Archaeometallurgy of Copper and Silver Alloys In

42 ADVANCED MATERIALS & PROCESSES | JULY/AUGUST 2021 M F i g . 1 — Ages Copper, Iron and the Bronze, as prehistory Neolithic by referring to the main periods of post- metals of technology in ancient societies is importanceshown The B.C. 6000 tive metals, may be traced back to about na- of consolidation and melting lurgy, ic times, and the earliest process metal- 60 years there have been international been have there years 60 to 50 last the Over archaeometallurgy. of remit major a is world ancient the in was effectively lost after 4000B.C. around 4500 B.C. 4500 around Balkans, the in earlier much smelted were bronzes tin complex that suggest data recent However, B.C.). (1500 iron per (6000 B.C.), bronzes (3500 B.C.), and nologies in the Near East areas are: cop- tech- these of beginnings the for dates Russell Wanhill, Emmeloord, theNetherlands Omid Oudbashi,ArtUniversity ofIsfahan, Iran providinglegacya lasting for modern civilizations. beganin the World Old about 8000 years ago and develo Theproduction and processing advanced of materials, na WORLDOLD THE IN ALLOYSSILVER AND COPPER OF ARCHAEOMETALLURGY Understanding process metallurgy Schematic of the main aspects of archaeometallurgi ment of societies from Neolith- from societies of ment develop- the of part essential an constitute alloys and etals [2] , but this technology technology this but , [1] .

Approximate bles, andfurnaces (Fig.1). cruci- slags, e.g., equipment, and tools, by-products, (ii) and production, final to materials raw from artifacts, and es metallurgicalprocess-(i) of studies tific scien- promote and establish to efforts a b ol prily ucsfl and successful partially only be may scientific knowledge these experiments modern with Even furnaces. and cibles cru- ancient-style in ores from metals smelt to experiments rometallurgical py- i.e., archaeometallurgy, perimental overcomedemonstratedwell are ex-by and restoration techniques aspect is directly linked to conservation latter This millennia. the over rosion) cor- (especially deterioration eventual their and themselves artifacts the also and involved, processes complex the understand better to instruments tory labora- and methods scientific modern h dfiute ta hd o be to had that difficulties The of range wide a use studies These cal studies. Adapted from Bayley et al. . ped over ped many millennia, [3] melymetals and alloys, . that the ores were mined from copper from mined were ores the that importanttonoteis It B.C. 4000 around copper of began copper derive to (smelting) ores reduction and B.C., 6000 around began casting and melting by ANCIENT COPPER ALLOYSCOPPER ANCIENT n Cuau ad s ae t about B.C. to 8000 dated is and Caucasus and East Near the from comes objects tive decora- and small make to copper tive burial before archaeological recovery. long-termto owing embrittlement, and corrosion including post-pro- problems cessing mentions also and World, nin bozs n sle i the in silver and of bronzes ancient processing and production the of cient metalworkers. an- of skills derived empirically the of appreciation veritable a enable iments exper-completely.failSuch sometimes h frt vdne f sn na- using of evidence first The This article gives a brief overview overview brief a gives article This [1] Poesn ntv copper native Processing . smelting. before (roasted) be oxidized to had these and its, mining depos- sulfide reachedthe continued ever, How- furnaces. and bles cruci- smelting to added simply be could These ide. of coppercarbonates ox-and mostly consisted ers lay- upper weathered where the deposits, sulfide Old ADVANCED MATERIALS & PROCESSES |JULY/AUGUST 2021 2 5 . . [4] . The [12] [4] . The sec- Tylecote [10] . Silver was [11] , and the Andean [8], after tin bronzes be- . There remains the possi- [10] . Beginning before 3000 B.C., ze Age. Adapted from from Adapted Age. ze 1000 years [12] Owing to native silver’s scarcity, The majority of Near East tin bility that smelting arsenical ores was abandoned in Eurasia owing to health concerns. However, arsenical es bronz- were still being produced in the LBA (Fig. 3), predominant. came bronzes have tin contents about less 12 than wt%, typically ranging from 5‒10 wt% from about 3000 B.C. there is limited evidence of its direct use direct its of evidence limited is there for artifacts, a few of which have been dated to 4300‒4000 B.C. ond hypothesis is disfavored by an extensive study and comparison of the mechanical properties of arsenical and tin bronzes region region study reinforces this ANCIENT SILVER ALLOYS lead lead obtained from smelting argentif- erous lead ores was further processed by cupellation the to This extract silver. process became the primary source of ancient silver and silver artifacts, though al- some artifacts were obtained ores smelting of silver direct from more more abundant as a minor component in the ores of other metals, especially lead earliest earliest EBA alloys have lower tin contents, 1‒3 wt%; and there sional are exceptions, occa- the high-tin alloys already mentioned. Hence most of the materials and artifacts would have had microstruc- single-phase homogeneous tures after working and annealing, very different from the inhomogeneous as- (Fig. 4). structures cast distinct distinct possibility . ? [7] [8,9] . [10] . [7] Schematic copper smelting furnace, Crete, Late Bron Late Crete, furnace, smelting copper Schematic

Returning Returning to Eurasia, two more The more intriguing question is important important questions arise. Why did tin bronzes become the main type, large- ly replacing arsenical bronz- antimony did why and B.C., 2500 bronzes after es almost disappear after 2000 B.C. Also, Also, although digressing here from the Old World, there is evidence convincing that the Andes region arsenical bronzes containing 0.5‒2 wt% arsenic were intentionally produced from 850 A.D. for cold-hammering about into cul- turally desirable small implements and materials thin sheet Possible Possible answers have been given, but antimony there is no Firstly, consensus. may bronzes have been supplanted because their lesser hardness, and hence lesser strength, made them unsuitable for tools or weapons. This could have resulted in a lack of demand and trade not is this though bronzes, tin of favor in known (yet) the predominance of tin bronzes over later the in beginning bronzes, arsenical EBA. There are three basic hypotheses: (i) tin bronzes were intentional alloys but arsenical bronzes were not; (ii) tin had bronzes superior mechanical properties; (iii) smelting arsenic-containing ores resulted in poisonous fumes that became recognized as a health ard. The first haz- hypothesis has been dis- cussed already: intentional alloying to obtain Eurasian arsenical bronzes is a were were often present in copper-bearing Ear- of analysis hand, other the On ores. that shows Iran from slags Age Bronze ly speiss, an iron-arsenic alloy, was probably added to copper ore or during re- melting to obtain arsenical bronzes Fig. 2 — 2 Fig. . [6] , and these could could these and , [4] , since these elements [5,7] , and with intermittent an- intermittent with and , [5] The history of ancient bronzes Large Large “oxhide-shaped” copper in- Early processing was done us- Near Near East Early Bronze Age (EBA: is complex, spanning a “classic” peri- od of more than 2000 years in Eurasia (3300‒1200 B.C.). Many issues are still unresolved, despite extensive the studies Perhaps century. 20th early the since the whether is question important most copper in elements alloying of presence inten- became or accidental always was types main three the Considering tional. of bronzes, antimony bronze, arsenical bronze, and tin bronze, the evidence of in- is bronzes tin for alloying intentional controvertible. However, deliberate alloying with antimony and arsenic can be questioned gots gots were widely used LBA the in in items trade Eurasia as ing crucibles containing crushed ores and charcoal, with forced airflow pro- vided by bellows-powered blowpipes. Later on, crucible and hearth furnaces using forced air via tuyères provid- ed Figure conditions. 2 more controlled is a schematic of a from smelting the furnace Near East Late (LBA: Bronze 1550‒1200 Age B.C). Besides metal production, such a furnace could initial be used to remelt additions of copper metal other before tapping into clay or stone molds to cast ingots or facts, arti- for example, vessels, tools, and ornaments. 3300‒2100 B.C.) ingots were bly forged by cold-working rather than proba- hot-working be remelted with additions of tin or tin oxide (cassiterite), and possibly other alloying metals, to produce bronze ingots or cast artifacts including vessels, and ornaments. weapons, tools, ANCIENT BRONZES nealing, depending on the metalsmith’s metalsmith’s the on depending nealing, experience with the materials and the required artifacts. This practice tinued con- well into the Iron Age, grad- beyond have would Hot-working B.C. 1500 ually developed as an alternative, ex- cept for high-tin bronzes, because “hot tem- high at cracking (brittle shortness” peratures) would become increasingly 8 wt% above with tin contents likely 62 ADVANCED MATERIALS & PROCESSES | JULY/AUGUST 2021 and coring. some retained cold-work, and (b) shows interdendrit annealed vessel, 8.67 wt% Sn. (b) An Iron Age I as- Conophagos Fig. 5 — Fig. 4 Fig. 3 2.17 wt% As. (b) An LBA worked and annealed bowl, 2 ec te ed o a ih tempera- surface a via drained high litharge The ture. a for need the hence 880°C, at melts which (PbO), litharge to lead Bellows- the oxidized tuyères fuel. powered wood using perature tem- high a to remelted was This lion. bul- lead smelted enriching for hearth stage first a of schematic a is 5 Figure separatehearths. three employing cess (a) (a) Cupellation was a multistage pro- multistage a was Cupellation — —

Schematic of Stage I cupellation about 500 B.C., La Metallographs of two binary Cu-Sn alloy artifacts f Metallographs of two binary Cu-As alloy artifacts f [13] . (b) (b) cast tool, 10.83 wt% Sn. Note that (a) shows hs rd wr rpael removed, repeatedly were rods These form layered litharge cones on the rods. to poles) wooden B.C., 1000 (before it into rods iron dipping lithargeby removed was the here but oxidized, again and hearth second a transferredto was lead enriched the Then stage. second sil- sufficient the for obtained until was lead ver-enriched added was lion bul- More discarded. was and groove ic (α + δ) eutectoid, shrinkage porosity .10 wt% As. rom Iran. (a) An EBA as-cast axe head, rom Iran. (a) An EBA worked and urion, Greece. Adapted from 00 B.C. 3000 about since done been have to appear additions Copper coins.artifacts and ty lower-quali- make to amounts larger in resistancealso and alloys, high-silver in wear and strength the increase to ably prob- most additions, deliberate cate indi- wt% 0.5‒1 above contentscopper that shown have studiesSeveralnickel. and zinc, tellurium, arsenic, antimony, ally below 1 wt% for each), and traces of (gener-leadcopper, and bismuth, gold, containsally minor-to-trace amounts of usu- It purity. 95% above silver ducing in thecupelwall pores by absorbed being PbO maining re- the ingots, obtain to hearth other an- in refined further and melted were globules of stage,thirdnumber the a In hearth.the stageon globule silver a left second this Eventually re-dipped. rods the and discarded, cones litharge the exquisite craftsmanship. with vessels were thin-walled artifacts high-quality Many produced. also were objects silver cast although ing, anneal- intermittent with working cold by ingots from made commonly were AND EMBRITTLEMENTAND CORROSION PROBLEMS: POST-PROCESSING tural embrittlement, most probably due microstruc- intergranular of evidence planes). Also, some silver artifacts show slip (along transgranular and granular inter- both is damage SCC The burial. during cups) and vessels (e.g., artifacts hollow thin-walled on forces external also and work cold retained by moted pro- is which (SCC), cracking corrosion stress and corrosion general both go under- may silver and bronzes ancient both Basically, damage. the of details on not but techniques restoration and conservation on concentrate usually they and damage, burial the on tions publica- numerous are There 6. Fig. in given is B.C., century 2nd or 1st the to high-silver Gundestrup Cauldron, dated famous the from example An covery. re- before burial of millennia to owing damage embrittlement and corrosion bronze and silver artifacts have suffered Cupellation is effectivevery in pro- notntl, ay ancient many Unfortunately, [14] Te riat themselves artifacts The . . ADVANCED MATERIALS & PROCESSES |JULY/AUGUST 2021 2 7

ISIJ (Iron University University Studies Studies in Journal of to to be pub- Ligas Metálicas, 67, p 103-115, ASM ASM Handbook, (1), p 41-49, 2013. Verlag Marie Leidorf Von Von Baden bis Troia: Fractography, Fractography, Join the conversation about (5), p 1085-1092, 2014, https://doi. historic historic metals. The archaeometallurgy tallurgy community on ASM Con- nect welcomes new participants. Visit connect.asminternational.org, search for the Archaeometallurgy Committee, and conversation! get into the GET GET ENGAGED, GET INVOLVED, GET CONNECTED Die Anfänge in der Eura-sien, Silbermetallurgie Ressourcennutzung, Metallurgie Wissenstransfer, und GmbH, Rahden, Germany, 2016. p 41-58, 12. P. Craddock, Production of Silver Across the Ancient World, and Steel Institute of Japan) 54 Journal, org/10.2355/isijinternational.54.1085. 13. C.E. Conophagos, Antique et la Le Technique Grecque la de Laurion Production del’ 1980. Greece, Hellados, Athens, Argent, Ekdotike 14. N.H. Gale Ancient and Egyptian Z.A. Silver, Egyptian Archaeology, Stos-Gale, 1981. 15. R.J.H. Wanhill, T. Hattenberg, and J.P. Northover, EBSD Deformation of and Precipitation Corrosion, in the Gundestrup Cauldron, Investigação e Conservação, p 47-61, 2008. Portugal, of Porto, 16. R.J.H. Wanhill, Cracking Stress in Corrosion Ancient Silver, Conservation, 58 17. R.J.H. Wanhill and Fractography O. of Ancient Oudbashi, Metallic Arti- facts: Archaeometallurgical Analysis Fracture and Restoration servation and Aspects, Con- Volume Volume 12, lished in 2022.

(7): (3), (3), 15 22

Journal of Old World equiaxed equiaxed (6), p 1717- 39

Journal of Field Deutsches Bergbau (4), p 477-514, 1996. Journal of the Institute of Metals, of Institute the of Journal corrosion damage in a sample sample a in damage corrosion for for Archaeology Guidelines, Heritage, National Monuments English Record Centre, Great Western Village, Kemble Swindon, UK, 2001. Drive, 4. R.F. Tylecote, A lurgy, Second History Edition, of Maney Publish- Metal- UK, 2002. ing, London, 5. J.P. Northover, Properties and Use of Arsenic-copper Alloys, Archaeometallurgy, 111-118, p Germany, Bochum, Museum, 1989. 6. W.L. Kent, Bronze, The Brittle Ranges of 35, p 45-53, 1926. 7. G. Dardeniz, Why Antimony-bearing Alloys in Bronze Did Age the Use of Anatolia fall Dormant After the Bronze Early Age? A Case (Çorum, from Resuloğlu Turkey), PLoS One, 1727, 2012. 9. C.P. Thornton, The Complex Metallurgy on the Emergence Iranian Pla- of Escaping the Paradigm, teau: Levantine Journal of World Prehistory, e0234563, 34 pages, 2020, https://doi. org/10.1371/journal.pone.0234563. 8. T. E. Pernicka, Rehren, Large Scale Smelting Speiss L. of and Arsenical Copper Boscher, at Bronze Early Age Arisman, and Iran, Archaeological Science, p 301-327, 2009. 10. H.N. Lechtman, Dirty Arsenic Copper or Chosen Bronze: Alloy? A View from the Americas, 23 Archaeology, 11. S. Hansen and B. Helwing, ngle map showing retained cold-work as as cold-work retained showing map ngle spaced irregular yellow boundaries). The The boundaries). yellow irregular spaced d cold work and has been identified as stress stress as identified been has and work cold d (b)

ASM ASM . (a) Inverse pole figure color-coded map, showing showing map, color-coded figure pole Inverse (a) . 87, p 1030- [15] . The impli- Fractography, [17] . Omid Oudbashi, Oudbashi, Omid [16] Antiquity, Antiquity, 83, p 1012-1022, 2009. [email protected], [email protected], www. Volume 12,

Antiquity, Bayley, Bayley, D. Dungworth, and

Electron back-scatter di"raction metallographs of of metallographs di"raction back-scatter Electron The authors recently prepared an prepared recently authors The . 6500 Ago, Years (a) 1045, 2013. 3. J. which discusses the types of damage in ancient metals, including tin bronze and high-silver artifacts 2. 2. M. Radivojević, et al., Tainted Ores and the Rise of Tin Bronzes in Eurasia, c V.C. Pigott, Development Pigott, of V.C. Metallurgy in Eurasia, ~AM&P Note: article for a new (2022) edition of Handbook, to to long-term segregation of lead, origi- nally in retained solid solution after cupellation and working and annealing. Fig. 6 — 6 Fig. from the Gundestrup Cauldron Gundestrup the from For more information: information: more For cations cations of all these types of damage for conservation and restoration are dis- cussed in the article. associate professor, department conservation of cultural and of historical properties and archaeometry, Art Uni- versity of Isfahan, P.O. Box 1744, Isfa- han, Iran, 1. B.W. Roberts, C.P. Thornton, and References aui.ac.ir. aui.ac.ir. Russell principal Wanhill, research scientist, aerospace emeritus vehicles division, Mark- and Royal Amsterdam Centre, Aerospace Netherlands nesse, the Netherlands, gmail.com rjhwanhill@ grains and annealing twins. (b) Boundary rotation a rotation Boundary (b) twins. annealing and grains (narrowly twins deformation and (red) dislocations retaine with associated preferentially is corrosion (SCC) cracking corrosion S. Paynter, Archaeometallurgy, Centre