materials [Gale, 1989]; it is based on differences in the isotopic composition depending on the geological age and features of the ore deposit formation conditions. The isotopic composition is almost constant for a particular mine and does not change during metallurgical processing or weathering [Barnes et al., 1978; Pollard and Heron, 1996].

The purpose of this study is to determine the geographical location of the source of lead ore used in the production of the studied slingshot projectiles.

Presentation of the research object

The article examines lead artifacts, many of which were discovered accidentally in the coastal zone of Dobrudja, near the ancient Greek cities of Istria, Tomis, Kallatitis and Argamum on the coast of the Euxine Pontus (Black Sea). Several lead projectiles were found in the area of Northern Dobrudja (Tulcija region between the Danube and Euxine Pontus). The finds can be attributed to the period between the IV century BC and the late period of the Roman Empire (I-III centuries AD). Today, these objects are kept in the Museum of National History and Archeology (Romania, Constanta).

According to modern data, lead projectiles were made during military campaigns and even directly on the battlefield by casting in a double-leafed clay mold, similar to the samples found in Phanagoria, north of the Euxine Pontus, or in the area of the Greek city of Olintos (Veiling, 1990). Most likely, lead was delivered from the deposits in the form of ingots. The artefacts studied were made and used on the territory of Dobrudja by soldiers of the Macedonian and Roman armies from the time of Alexander the Great, as evidenced by the inscriptions with his name on some samples (335 BC), as well as by the commander Zopyrion, who led a campaign against Olbia north of the Danube (Suceveanu, 1993).

Several studies have been devoted to inscriptions on lead projectiles (Guarducci, 1987; Tuck, 1999-2001). Projectiles that probably date back to the Roman Empire have no inscriptions; they are spindle-shaped, in contrast to the egg-shaped products of the Greek period (Veiling, 1990).

Table 1. Composition of trace elements in projectiles, mcg / kg

The problem of identifying the source of lead used for the production of projectiles occupies a special place in historical and archaeological research, especially when determining the geographical location of a lead deposit. A solution to this problem would help clarify some of the issues concerning the movement of Macedonian and Roman armies during military conflicts in the territory between the Danube and the Black Sea for several centuries.

For analysis, 11 samples were taken: six of the projectiles that are egg-shaped and the corresponding inscription refers to the Macedonian army, (group A) and five of the elongated artifacts with pointed ends, dated to the Roman era, (group B).

Chemical analysis

Trace element analysis was performed using inductively coupled plasma quadrupole mass spectrometry (ICP-MS, Perkin Elmer / Sciex Elan 6000). Samples (70-110 mg) were treated with 1 ml of concentrated nitric acid (65%) and heated for 1 h at 60 °C. The resulting solution was carefully evaporated almost dry, the remainder was placed in 5 ml of 2% HNO3 solution and heated for several minutes at 100-120 °C. The cooled solution was placed in a 50-milliliter volumetric flask and brought to the required volume with a 2% HNO3 solution. The resulting solution was completely transparent and did not contain any sediment.

For ICP-MS analysis, 0.5-5.0 ml aliquot of the test solution was diluted to 10 ml with a 2% HNO3 solution. To minimize the effects of shifting the instrument readings, an internal standard of 10 ng/ml Rh was added to the solution. The final dilution factor was 5 000 - 14 000 depending on the concentration of the element.

For calibration, a multi-element standard solution (Merck) was used, diluted according to the concentration of the sample. Each series of two or three samples is twice calibrated. All reagents used, including water, were of high purity. The results of the analysis are presented in Table 1.

Lead isotope analysis

The source of raw materials was determined by the method of lead isotope analysis; it allows you to study the smallest samples and does not require their purification. Nu Instruments MK-ICP-MS (multicollector mass spectrometry with inductively coupled plasma) was used for the analysis.

NIST SRM 981 reference materials were used as standards. The following values were obtained (in all cases, the error is 2 o):

²⁰⁶Pb/²⁰⁴Pb = 16,937 ± 0,002

²⁰⁷Pb/²⁰⁴Pb = 15,493 ± 0,002

²⁰⁸Pb/²⁰⁴Pb = 36,708 ± 0,004.

Samples (approximately 1 mg) were dissolved in concentrated HNO3. The resulting solutions are diluted to a concentration of 100 billionths.

The results of isotope analysis were processed by two-dimensional discriminant analysis (Table 2, Figs. On the same charts, with the goal of

Table 2. Ratio of lead isotopes in projectiles

Figure 1. Ratio of ²⁰⁶ Pb/²⁰⁴ Pb vs. ²⁰⁷Pb/²⁰⁶Pb.

Figure 2. Ratio of ²⁰⁸ Pb/²⁰⁶ Pb vs. ²⁰⁷ Pb/²⁰⁶ Pb. See Fig. 1 for the conditional obozn.

In order to determine the sources of the metal, literature data on lead deposits on the Halkidiki peninsula, Thassos Island, and in the Lavrion area (Greece) are presented (Stosgale, Gale, and Annetts, 1996). The results obtained by us are in good agreement with the isotopic values of the Halkidiki and Thassos deposits.

Results and discussion

Chemical analysis

Among the elements found in trace amounts in lead products, the most important are silver and bismuth, since they, being in metallic lead in the form of a solid solution, can characterize the ore. Other impurities are arsenic, copper, and antimony, with copper usually accompanied by nickel. The silver content is low and averages 109 micrograms / g (varies slightly in the samples). Given that at that time mainly silver-bearing lead ores were used, the results obtained can be considered as evidence of a preliminary silver separation process or the use of lead oxide, a by-product of the cupellation process. This assumption is also confirmed by the relatively low content of bismuth in the samples (an average content of 94 µg/g), which was previously indicated by I. Kuleff [Kuleff et al., 2006]. Projectiles are characterized by

as well as the arsenic content, which varies from 0.1 to 700 micrograms/g, and the copper concentration from 0.02 to 1.2%. Our results are comparable with the indicators for lead products of the Hellenistic period from the territory of ancient Thrace (now Bulgaria) [Ibid.].

We compared the data obtained with the results of a study of anchors made of wood and lead from the sixth century BC to the second century AD found on the Black Sea coast of Bulgaria [Kuleff et al., 1995, p. 64]. After processing the isotopic parameters of the main components of the armature, taking into account the probable source of lead, they were divided into four groups. The results were compared with data on lead deposits that were developed in the Aegean Sea region at that time.

According to the composition of impurities, lead from the projectiles studied by us is similar to that from the known deposits on the Halkidiki peninsula (in terms of Ag and Au content), Phasos Island (in terms of As, Bi, Cu and Sb content) and near Lavrion (in terms of As and Bi content). However, the inconstancy of the content of impurities in lead within deposits does not allow us to unambiguously determine the source of ore, especially since the content of various elements varies greatly during processing.

Lead isotope analysis

Analyzing the data obtained, we note that in the graphs all samples are grouped in one narrow area (Figures 1, 2). The samples of group B in terms of the ratio of lead isotopes coincide with those of group A. If we assume that both armies purchased supplies directly at the site of military operations, this coincidence is quite expected.

Within the narrow region that all the samples form, you can select objects that are even closer to each other. So, within Group A, four samples show exceptionally close isotopic composition, which implies a single production process. An obvious problem is the significant spread of data on the trace elements of these products. Even if there was only one source of ore, with a low level of production technology, differences in the concentration of impurities in lead projectiles are inevitable.

The close isotopic composition of the samples of groups A and B indicates that all the studied products are made of lead produced from the ore deposits of the same region. High values of the ²⁰⁶ Pb/²⁰⁴ Pb ratio are a sign of a geologically young "Alpine" type of ore, characteristic of the Balkan-Greek region. The samples are quite consistent with the isotopic composition of ore from either one deposit or a group of mines located in the same area. Both graphs support the hypothesis that there are at least two close metal sources.

It is rather difficult to identify these deposits, since most of the published data relate to deposits that have been developed since the Bronze Age. However, it is possible that some of these mines were used in the last centuries BC. It should also be emphasized that so far no traces of ore processing and deposits have been found in Dobrudja during archaeological surveys, despite numerous finds in the form of lead products.

The published data related to the lead ore deposits in Halkidiki and Thassos are consistent with most of our results. Thus, these deposits are the most likely sources of ore for the projectiles we study. Given the low ²⁰⁶ Pb / ²⁰⁴ Pb ratio in samples A5 and B9, it can be assumed that the source of ore for these products was another deposit located near the Rhodope region, which is geologically older.

Group B artefacts may be Roman projectiles made from the ore of Halkidiki and Thassos. It is also possible that they were obtained by recycling old lead projectiles. The relatively straight-line arrangement of the values of Group B samples seems to indicate a two-component mixture, with the data of pure components appearing as two extrema on the right line, and the intermediate points are mixtures of different proportions (Pinarelli, 2004).

All the above mentioned areas were under the influence or were subject to Macedonia and later the Roman Empire. Thus, the simplest hypothesis is that the metal for the shells of both armies was extracted from lead ore directly in the deposits, and the projectiles themselves were usually made directly during military campaigns; only weapons that required high skill (for example, swords or shields) were delivered from afar. Rear supply of troops made it possible to make low-tech weapons right on the march. Although the lead used had a variable chemical composition and many impurities, the homogeneous isotopic composition of the metal nevertheless allows us to establish its geographical origin.

Conclusion

ICP-MS and MK-ICP-MS-methods established the origin of two groups of lead projectiles for slings of the Macedonian and Roman armies,

found on the territory of Dobrudja. The most likely sources of ore were the mines of Halkidiki and Thassos. This implies the use of both Macedonian and Roman armies of ore from the same deposits for smelting lead, which was used to make throwing projectiles for slings during military campaigns. The use of isotopic analysis of lead in determining its origin provides important information about the metal trade and turnover in an era when the sling was a deadly weapon.

List of literature

Barnes I.L., Gramlic J.W., Diaz M.G., Brill R.H. The possible change of the lead isotope ratios in the manufacture of pigments; a fractionation experiment // Archaeological Chemistry II. Advances in Chemistry. - 1978. - Vol. 171. - P. 273 - 279.

Gale N.H Lead isotope analyses applied to provenance studies: A brief review // Archaeometry, Proceedings of the 25^th International Symposium. - Amsterdam, 1989. - P. 469 - 502.

Guarducci M. L'epigrafia greca dalle origini al tardo impero. - Roma: Institute Poligrafico e Zecca dello State, 1987. - 561 p.

Kuleff I., Djingova R., Alexandrova A., Vakova V., Amov B. INAA, AAS and lead isotope analysis of ancient lead anchors from the Black Sea // J. Radioan. Nucl. Ch. - 1995. - Vol. CXCVI, N1. - P 65 - 76.

Kuleff I., Iliev I., Pernicka E., Gregova D. Chemical and lead isotope compositions of lead artefacts from ancient Thracia (Bulgaria) // J. Cult Herit. - 2006. - Vol. VII. - P. 244 - 256.

Pinarelli L. Lead Isotope Characterization of Copper Ingots from Sardinia (Italy): Inferences on their Origins // Bull, of Geological Society of Greece. - 2004. - Vol. XXXVI. - P. 1173 - 1180.

Pollard A.M., Heron C. Lead isotope geochemistry and the trade in metals // Archaeological Chemistry. - Cambridge: The Royal Society of Chemistry Publication, 1996. - P. 302 - 336.

Stos-Gale Z.A., Gale N.H, Annetts N. Lead isotope data from the Isotrace Laboratory, Oxford: Archaeometry data base 3, ores from the Aegean, part 1, Lead isotope geochemistry and the trade in metals // Archeometry. - 1996. - Vol. 38. - P. 381 - 390.

Suceveanu Al. Alexandra eel Mare. - Bucuresti: Editura Academiei Romane, 1993. - 240 p.

Tuck S.L. "Ouch!" Inscribed Greek sling projectiles // Missouri Annual of the Museum of Art and Archaeology. - 1999 - 2001. - Vol. 33/35. - P. 15 - 31.

Tylecote R.F., Ghaznavi H.A., Boydell P.J. Partitioning of trace elements between ores, fluxes, slags and metal during smelting of copper // J. Archaeological Science. - 1977. - Vol. 4. - P. 305 - 333.

Veiling T. Funditores in romischen Heer // Saalsburg Jahrbuch. - 1990. - Vol. 45. - P. 20 - 58.

The article was submitted to the Editorial Board on 04.02.10.

Permanent link to this publication: