Platinum-group elements (PGE) are strongly chalcophile and are therefore potentially sensitive indicators of processes involving segregation and accumulation of sulfide melts from silicate magmas. Over 500 new high-precision PGE data for komatiites and komatiitic basalts, spanning a wide range of emplacement and crystallization histories, have been combined with literature data on PGE in magmatic systems from other barren and variably mineralized environments, to test the effectiveness of PGE geochemistry as an indicator of processes forming magmatic sulfide ores.<br/><br/>Results show that PGE depletion in S-poor komatiites and komatiite basalts spatially and genetically associated with Fe-Ni-Cu sulfide mineralization is not as common or as strong as expected: samples displaying orders of magnitude depletion in PGE represent less than 10 percent of any given data set from any location. The data confirm that most, if not all, komatiites were sulfide undersaturated when they separated from their sources and remained undersaturated on eruption. Some ore-bearing komatiite sequences display no detectable depletion, and the degree of PGE depletion is commonly less than expected based on modeling using experimentally determined partition coefficients. PGE enrichment is more common and spatially widespread than PGE depletion, commonly representing a better approach to lithogeochemical exploration, even where samples containing anomalous Ni or S contents are absent. PGE enrichment and/or depletion associated with sulfide enrichment and/or segregation can be discriminated from secondary hydrothermal and/or metamorphic processes by covariance of all PGE, with the exceptions in some cases of Ir, Ru, and Os whose abundances may be complicated by the presence of saturation in and accumulation of Ir-Os-rich liquidus phases. Variations attributable to other magmatic processes, such as olivine accumulation and fractionation, can be distinguished by variations in PGE/Ti ratios and strong correlations between Pt/Ti, Pd/Ti, and Rh/Ti ratios in mineralized systems.<br/><br/>The degree of PGE depletion is consistent with the relatively low R factor estimated for many komatiite-hosted deposits, which fall in the range of 20 to 200 for Thompson, 100 to 500 for Kambalda, and 300 to 1,100 for Raglan, implying that the volume of silicate magma that interacted with sulfide liquid was relatively small. This is also consistent with the relatively small proportion of komatiites displaying PGE depletion within ore-bearing flow sequences, as only magmas in ore-forming channels or conduits will interact with sulfides. False negatives, i.e., mineralized komatiite sequences with no detectable PGE depletion, are associated with systems characterized by high R factors.<br/><br/>Basalts and komatiitic basalts show more complex patterns of variation, which can broadly be divided into three categories: (1) systematic PGE depletion over a range of Mg numbers, as in MORB suites, consistent with retention of sulfide in the mantle during partial melting; (2) increasing PGE depletion with decreasing Mg numbers in large igneous province (LIP)-associated basalts, interpreted to reflect attainment of sulfide saturation during fractionation with subsequent cotectic olivine-sulfide segregation; and (3) variable PGE depletion over a range of Mg numbers in komatiitic basalts (e.g., Raglan) interpreted to reflect ore-forming sulfide incorporation and segregation processes. The results of this study confirm that the PGE geochemistry of komatiites and basalts is a powerful indicator of sulfide saturation and ore-forming processes, but that it must be interpreted with the context of physical volcanologic and fluid dynamic processes.
