Contents lists available at ScienceDirect Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep Reconstructing bronze production technology from ancient crucible slag: experimental perspectives on tin oxide identification Frederik W. Rademakers a, ⁎ , Carlotta Farci b a Division of Geology, KU Leuven, Leuven, Belgium b Department of Archaeology, University of Exeter, Exeter, UK 1. Introduction Bronze production has been marked as one of the major technolo- gical transitions in the archaeological record worldwide. From the onset of archaeological materials analysis, copper alloys have been a favoured subject (Pollard, 2013), and they remain so to this day. While major diachronic trends in the use of different copper alloys such as arsenical, tin and leaded bronze have been mapped for various archaeological cultures, the basic production technology of these alloys has not en- joyed anywhere near as much attention (Pigott et al., 2003; Rovira, 2007). However, these production techniques may vary significantly, and their identification provides insights into shifting technological choices within particular contexts, as well as the spread of metallurgical technology in a broader perspective. Furthermore, these considerations inform debates on the interpretation of metal provenance as well. Essential evidence towards understanding the technological choices underlying bronze production is reflected in crucible remains, in ad- dition to casting and shaping techniques attested in moulds and final objects. This paper focuses on the analysis of crucible slag as a tool to distinguish between different tin bronze (henceforth: bronze) produc- tion processes. To this end, controlled laboratory experiments are em- ployed to highlight important tin oxide (SnO 2 unless otherwise noted) phases embedded in crucible slag and their diagnostic value for re- cognising different production techniques. This research is prompted by the authors' previous research on an- cient bronze production evidence, whereby the distinction between bronze recycling and alloying was elaborated based on archaeological remains from ancient Egypt, Phrygia and Spain (Farci et al., 2017, Rademakers, 2015, Rademakers et al., 2017a, 2017b, 2018 1 ), and an increasing attention for this subject witnessed in recent publications (discussed below). The main ancient techniques for bronze production are: • Alloying of copper and tin metal • Alloying (cementation) of copper with mineral cassiterite (SnO 2 ) • Co-smelting of copper and tin ores • Recycling of existing bronze, with the possible addition of copper/ tin metal/ore Bronze melting and alloying crucibles have previously been ana- lysed by other researchers in varying degrees of detail. Of particular interest here are attempts to distinguish between bronze production modes by recognising particular phases in crucible slag: tin oxide is most often cited in this context (see, for example, Adriaens, 1996, Bandama et al., 2015, Benvenuti et al., 2000, 2003, Chirikure et al., 2010, Cooke and Nielsen, 1978, Denbow and Miller, 2007, Dungworth, 2000, 2001, Eliyahu-Behar et al., 2012, Erb-Satullo et al., 2015, Figueiredo et al., 2010, 2017, Garbacz-Klempka et al., 2017, Mahé-Le Carlier et al., 2001, Merideth, 1998, Murillo-Barroso et al., 2010, Nezafati et al., 2009, Papadimitriou, 1992, Renzi et al., 2009, Renzi and Rovira, 2016, Rostoker et al., 1983, Rovira, 2007, Rovira et al., 2009, Rovira, 2011, 2011–2012, Rovira et al., 2009, Yener and Vandiver, 1993, Valério et al., 2013, Wang et al., 2016). While tin oxide is noted upon in a multitude of publications, the diagnostic significance of its variable occurrence (particularly morphology) is only explicitly dis- cussed in more detail by few. Dungworth (2000) highlights the appearance of “highly euhedral [tin oxide] inclusions”, “present as rhomboids or as needles” in experimen- tally cast bronze. They are interpreted as resulting from oxidation during casting, rather than the use of cassiterite for alloying. An important https://doi.org/10.1016/j.jasrep.2018.01.020 Received 6 October 2017; Received in revised form 12 January 2018; Accepted 14 January 2018 ⁎ Corresponding author. E-mail address: frederik.rademakers@kuleuven.be (F.W. Rademakers). 1 Rademakers et al. (2018) was formerly cited in other publications as Rademakers et al. (2018) while still in press, but the final revised version of this paper has recently appeared in print. Journal of Archaeological Science: Reports 18 (2018) 343–355 2352-409X/ © 2018 Elsevier Ltd. All rights reserved. T