Research

9 pages
179 views

Bronze Jewellery from the Early Iron Age urn-field in Mała Kępa. An approach to casting technology

of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
This study characterizes the bronze jewellery recovered from the Lusatian culture urn-field in Mała Kępa (Chełmno land, Poland). Among many common ornaments (e.g. necklaces, rings, pins) the ones giving evidence of a steppe-styled inspiration (nail
Transcript
    ARCHIVES of FOUNDRY ENGINEERING DOI: 10.1515/afe - 2017 - 0112   Published quarterly as the organ of the Foundry Commission of the Polish Academy of Sciences ISSN (2299-2944) Volume 17 Issue 3/2017 175  –   183  ARCHIVES of FOUNDRY ENGINEERING Volume 17, Issue 3/2017, 175-183 175   Bronze Jewellery from the Early Iron Age urn- field in Mała Kępa.  An approach to casting technology   A. Garbacz-Klempka a, *, Ł. Kowalski b , J. Gackowski b , M. Perek-Nowak c   a  AGH-University of Science and Technology, Faculty of Foundry Engineering, Historical Layers Research Centre Reymonta 23, 30- 059 Kraków, Poland   b   Nicolaus Copernicus University, Institute of Archaeology, Szosa Bydgoska 44/48, 87- 100 Toruń, Poland   c  AGH-University of Science and Technology, Faculty of Non-Ferrous Metals, Mickiewicza 30, 30- 059 Kraków, Poland   * Corresponding author. E-mail address: agarbacz@agh.edu.pl Received 13.09.2016; accepted in revised form 04.01.2017   Abstract   This study characterizes the bronze jewellery recovered from the Lusatian culture urn-field in Mała Kępa (Chełmno land, Poland). Among many common ornaments (e.g. necklaces, rings, pins) the ones giving evidence of a steppe-styled inspiration (nail earrings) were also identified. With the dendritic microstructures revealed, the nail earrings prove the implementing of a lost-wax casting method, whereas some of the castings were further subjected to metalworking. The elemental composition indicates the application of two main types of  bronze alloys: Cu-Sn and Cu-Sn-Pb. It has been established that the Lusatian metalworkers were familiar with re-melting the scrap bronze and made themselves capable of roasting the sulphide-rich ores. The collection from Mała Kęp a has been described in terms of its structure and composition. The investigations were made by means of the energy dispersive X-ray fluorescence spectroscopy (ED-XRF), scanning electron microscopy (SEM) coupled with an energy dispersive X - ray analysis system (EDS) and optical microscopy (OM). In order to fingerprint an alloy profile of the castings with a special emphasis on the nail earrings, the data-set (ED-XRF, EDS) was statistically evaluated using multidimensional analyses (FA, DA). Keywords:  Non-Destructive Testing, Investment casting, X-ray fluorescence spectroscopy, Archaeometallurgy, Lusatian culture 1. Introduction  During excavations made between 1956-1958   by the Toruń District Museum at site 1 in Mała Kępa (Dąbrowa Chełmińska commune, Bydgoszcz county) the Early Iron Age cemetery of the Chełmno  group society of the Lusatian culture was recovered. It has been preserved in a form of an urn-field with 122 cremation graves. Besides numerous ceramic vessels, metal artefacts giving evidence of bronze valorisation manifested as grave offerings (e.g. necklaces, rings, pins) were also identified. According to the archaeological data obtained so far, the urn- field in Mała Kępa was established around 900BC and was used for over 450 years (Ha B1/B2 - Ha D3) [1, 2]. The jewellery investigated here was manufactured by two techniques: (1) the investment casting method and (2) a plastic working. Since the Lusatian metalworkers were fully aware of the alloying limitations, they applied the plastic working as a deliberate treatment for improving mechanical properties of the castings (e.g. necklaces), while the lost-wax method was used rather to cast small objects (e.g. nail earrings) and those more advanced in shape, like decorative pendants [3-8]. UnauthenticatedDownload Date | 10/21/17 10:41 AM  176 ARCHIVES of FOUNDRY ENGINEERING Volume 17, Issue 3/2017, 175-183   2. Materials and methods   2.1. Materials  As the funeral rite practi ced by the Chełmno group of the Lusatian culture required burning the dead member body together with the grave offerings, the majority of artefacts investigated here is thermally deformed and defragmented (Fig. 1). Among numerous body ornaments (e.g. earrings, necklaces) including items of clothing (e.g. pins), the ones preserved in unrecognizable form (e.g. wire, lumps) were also selected. Fig. 1. The investigated collection (by courtesy of the Toruń  District Museum)   2.2. Methods  The elemental composition was established by means of the X-ray fluorescence spectrometry with energy dispersive X-ray fluorescence (ED-XRF) spectrometer Spectro Midex equipped with a molybdenum X-ray lamp of an excitation energy of 44.6kV and a Si Drift Detector (SDD) with a resolution of 150eV. For the nail earrings the elemental composition was completed by the SEM-EDX investigations using a SEM microscope Hitachi S 3400N equipped with the EDS spectrometer by Thermo Noran. The mechanical removal of the corrosion products (reaching the metallic core level) followed by surface degreasing using acetone were prior to the ED-XRF investigations. The microstructure of the nail earrings was observed with a Nikon Eclipse LV150 optical microscopy (OM) equipped with a Nikon Digital Sight DsFi1 microscopic camera and it was completed by the SEM investigations. The macrostructure observations were made using a Nikon SMZ 745Z stereoscopic microscope (OM) equipped with the  Nikon Digital Sight DsFi1 microscopic camera. 3. Results  The results were divided with a special emphasis on the nail earrings and necklaces. The rest of the artefacts were taken as a comparative background and therefore discussed in further parts of this work. The averaged and normalized results of the elemental composition determined for the jewellery, including 39 artefacts (ED-XRF) are summarized in Table1. 3.1. Nail earrings  While the MaKa_27 is preserved almost in a full form, the MaKa_28 underwent strong defragmentation and lost its srcinal circle-like shape (Fig. 2). The nail earrings do not, however exhibit many surface casting defects (e.g. blows or drops) what could be due to the proper pattern design and casting practice. Fig. 2. The nail earrings with reconstruction of an srcinal form The nail earrings were cast in tin-bronze. In fact, the castings contain the levels of tin (7.2wt% and 13wt%) suggesting that the alloys could have been kept in their as-cast condition [9]. When looking at the macrostructure of the nail earrings, it is almost evident that such ornaments could be manufactured only using a lost-wax casting method. With dendrites and well-developed grain boundaries (partially degraded by corrosion), the microstructures are typical for the cast structures (Fig. 3, 4). Moreover, the large and hexagonal grains are indicative of slight forging followed by annealing. Fig. 3. The OM images of the nail earrings microstructure The nail earrings exhibit a dendritic α -phase containing tin dendrites forming a solid solution with copper. Whereas the inclusions brought to the MaKa_28 by Pb were captured in the interdendritic phase and appear white, no significant inclusions can be found in the MaKa_27 matrix (Fig. 4, Table 2). The high amounts of iron (22wt%) and sulphur (22wt%) in the EDS microareas of the MaKa_28 should be noticed. While the elements such as Ag, As and Sb tend to agglomerate as impurities on the grain boundaries, Fe, Pb and S usually form Cu 2−x Fe x S-inclusions in the tin  –   bronze matrix [10]. It is commonly assumed that copper containing sulphur was produced by the smelting of chalcopyrite ore [9] and therefore it is likely that sulphide-rich ore was used to manufacture the MaKa_28 nail earring. UnauthenticatedDownload Date | 10/21/17 10:41 AM   ARCHIVES of FOUNDRY ENGINEERING Volume 17, Issue 3/2017, 175-183 177 Table 1. The elemental composition (wt%) of the jewellery by means of ED-XRF Signature Artefact Grave no  Fe Co Ni Cu Zn As Ag Sn Sb Pb Bi  MaKa_1 Necklace 20  0.086 0.056 0.42 92 0.14 0.48 0.62 5.0 0.95 0.030 0.070  MaKa_2 Necklace 95  0.042 0.065 0.23 88 0.13 0.31 0.17 10 0.38 0.88 0.025  MaKa_3 Necklace 102  0.088 0.072 0.23 88 0.13 0.36 0.16 10 0.35 0.85 0.037  MaKa_4 Pin 89  0.21 0.061 0.17 87 0.13 0.12 0.083 7.9 0.093 4.3 0.28  MaKa_5 Rod coil 89  0.15 0.091 0.10 87 0.12 0.19 < 0.020 12 < 0.051 0.28 0.021  MaKa_6 Necklace 89  0.056 0.059 0.14 91 0.12 0.15 0.14 4.7 0.094 3.3 0.37  MaKa_7 Lump 89  0.84 0.088 0.15 79 0.16 0.37 0.15 8.5 0.23 10 0.66  MaKa_8 Earring 89  0.084 0.080 0.079 86 0.13 0.26 < 0.020 13 < 0.051 0.083 0.015  MaKa_9 Earring 89  0.077 0.075 0.081 87 0.12 0.24 0.025 12 < 0.051 0.062 0.023  MaKa_10 Earring 89  0.19 0.073 0.24 87 0.12 0.23 0.10 10 0.15 2.1 0.079  MaKa_11 Earring 89  0.067 0.099 0.24 87 0.12 0.30 0.12 10 0.18 1.6 0.31  MaKa_12 Earring 89  0.090 0.072 0.084 87 0.13 0.24 0.023 12 < 0.051 0.12 0.023  MaKa_13 Earring 89  0.072 0.087 0.10 87 0.16 0.30 < 0.020 11 < 0.051 0.98 0.011  MaKa_14 Earring 89  0.32 0.11 0.12 84 0.18 0.24 < 0.020 12 < 0.051 3.4 0.033  MaKa_15 Earring 89  0.21 0.080 0.077 82 0.11 0.36 0.041 16 < 0.051 0.39 0.035  MaKa_16 Earring 89  0.48 0.056 0.27 90 0.13 0.43 1.4 3.5 2.2 1.3 0.034  MaKa_17 Earring 89  0.17 0.075 0.062 88 0.15 0.23 0.082 11 < 0.051 0.30 0.039  MaKa_18 Earring 89  0.20 0.071 0.10 89 0.13 0.37 0.027 10 < 0.051 0.27 0.042  MaKa_19 Necklace 80  0.11 0.066 0.34 88 0.12 0.053 0.034 11 < 0.051 0.058 0.010  MaKa_20 Pin 80  0.080 0.062 0.20 92 0.13 0.33 0.18 6.4 0.52 0.36 0.021  MaKa_21 Ring 38  0.36 0.11 0.34 70 0.17 1.7 0.26 22 1.2 4.0 0.13  MaKa_22 Ring 38  0.013 0.077 0.16 92 0.18 0.99 0.17 5.2 0.63 0.78 0.028  MaKa_23 Pin 78  0.037 0.058 0.10 89 0.12 0.051 0.085 10 0.064 0.29 0.080  MaKa_24 Plate 84  0.041 0.064 0.52 93 0.12 0.36 0.50 3.8 1.4 0.27 0.015  MaKa_25 Salta leone 103  0.068 0.070 0.36 82 0.12 0.61 0.10 16 < 0.051 0.82 0.24  MaKa_26 Earring 40  0.10 0.056 0.20 98 0.11 0.041 0.35 0.28 0.98 < 0.020 0.0056  MaKa_27 Nail earring 95  0.12 0.096 0.25 91 0.12 0.35 0.071 7.2 0.13 0.18 0.037  MaKa_28 Nail earring 95  0.21 0.10 0.39 84 0.17 0.28 0.37 13 0.65 1.0 0.032  MaKa_29 Wire 95  0.053 0.074 0.26 90 0.13 0.30 0.18 8.0 0.44 0.42 0.022  MaKa_30 Bracelet 91  0.060 0.087 0.17 89 0.13 0.12 0.090 10 0.052 0.71 0.11  MaKa_31 Bracelet 91  0.056 0.069 0.19 89 0.12 0.20 0.090 9.2 0.14 0.42 0.062  MaKa_32 Bracelet 91  0.15 0.086 0.20 86 0.11 0.20 0.078 12 < 0.051 0.65 0.12  MaKa_33 Band 91  0.25 0.067 0.16 90 0.13 0.17 0.11 6.9 0.11 2.1 0.047  MaKa_34 Plate 96   0.047 0.054 0.43 94 0.11 0.97 1.4 1.3 1.6 0.33 0.020  MaKa_35 Plate 96   0.36 0.067 0.68 91 0.17 2.1 1.1 1.6 2.6 0.26 0.067  MaKa_36 Plate 96   0.057 0.071 0.40 95 0.12 0.46 0.93 1.4 1.4 0.070 0.036  MaKa_37 Lump 72  0.30 0.070 0.11 77 0.12 0.13 0.10 22 < 0.051 < 0.020 0.011  MaKa_38 Lump 72  0.056 0.096 0.16 85 0.13 0.40 0.11 11 0.22 3.0 0.16  MaKa_39 Lump 72  0.25 0.069 0.10 80 0.13 0.17 0.086 19 < 0.051 < 0.020 0.016 Fig. 4. The SEM images of the nail earrings microstructure with the EDS microareas spots Table 2. The elemental composition (wt%) of the nail earrings by means of SEM-EDS  Microarea  O S Fe Cu Sn Pb  MaKa_27 (1)  14 ... ... 86 ... ...  MaKa_27 (2)  13 ... ... 87 ... ...  MaKa_27 (3)  ... ... ... 95 5.0 ...  MaKa_27 (4)  ... ... ... 95 5.0 ...  MaKa_27 (5)  ... ... ... 94 6.0 ...  MaKa_27 (6)  15 ... ... 82 3.0 ...  MaKa_27 (7)  16 ... ... 78 6.0 ...  MaKa_27 (8)  15 ... ... 78 7.0 ...  MaKa_27 (9)  16 ... ... 76 8.0 ...  MaKa_28 (1)  ... ... ... 92 8.0 ...  MaKa_28 (2)  ... ... ... 91 9.0 ...  MaKa_28 (3)  ... ... ... 90 10 ...  MaKa_28 (4)  ... ... ... 8.0 ... 92  MaKa_28 (5)  ... ... ... 5.0 ... 95  MaKa_28 (6)  ... 22 22 76 ... ...  MaKa_28 (7)  ... 22 3.0 75 ... ...  MaKa_28 (8)  ... 22 3.0 75 ... ...  MaKa_28 (9)  8.0 ... ... 12 ... 80  MaKa_28 (10)  3.0 ... ... 7.0 ... 90 UnauthenticatedDownload Date | 10/21/17 10:41 AM  178 ARCHIVES of FOUNDRY ENGINEERING Volume 17, Issue 3/2017, 175-183   The microstructures observed on cross-sections are coherent and indicative of implementing a 3-step manufacturing process: (1) casting by a lost-wax method followed by (2) annealing and finished with (3) slight forging to a final shape. 3.2. Necklaces  The necklaces belong to the same typological category of the  band ornaments and are indicative of Ha D (Fig. 5). This category tends to concentrate in the Upper and Middle Odra and Warta  basins reaching the regions of Cuyavia and Eastern Pomerania. Fig. 5. The necklaces MaKa_1-3 with close-up on the decoration While the MaKa_1-3 and MaKa_19 were cast in tin-bronze, the MaKa_6 was made in tin-lead-bronze. A suggestive indication for the casting use lies in the numerous inclusions brought to the alloys. Thus it can be assumed that the bronze in-use had been molten and relatively rapidly solidified [11]. The macrostructures are indicative of implementing an intense metalworking. With no visible casting seams left on their surfaces, the necklaces are suggestive of using a lost-wax casting method prior to the plastic working. In conclusion, an identical chaîne opératoire  was applied during manufacturing all the necklaces: (1) casting of a square in a cross-section rod using a lost-wax method, (2) elongating the rod by hammering and annealing with potential water quenching in order to avoid brittle phases, (3) one-way twisting the rod around its own axis followed by (4) forming the circle-shaped ears at the flattened endings and finished with (5) [10]. 4. Discussion  Since long-term tradition of emphasizing the cremation funerary rite by the Chełmno group society required the grave offerings to be combusted with a corpse before burial act was done (Fig. 6a), the majority of jewellery recovered from the urn- field in Mała Kępa did not resist high -temperature corrosion. In general, this type of damage tends to spread equally into the entire surface of the metal to form a final corrosion layer (usually of uniform thickness) with mill-scale and intergranular products. It can be seen from Fig. 6b that after exposed at high-temperature atmosphere, the mill-scale formed on the outer surface of the MaKa_27. It can also be seen in Fig. 6b that intergranular corrosion products accumulated alongside grain boundaries of the artifact. The intergranular corrosion led to a significant reduction in tensile strength and ductility of the metal whereas the mill-scale  brought to the MaKa_27 by high-temperature corrosion reduced the aesthetic quality of the artifact, which in fact is desirable in the case of exhibition and presentation purposes. Fig. 6. a) Cremated human bone with incorporated bronze drops  b) Mill-scale formed on the MaKa_27 with visible intergranular corrosion products The elemental composition results lead to the conclusion that the jewellery from the urn- field in Mała Kępa had been cast with the emphasis on tin- and tin-lead-based bronzes (Table 3). The alloys contained numerous impurities of Ni, As, Ag and Sb and thereby impacting the technological properties. Table 3. The characteristics of the alloys  Alloy   Type  %  Class  %   Cu  8  N/A  8   Cu-Sn  64  Binary  64   Cu-Sn-Pb  22  Ternary  28   Cu-Sn-Sb  3 Cu-Sb-As  3   Interestingly, some of the artefacts have reached relatively low-tin levels of less than 2wt% Sn (MaKa_26 and MaKa_33-35) suggesting that bronzes in-use had been recycled [9].   It is only 2wt% Pb that is required for optimal casting fluidity, however more than 2wt% has been noticed in several alloy  profiles (i.e. MaKa_4, MaKa_6-7, MaKa_21). Although lead improves fluidity and castability, it also brings a deleterious effect to the ductility of the alloys [9]. Nevertheless it was probably an additive used by the Lusatian metalworkers not for economic but rather technological reasons. This, indeed, is the case of the Chełmno group metalworkers where lead-contaminated alloys  become common, alongside mainstreamed tin-bronzes [6, 8]. In order to fingerprint the alloy profile of the nail earrings , the EDS data-set (Table 2) was statistically evaluated using a factor analysis (FA) with a maximum likelihood (ML) extraction method completed by a Varimax rotation with two factors to include in a model. All available elements (O, S, Fe, Cu, Sn and Pb) were used to determine the FA model. The first factor determined by Cu, Sn and Pb impacts a significant intergroup diversity within the Pb inclusions (captured in the MaKa_28 Cu-matrix) and the remaining EDS microareas (Fig. 7), which in fact tend to share the FA space generated by the first factor (α - Cu solution, α -Cu-Sn eutectic and UnauthenticatedDownload Date | 10/21/17 10:41 AM   ARCHIVES of FOUNDRY ENGINEERING Volume 17, Issue 3/2017, 175-183 179 Cu-sulphides). Correlated strongly with the second factor, S and Fe clearly separated the MaKa_28 Cu-sulphides. It may be explained by using the sulphide-rich ore to manufacture the MaKa_28 nail earring, as it has been already suggested. Fig. 7. The FA projection of the nail earrings EDS microareas As the sulphide-rich ores cannot be reduced directly but need to be oxidized, then it became essential for the Lusatian metalworkers to implement roasting prior to smelting. The roasting had taken place at about 350°C in an oxidizing atmosphere and thereby the resulting ore was partially oxidized and desulphurised [9, 11, 12]. This is in agreement with the  presence of sulphur inclusions and above-average iron weight fractions (0.21wt%) in the alloy profile of the MaKa_28. After being cast, the castings underwent annealing cycles improving their ductility and thereby enabling further plastic deformation [9]. Both the nail earrings were manufactured according to the same technological process, using a similar tin- bronze alloy and final shaping by the process of slight forging. In the case of the MaKa_2 and MaKa_3 alloy profiles, the distribution of the minor and trace elements is fairly uniform (Fig. 8). The values of a relevant standard deviation (RSD) lend unequivocal support to these observations. Obviously, the necklaces were made with the same chaîne opératoire  using consistent tin-bronze alloy and final shaping with intense forging, as may be expected if they were worked together in the same casting workshop. This, indeed, reflects the cultural changes initiated during HaD in Central Europe by the Northern Barbarians practicing their own and srcinal style of bronze manufacturing [13]. Both necklaces contain relatively low amounts of arsenic and antimony (0.31wt% - 0.38wt%) with the total sum of impurities averaging 2wt%. Since two cycles of re-melting and hot-working in air can impact the loss of As and Sb content until less than 1wt% is left, it is likely that re-melting of scrap bronze and subsequent smelting could have been indeed applied here [11]. Furthermore, evaluating the ornaments by means of a   discriminant analysis (DA) with regard to the level of the implemented plastic forming yields some interesting conclusions. Based on the macrostructural observations the ornaments used here have been arbitrally classified into two groups (weak or intense plastic working). The DA model was dedicated to the artefacts preserved in a well-distinguishable form and it was determined using all avaible elements (ED-XRF), excluding Cu. A greater tendency shared by the ornaments treated with intense  plastic forming to contain relatively higher amounts of iron and cobalt is responsible for the significant (Wilk's lambda Λ W =0.03) intergroup discrimination (Fig. 9). Fig. 8. The alloy profiles of necklaces MaKa_2 and MaKa_3 with corresponding values of RSD (Sn content given as Sn·10 -1 wt%) Therefore, taking into account the manner the ornaments were formed, it is almost certain that the smelting process had impact on the differences in the iron content. Cobalt, having both  provenance and technology marking potential, may have also contributed to the intergroup discrimination apparent in the DA scatterplot [12, 14]. While the ornaments where intense metalworking was implemented average Fe around 0.08wt% (with Co about 0.06wt%), the Fe content in those which underwent weak plastic forming amounts to approximately 0.2wt% (with Co about 0.1wt%). This points to the alloying technology approaches taken by the Lusatian metalworkers, one that suggests implementing short refining process and the other with the more sophisticated process involving slagging off [14]. It must also be stressed here that the first scenario is in a good relation to the necklaces cast in re-melted scrap bronze (MaKa_2 and MaKa_3), while the second is UnauthenticatedDownload Date | 10/21/17 10:41 AM
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x