Key Engineering Materials Vol. 682 (2016) pp 151-159 © (2016) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/KEM.682.151 Submitted: 2015-09-15 Revised: 2015-10-08 Accepted: 2015-10-10 Online: 2016-02-20 Metallographic studies of selected Eneolithic and Bronze Age artifacts from Poland Józef Szczepan Suchy 1, a, Aldona Garbacz-Klempka 1, b*, Kamil Adamczak 2, c, Łukasz Kowalski 2, d, Janusz Kozana 1, e, Małgorzata Perek-Nowak 3, f, Michał Szucki 1, g and Marcin Piękoś 1, h 1 AGH University of Science and Technology, Faculty of Foundry Engineering, Historical Layers Research Centre Reymonta 23, 30-059 Cracow, Poland 2 3 a Nicolaus Copernicus University, Institute of Archaeology Szosa Bydgoska 44/48, 87-100 Toruń, Poland AGH University of Science and Technology, Faculty of Non-Ferrous Metals, Mickiewicza 30, 30-059 Cracow, Poland jsuchy@agh.edu.pl, b* agarbacz@agh.edu.pl, c adamczak@umk.pl, d lukaszkowalskipl@gmail.com, e jkozana@agh.edu.pl, f mperek@agh.edu.pl, g mszucki@agh.edu.pl, h mpiekos@agh.edu.pl Keywords: Archaeometallurgy, XRF, SEM-EDS, 3D scanning, Eneolithic, Bronze Age Abstract This work presents the results of metallographic studies performed on four Eneolithic and Bronze Age artifacts from Poland. All of them are of none archaeological context therefore its academic value is strongly reduced. The aim of this work is to deal with such a reduced data in a way of improving and verifing current state of knowledge about the artifacts. In order to achieve this goal elemental composition (XRF), microstructure analysis (SEM-EDS), macrostructure analysis (optical microscopy) and 3D scanning were performed. Introduction The earliest metal objects known from Polish land were made of native copper either copper ores. They are dated to a period about 4500-3500 BC and are connected with local Neolithic societies. The next flow of metals (made of arsenic, antimony and tin bronze) took a place in the beginings of the Bronze Age about 2300-2200 BC [1, 2]. For many metal objects information about their archaeological context is strongly limited or unknown. Hence, an academic value of the objects is reduced to a general place of discovery, its morphological and technological characteristic. Nevertheless such a reduced data can be sucessfully improved. Investigation of chemical composition allows to obtain a chemical profile of the objects. It may be then compared with another Neolithic and Bronze Age artifacts. By micro- and macrostructure analysis supported by 3D scanning the workmanship technique production and the ways the particular objects were used may be also established. Materials Investigated here collection of the artifacts consist of an Eneolithic hammer-axe of Szendro A type (Fig. 1a; stored in the Nysa Museum under ACCN: MNa/A/690) discovered in Karłowice Małe (Silesia) [3], an Eneolithic flat axe of Bytyń A type (Fig. 1b; stored in the Toruń District Museum under ACCN: MT/A/868) discovered in Kamionka (Chełmno land), Bronze Age flanged axe (Fig. 1c; stored in the the Toruń District Museum under ACCN: MT/A/719) discovered in Gniew (Pomorania) and a double spiral ornament (Fig. 1d; stored in the Warsaw National Archeological Museum under ACCN: PMA/III/5990). The place of discovery of the last artifact is unknown. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 149.156.96.146-12/12/15,15:20:45) 152 Engineering and Technology on Non-Ferrous Metals All of the artifacts are of none archaeological context therefore we are not able to determine if they were a part of a hoard, either left or deposited on a prehistoric settlement. In consequence all of them should be treated as a single finds. Fig. 1. Investigated artifacts: (a) the hammer-axe; (b) the flat axe; (c) the flanged axe; (d) the double spiral ornament. Experimental method XRF The quantitative determination of the elemental composition was performed by X-ray fluorescence spectrometry (XRF) using energy dispersive X-ray fluorescence spectrometer SPECTRO Midex. The objects were prepared by the chemical removal of conservation layers and mechanical removalof the corrosion products. Microstructure analysis The microstructure observations were performed at a scanning electron microscope (SEM) Hitachi S3400N equipped with Energy-Dispersive X-ray Spectrometer (EDS) by Thermo Noran allowing for phase analysis in microareas. The surface of the objects was prepared by the chemical removal of conservation layers, followed by mechanical polishing with sand papers and final polishing with 1µm diamond paste. The microstructure of the artifacts was observed with respect to the state of preservation, surface quality and elemental composition in microareas. Macrostructure analysis The macrostructure analysis was performed using a NIKON SMZ 745Z stereoscopic microscope with a Nikon Digital Sight DsFi1 microscopic camera and a Nis-Elements BR picture analysis system. The macrostructure of the artifacts was observed with respect to the state of preservation and surface quality. 3D Scanning The hammer-axe was the only reference model used in the 3D scanning. In the research Creaform measurement system with a tracking system C-Track 780 and a scanning head MetraSCAN were applied. The hammer-axe was scanned in three separate parts: top, side and bottom. Firstly it was located on a stationary surface, then around it eight reference points (markers) were randomly selected. The aim of the scanning was to make individual scans partially overlapping each other. The obtained triangle meshes (polygons) were elaborated using VXElements software in order to remove the unnecessary background elements from the scans and to combine them into a whole. Key Engineering Materials Vol. 682 153 Results XRF With respect to the conservation recommendations the XRF analysis was limited to a destruction area of the double spiral ornament. However, during the research it was necessary to expose additional analythical area, therefore, a chemical profile of the double spiral ornament was based on four analythical areas (Fig. 2). Fig. 2. Analythical areas of the double spiral ornament: (a, c) core structure; (b, d) non-core structure. Table 1 presents the results of elemental composition quantification of the artifacts including the quantification results of the flat axe from Kamionka published in 1906 by Chmielecki [3, 4]. Table 1. The elemental composition (wt%) of the artifacts. Artifact Cr Fe Co Ni Cu Zn As Ag Sn Sb Pb Bi The hammer-axe <0.020 <0.025 0.04 <0.015 98.7 <0.010 1.3 <0.020 <0.051 <0.05 <0.020 <0.0010 The flat axe (2015) <0.020 <0.025 0.04 <0.015 99.7 0.049 0.24 <0.020 <0.051 <0.051 <0.020 <0.0010 The flat axe (1906) N/A + N/A N/A 98.5 + N/A N/A 0.8 + + + <0.020 <0.025 0.1 1.2 93.4 0.052 0.48 1.1 0.10 3.5 <0.020 <0.0010 a* <0.020 0.11 0.09 0.45 87.9 0.18 0.54 0.013 9.4 1.1 0.14 0.037 b* The double spiral ornament c* <0.020 0.060 0.09 0.030 2.83 0.21 <0.00051 0.010 56 0.65 40 0.13 0.32 0.090 0.09 0.36 82.6 0.17 0.31 0.010 10 0.79 5.0 0.075 5.4 0.33 0.08 0.24 55.6 0.41 0.13 0.026 19 0.84 18 0.088 The flanged axe d* * an analythical area Having a content of mercury (Hg) below a detection limit (<0.001 wt%) together with a higher level of cobalt (Co), the chemical profiles suggests that all the artifacts were made of recast copper ores [5]. The average iron (Fe) content in the artifacts did not exceed 1 wt% what seems to be typical for the Eneolithic and the Early Bronze Age objects [6]. The major contribution in a chemical profile of the hammer-axe and the flat axe was made by copper (Cu) and in both cases it was completed by arsenic (As). The results suggest that they were made of arsenic copper [3,7-9]. Chemical profile of the flanged axe was based on Cu with antimony (Sb), silver (Ag) and nickel (Ni) making a significant contribution. The profile indicates that it was made of antimony bronze. 154 Engineering and Technology on Non-Ferrous Metals The analythical areas of the double spiral ornament demonstrate a significant variation in elemental composition. Its chemical profile was displayed only in area „a”. The major contribution in the profile was made by Cu and it was completed by tin (Sn). The results suggest that the double spiral ornament was made of tin bronze. Weight fractions of some elements, particularly Sn and lead (Pb), in the remaining analythical areas may indicate that they were not originally included in the matrix of this artifact. Microstructure analysis Microstructure of the hammer-axe is consists of crystallites with developed grain boundaries (Fig. 3a and b). Characteristic pattern in the microstructure (marked with an arrow) suggests an object underwent plastic deformation processes (Fig. 3c). On the other hand, microstructure presented in Figure 3d indicates its dendritic character. Since As content in the structure is in the range 0.56 wt% – 3.4 wt%, the only stable phase is solid solution α according to Cu-As phase diagram [10]. Fig. 3. The microstructure of the hammer-axe: (a-c) SEM images; (d) an optical microscope image. In both flat axes, the microstructure with well-developed grain boundaries was revealed (Fig. 4). The large grains, some of them including twins, that exist in the microstructure of the flat axe suggest that an object was forged followed by annealing (Fig. 4a and b). In the case of the flanged axe the small grains indicate recrystallization processes in the structure (Fig. 4c and d). Therefore it can be concluded, that the axe was also annealed after metal working, most probably forging. Analysis of chemical composition at microarea II of flat axe (Fig. 4b and Table 2) indicate arsenic copper. In microarea II of the flanged axe, phases enriched with antimony are present (Fig. 4d and Table 2). Key Engineering Materials Vol. 682 155 Fig. 4. The microstructure of the flat axe (a, b) and the flanged axe (c, d) Besides, sulfides crystallites are present in the structure of the ornament (Fig. 5a). Chemical composition of I and II microareas (Table 2) allow to state that the raw material used for manufacturing of the analyzed object originated in sulfide ores. Table 2. The elemental composition (wt%) of the artifacts within the microareas. Artifact Fig. 5. The microstructure of the double spiral ornament: (a) the core; (b) the coarse solder. MA S Fe Ni Cu As Ag Sn Sb Pb I - - - 100 0.0 - - - - The flat axe II - - - 99.1 0.9 - - - - The flanged axe I - - 8.1 49.9 0.0 0.00 - 42 - II - - 3.7 87.1 3 2.3 - 4.2 - I 22 1.2 - 77.2 - - - - - II 16 1.1 - 80.5 - - - - - 0.00 0.00 - 92.9 - - - - - The double spiral ornament III IV - - - - - - 12.8 - 87.2 V - - - - - - 100 - 0 VI - - - - - - 13.4 - 86.6 * MA – the analythical microarea. Microstructure of microareas IV, V, VI (Fig. 5 and Table 2) indicates that a tin-lead solder was applied in a place of damage. This is in good relation with observations of the core and surrounding it structure composed of elements with higher atomic number than Cu (Fig. 6a and b) and with XRF analysis in the area b (Table 1). Besides, a corrosion cell is observed in the core microstructure (Fig. 6c). 156 Engineering and Technology on Non-Ferrous Metals Fig. 6. The SEM image of the double spiral ornament: (a,b) the coarse solder; (c) the corrosion centre. Macrostructure analysis The macroscopic observations confirmed the presence of the casting defects on the surface of the flat axe. They were identified as gating system (Fig. 7a) and cast seam (Fig. 7b). Slag inclusions (Fig. 8a) were present on the surface of the flanged axe. Such a structures allow us to asume that both axes were casted in two part closed moulds. The macroscopic observation confirmed the presence of the usage traces on the flanged axe's blade (Fig. 8 b) which probably resulted from its usage in the prehistory. The macroscopic observations signalize the plastic forming of both axes. For the metal structure improving and refining its shapes the axes were probably cold hammered just after being casted. Fig. 7. The macrostructure of the flat axe: (a) the gating system; (b) the cast seam. Fig. 8. The macrostructure of the flanged axe: (a) the slag inclusions; (b) the usage traces blade. The presence of the usage traces left on the hammer-axe surface was confirmed and identified as deformations of the blade and the butt-end part. The presence of lips deformations around the shafthole (with Fig. 9a) and around the cutting edge (Fig. 9b) supports a conclusion that the hammer-axe was probably cold worked. On the surface of the hammer-axe nor cast seams neither gating system were noticed, therefore it is likely that it was cast in open (sand) mould and metal stream was directed into the mold cavity. The double spiral ornament was made by coiling regular and round in a cross section wire Fig. 10a) and it was probably cold hammered what can be supported by the traces left on the coil turns (Fig. 10b). The surface analyses undoubtedly made the possibility of the double spiral ornament having been initially given a pin mechanism unlikely. Key Engineering Materials Vol. 682 Fig. 9. The macrostructure of the hammer-axe: (a) the lips deformations; (b) the cutting edge deformations. 157 Fig. 10. The macrostructure of the double spiral ornament: (a) the coil turns; (b) the cold hammering traces. 3D Scanning 3D scanning confirmed presence of deformations around a shaft-hole (Fig. 11a) and lateral deformations around the blade part of the hammer-axe (Fig. 11b). The observations made by 3D scanning correspond with the results of the macrostructural analysis. Fig. 11. The 3D scan of the hammer-axe: (a) the shaft-hole deformations; (b) the blade deformations. Discussion The area around the shaft-hole of the hammer-axe from Karłowice Małe is deformed as it is shown particularly in the down side of the casting. This is not the result of the casting but provides rather possibility that the shaft-hole was not formed by clay core but punched using a round rod or pin when metal was still in a semi-molten state [3,7]. Due to a fact that Chmielnicki did not specify an applied analytical method [4] comparison of the chemical profiles of the flat axe from Kamionka encounter difficulties. However, both profiles differ from each other significantly (about 1 wt%) in the Cu weight fraction. The corresponding qualitatively results were obtained for Fe, Zn, Sb and bismuth (Bi). The results published in 1906 indicated an absence of Co and As which probably resulted from the instrumental limitations. With respect to the content of Sn amounting 0.8 wt% the axe from Kamionka has been classified into the Early Bronze Age. It is approximately eight times higher than the result obtained here. Verification of the chemical profile changes significantly chronological attribution of the flat axe from Kamionka, which should be placed now in the Eneolithic horizon of the flat axes. The major non-copper contribution in the profile of the flanged axe from Gniew was made by Sb amounting 3.5 wt%. The quantitive indicate that Ni, Ag and As made a significant contribution to 158 Engineering and Technology on Non-Ferrous Metals its chemical profile. This can be found as a consequence of their occurrence as natural surroundings of the secondary enriched sulfide ores in grey ores type [11,12]. The double spiral ornament was made of tin bronze melted from sulphide ore. It is very likely that a metalworker used widely available chalcopyrite. This is confirmed by the quantitative results indicating the presence of Ni, Fe, As, Ag, Sb and Bi, which all can be found as natural surroundings of the chalcopyrite [11]. The results of the elemental composition obtained by XRF for the double spiral ornament were statistically elaborated using the cluster analysis with two-way joining method (Fig. 12). It allowed to distinguish three areas: (a) a pure bronze core (with major contribution made by Cu completed by As, Sb and Ni), (b) a core – solder contact area (with major contribution made by Cr, Zn and Fe) and (c) a pure coarse solder (with major contribution made by Sn, Pb and Bi). The double spiral ornament belongs to the Doppelspiralscheiben mit Spiraligem Verbindungstück typological group [13]. It is significant that in comparison to the other objects of a such type, the analyzed one has a unique number (two) of turns in a cylindrical link, as in general the number is not less than the four (Fig. 13). Having the soft solder at the destruction area the double spiral ornament could be considered as being initially provided with a greater number of turns in the link. Fig. 12. The results of the cluster analysis for the chemical profile of the double spiral ornament using two-way joining method: (a) the pure bronze core; (b) the core – solder contact area; (c) the pure coarse solder. Fig 13. The selected double spiral ornaments from Poland: (a) Karmin, Milicz county; (b) the analyzed here ornament, place of discovery is unknown; (c) Kietrz, Głubczyce county; (d) Lubiąż, Wołów county; based on [13]. The increased content of Zn and Cr in the core – solder contact area may be consider as contributed by preparing the surface of the double spiral ornament with polish paste based on chromium oxide (Cr2O3) and soldering flux based on zinc chloride (ZnCl2) for the coarse solder application. This may indicate that its re-combining was a present (Sn-Pb-Bi solder) conservation treatment. Final remarks Prehistoric metal artifacts often come from random discoveries. Some of them were obtained even in the period of antiquarians activity at the 18th or 19th centuries. For many of these objects information about their archaeological context is strongly limited or unknown. As it was presented in this study such a reduced data can be improved. Applying advanced research methods based on metallurgical sciences allowed us to conclude on a raw materials used for production of the investigated objects. The obtained results were also used to establish the workmanship technique of all the artifacts (particularly the hammer-axe) and the ways how the they were used, modified or repaired (particularly the double spiral ornament). Moreover, it was possible to verify the results of quantifications performed for the axes at the beginning of 20th century. It allowed us to precise its chemical profiles and in consequence to change their chronological and technological attributions. Such conducted study gives a opportunity to look at the past communities not only as a producers but also as users of the earliest metal objects. Key Engineering Materials Vol. 682 159 Acknowledgments We would like to thank Wojciech Brzeziński - the Director of the Warsaw National Archaeological Museum, Edward Hałajko - the Director of the Nysa Museum and Marek Rubnikowicz - the Director of the Toruń District Museum for making the artifacts available to the research and giving the agreement to publish the results. Special thanks from the authors are due to Mirosława Andrzejowska – the Chief of Bronze and Early Iron Age Department in the Warsaw National Archaeological Museum and to Mariusz Krawczyk – the Chief of Department of Archaeology, Arts and Crafts in the Nysa Museum for Their kindness and helpfulness. The work has been implemented within the framework of statutory research of AGH University of Science and Technology, contract No 11.11.170.318 - 11 AGH. References [1] J. Kostrzewski, Skarby i luźne znaleziska metalowe od eneolitu do wczesnego okresu żelaza z górnego i środkowego dorzecza Wisły i górnego dorzecza Warty, Przegląd Archeologiczny 15 (1962) 5-133. [2] A. Garbacz-Klempka, S. Rzadkosz, J. Górski, Artefacts from Krakow-Nowa Huta as an Illustration of Selected Issues of Research Into Prehistoric and Mediaeval Casting, Metallurgy and Foundry Engineering 39 (2013) 23–28. DOI: mafe.2013.39.2.23. [3] K. Adamczak, Ł. Kowalski, A. Garbacz-Klempka, K. Dobrzański, Siekieromłot typu Szendrő z Karłowic Małych, woj. opolskie w świetle analiz archeologicznych i metaloznawczych, Śląskie Sprawozdania Archeologiczne 57 (2015) in press. [4] K. Chmielecki, Stare bronzy w zbiorach Towarzystwa Naukowego w Toruniu, Roczniki Towarzystwa Naukowego w Toruniu 3 (1906) 65-82. [5] E. Pernicka, F. Begemann, S. Schmitt-Strecker, H. Todorova, I. Kuleff, Prehistoric copper in Bulgaria, Eurasia Antiqua 3 (1997) 41-180. [6] S.R.B. Cook, S. Aschenbrenner, The Occurence of Metallic Iron in Ancient Copper, Journal of Field Archaeology 2/3 (1975) 251-266. [7] J. Heeb, Copper shaft-hole axes and early metallurgy in South-Eastern Europe, Archaeopress Archaeology, Oxford, 2014. [8] A. Hauptmann, The Archaeometallurgy of Copper. Evidence from Faynan, Jordan, Springer, Berlin-New York, 2007. [9] E. Pernicka, Gewinnung und Verbreitung der Metalle in prähistorischer Zeit (=Jahrbuch des Römisch-Germanischen Zentralmuseum 37), Römisch-Germanischen Zentralmuseum, Mainz, 1990. [10] T. B. Massalski (Ed.-in-Chief), H. Okamoto, P. R. Subramanian, L. Kacprzak (Eds.), Binary Alloy Phase Diagrams, ASM International, Materials Park, Ohio, 1990. [11] A. Bolewski, A. Manecki, Mineralogia szczegółowa, PAE, Warszawa, 1993. [12] A. Pike, Appendix : Analysis of Caucasian Metalwork – The Use of Antimonal, Arsenical and Tin Bronze in the Late Bronze Age, in: J. Curtis, M. Kruszyński (Eds.), Ancient Caucasian and Related Material in The British Museum, The British Museum Press, London, 2002, pp. 87-98. [13] M. Gedl, Die Fibeln in Polen (=Prähistorische Bronzefunde XIV: 10), Franz Steiner Verlag, Stuttgart, 2004.