Complexity in the Au–Ag–Hg system: New information from a PGE ('osmiridium') concentrate at Waratah Bay, Victoria, Australia

Au–Hg–Ag phases have been described from a variety of metallogenic orebodies and the placer deposits derived from them. In many documented placer deposits, the phases typically occur intergrown as ‘secondary’ rims to primary Au–Ag grains. The origin of these rims has been ascribed to supergene redistribution reactions during deposition or to the effects of amalgamation (i.e. use of mercury) during mining for gold. Difficulties in determining compositions and crystal structures on such a small scale have made full characterisation of these phases problematic. This paper describes a new occurrence of these phases, found by accident during investigation of a historical concentrate of ‘osmiridium’ containing a number of gold grains from beach sands at Waratah Bay, in southern Victoria, Australia. The phases occur as rims to gold grains and are intergrown on a scale of tens of micrometres or less. Application of electron microprobe analysis (EPMA) and limited electron back-scattered diffraction (EBSD) was required to characterise them. These techniques revealed the presence of the approved mineral weishanite (Au–Hg–Ag) and a phase with compositional range Au2Hg–Au3Hg surrounding primary Au–Ag (electrum) containing trace amounts of Hg. EBSD analysis showed weishanite is hexagonal P63/mmc and Au2Hg to be hexagonal P63/mcm. Comparison with published data from other localities (Philippines, British Columbia and New Zealand) suggests weishanite has a wide compositional field. Textures shown by these phases are difficult to interpret, as they might form by either supergene processes or by reaction with anthropogenic mercury used during mining. However, in the absence of any historical evidence for the use of mercury for gold mining at Waratah Bay, we consider the formation of the Au–Hg phases is most probably due to supergene alteration of primary Au–Ag alloy containing small amounts of Hg. In addition to revealing some of the reaction sequences in the development of these secondary Au–Hg–Ag rims, this paper illustrates methods by which these phases can be more fully characterised and thereby better correlated with the Au–Hg synthetic system.