Non-destructive surface analysis of accelerated corroded copper alloys Panayiotis Papandreopoulos, Maria Koui, Dimitrios Yfantis and Theophilos Theophanides School of Chemical Engineering, National Technical University of Athens, Athens, Greece Abstract Purpose – The purpose of this paper is to investigate the surface corrosion products of copper alloys by non-destructive techniques (NDT) and correlate them with their bulk composition. Design/methodology/approach – Specimens of copper alloys, whose compositions were close to those of ancient copper-based artefacts, were left to be corroded in simulated soil solution containing ammoniacal buffering solution of pH ¼ 10 in 1:1 ratio, in order to accelerate the corrosion rate. The elemental compositions of the surface corrosion products were determined versus time using X-Ray Fluorescence Spectroscopy, and the surface morphology by Scanning Electron Microscopy with Energy Dispersive X-Ray Micro-detector methods, and the results were compared to the bulk composition, as measured by Atomic Absorption Spectroscopy and Inductively Coupled Plasma Atomic Emission Spectroscopy. Findings – During the corrosion evolution of the copper alloys in the corrosive solution, transitional phenomena were observed such as an initial decrease of the copper concentration with a simultaneous increase of the concentrations of the secondary alloying metals (Sn, Zn and Pb). After 30-60 immersion days, the alloy concentrations were stabilised. Originality/value – The results of this research could contribute to the non-destructive characterisation of copper-based ancient artefacts (from which the taking of samples is not allowed). Keywords Alloys, Corrosion, Artefacts, Copper alloys, Corrosion in soil, X-Ray fluorescence spectroscopy, Scanning electron microscopy with energy dispersive X-Ray micro-detector Paper type Research paper 1. Introduction Copper was one of the first metals widely used in human technology, which gave its name to a whole era of human civilization, known as “Bronze Age”. Historically speaking (Yfantis, 2006), the use of copper alloys started in the Middle East and Egypt from the end of the fourth millennium. The “Bronze Age” was the period of Greek civilization between 3000 and 1050 BC. Copper is a metal that has the ability to passivate, namely the oxides and hydroxides of copper that are formed on its surface act as protective layers and do not allow any further corrosion. However, copper (Cu) can be corroded under humidity, CO 2 and SO 2 environments (Yfantis, 2008). From numerous analyses that were done in Greece (Mangou and Ioannou, 1997, 2002; Papadimitriou, 2001; Koui et al., 2006) it was concluded that in the Early Copper Age (3000-2000 BC) arsenic copper alloys were used with As contents between 1 and 6 per cent and tin (Sn) as a trace element (, 1 per cent). Bronze, which is mainly a copper/tin alloy, was used in the Middle Copper Age (2000-1600 BC), where Sn usually varied between 4 and 12 per cent. In the analyses that were done in ancient objects in other parts of the world (Araujo et al., 2004; Attanasio et al., 2001; De Ryck et al., 2003; Mattsson et al., 1996; Olariu et al. , 2003; Ponting, 2002; Vittiglio et al., 1999) the average ancient bronze contained: Cu 80-95 per cent, Sn 4-12 per cent, Pb 0-5 per cent and As 0-6 per cent. Brass, which is mainly a copper/zinc (Zn) alloy, was widely used by Romans and not to the same extent as bronze. Analyses (Mattsson et al., 1996; Ponting, 2002; Vittiglio et al., 1999) showed that the average ancient brass contained: Cu 70-90 per cent, Zn 5-25 per cent, Sn 0-5 per cent and Pb 0-5 per cent. The most widespread techniques and methods used in the analysis of metals objects are destructive methods (as is the atomic absorption and emission spectroscopy). In the case of archaeological artefacts, where sampling is not allowed, there is a need for the use of non-destructive techniques (NDT). Until recently there were not enough comparative analyses that correlated the surface corrosion products of copper-based objects determined by NDT with their bulk composition. The present work was a continuation of a research attempt (Koui et al., 2007; Papandreopoulos et al., 2009) towards this target. 2. Methodology 2.1 Materials and methods Three alloys with the following commercial names and compositions close to the respective ones of ancient copper- based artefacts were selected: G12 (Cu 86-87 per cent, Sn 11-12 per cent and Pb , 1 per cent), RG7 (Cu 82-87 per cent, Sn 6-7 per cent, Pb 4-7 per cent and Zn 3.5-4 per cent) and MS58 (Cu 58-59 per cent, Zn , 40 per cent and Pb 1-2 per cent). The specimens were cut in form of disks with a diameter between 40 and 45 mm and thickness of about 3 mm in order to be adapted in all holders of the analytical instruments. The current issue and full text archive of this journal is available at www.emeraldinsight.com/0003-5599.htm Anti-Corrosion Methods and Materials 59/3 (2012) 110–120 q Emerald Group Publishing Limited [ISSN 0003-5599] [DOI 10.1108/00035591211224654] 110