Journal of Archaeological Science 156 (2023) 105805 0305-4403/© 2023 Elsevier Ltd. All rights reserved. Are crucible steel ingots isotopically homogenous? AMS radiocarbon measurements on ingots from Telangana, India Meghna Desai a, * , S. Jaikishan b , Thilo Rehren a, ** a The Cyprus Institute, 2121, Aglantzia-Nicosia, Cyprus b Bhavan’s New Science College, Hyderabad, India A R T I C L E INFO Keywords: Crucible steel ingots Fractionation Isotope Homogeneity AMS Radiocarbon Wootz ABSTRACT Radiocarbon analysis is increasingly used to directly date archaeological and historical metal objects, ranging from low-carbon bloomery iron to steel and cast iron. However, little is known about the isotopic homogeneity of iron-carbon alloys, particularly relating to the formation of primary cementite during crystallisation. Here, we present 14 C measurements for five crucible steel ingots and one crucible steel object from Telangana in south- central India. Two of the ingots were analysed twice. The results show a very wide scatter of 14 C dates, far exceeding the expected age range for this assemblage. The repeat analysis of one of the ingots gave also widely different results, indicating a fundamental problem with the 14 C analysis of crucible steel. We discuss the various factors that could have influenced the measured isotopic values, including variability in raw material, sampling contamination, and fractionation during the cleaning of the metal in hot acid leading to excessive sample mass losses prior to the extraction of carbon from the metal. We argue that mass-dependant fractionation of the different carbon isotopes between austenite and cementite during solidification of the ingot, and subsequent selective dissolution of one metal phase over the other, led to a distortion of the 14 C signature to seemingly older ages. We recommend further research to explore the compound-specific isotopic signature of high-carbon iron alloys and the effect of selective corrosion on such material, to reduce potential errors in 14 C dating of steel and cast iron. 1. Introduction Crucible steel is the only pre-industrial steel that was liquid during its production, ensuring an unparalleled homogeneity and purity of the metal. Its high carbon content made it harder than normal bloomery iron and steel, without having the brittleness of cast iron. It was therefore highly prized and used for sharp tools and weapons, famously the Damascus blades of the Middle Ages, but also for flint strikers, protective armour, metal files, agricultural knives, and many other objects. The first heyday of crucible steel making lasted from the 10th to early 13th century CE in Central Asia. Following the destructions wrought by the Mongol invasions in the 1220s, crucible steel making all but dis- appeared, with the only known surviving production in southern Iran (Alipour and Rehren 2015). A second flourishing occurred in late me- dieval and early modern India and Sri Lanka, roughly from the mid-2nd millennium CE to the late 19th century CE, probably building on 1st millennium CE roots in Sri Lanka (Juleff 2015). The outstanding quality and perceived mysterious production of crucible steel have fascinated European scientists and entrepreneurs alike, stimulating ground-breaking research in iron and steel metallurgy during the In- dustrial Revolution, in Russia, France and the UK. Almost all analytical research on crucible steel focussed either on finished objects such as blades and other arms and armour, or on a few ingots that were procured as curios by European travellers, colonial agents, and scientists. The chronology and geography of crucible steel making are relatively recent topics of research (Craddock 1998; Rehren and Papachristou 2003), and much new and unexpected evidence is still emerging (Rehren and Nixon 2017; Güder et al., 2022). Here, we focus on the dating of the largest-known hoard of crucible steel ingots, worldwide, from Konasa- mudram, Telangana, south-central India. In the last five decades, crucible steel or wootz production in Telan- gana (see Fig. 1) has gained considerable attention, producing a wealth of research from this region (Lowe 1989; Lowe 1995; Jaikishan 2007; Jaikishan and Balasubramaniam 2007; Juleff et al., 2011; Girbal 2017, * Corresponding author. ** Corresponding author. E-mail addresses: m.desai@cyi.ac.cy (M. Desai), th.rehren@cyi.ac.cy (T. Rehren). Contents lists available at ScienceDirect Journal of Archaeological Science journal homepage: www.elsevier.com/locate/jas https://doi.org/10.1016/j.jas.2023.105805 Received 29 May 2023; Accepted 1 June 2023