ADVANCED MATERIALS & PROCESSES | OCTOBER 2023 20 were used to produce the silver bull. Also, two silver solder samples were debased, with a copper content of 4.6 and 13.3 wt%. It is worth noting that besides Ag and Cu, only gold is measured as a trace element by NAA. Nevertheless, the compositional data achieved with µ-XRF and NAA are significantly similar. Figure 2 shows a bivariate scatter plot of Ag versus Cu in the composition of eight analyzed areas as well as nine analyzed data provided by NAA[1]. The analysis indicates that the pieces are made of two different types of silver: unalloyed silver with a very low concentration of impurities and debased silver-copper alloys with variable copper content. Specifically, the ancient metalworker used debased silver-copper alloys to create most of the upper parts of the object with the exception of the bull’s horns. The horns as well as the solders are made with unalloyed silver, which was also used for the lower parts including the patterned cloth, base, and hind hoof. The previous NAA analysis was performed using a total of nine samples that weighed approximately 100 µg each. Due to sampling limitations, not all parts were analyzed during the NAA analysis and only some of the ancient solders were sampled. Very few analytical studies have been performed on ancient silver production in the Bronze Age and the Proto-Elamite period in Iran[9,10] and the available data are not sufficient to reconstruct the traditions and techniques of manufacturing silver objects in this region. The use of unalloyed silver has been confirmed by the chemical composition of a Bronze Age silver strip from southwestern Iran, ca. 2500 BCE[11]. The available data from the present µ-XRF and previous NAA analyses of this multi-piece object, however, indicates that both unalloyed and debased silver could be used in the same object. The reason for using such a combination is not entirely clear but may be due to the fact that debased silver is harder than unalloyed silver, and this hardness may have been desired in certain sections of the figure[12]. Given the noninvasive nature of this analytical approach, µ-XRF allowed for the collection of a more comprehensive set of data from this object than was possible in the previous NAA study. In addition, µ-XRF analysis enabled a larger number of elements (Fe, Ni, Zn, As, Sn, Sb, Pb, and Bi) to be detected and measured simultaneously, as compared to the previous analyses. Development of noninvasive microanalytical methods such as µ-XRF enables conservators, scientists, and archaeologists to study ancient metallurgy and production practices without the need to collect samples from rare and precious objects. With calibrated quantification methods, noninvasive analysis can attain the precision and accuracy levels comparable to sampling methods used in the past. ~AM&P Acknowledgments: The authors are grateful to Dr. Kim Benzel, Dr. Michael Seymour, and Daira Eden Robert from The Metropolitan Museum of Art for their interest and support of this investigation. For more information: Omid Oudbashi, senior lecturer in conservation science, University of Gothenburg, Sweden, +46.76.618.37.44, omid.oudbashi@gu.se. TABLE 2 – NAA ANALYSIS OF SILVER BULL, WT% Place of Sampling Sample Code Ag Cu Au Solder (neck) U1 95.3 4.6 0.1 Solder (vase) U4 86.7 13.3 0.004 Left hoof SB1 99.4 0.6 0.008 Left haunch SB2 98.3 1.7 0.01 Head between horns U2 99.1 0.9 0.006 Left haunch below joint U6 99.3 0.1 0.03 Edge of sternum U8 98.6 1.4 0.04 Vase U9 96.8 3.2 0.01 Silver pin U3 95.5 4.4 0.07 Fig. 2 – Bivariate scatter plot of Ag versus Cu based on results of µ-XRF and NAA chemical analysis of MMA 66.173 (wt%).
RkJQdWJsaXNoZXIy MTYyMzk3NQ==