Understanding genetic processes at Norilsk by combining in-situ LA-ICP-MS and S isotopic data on different sulfide textures

This study presents the results of in-situ characterisation of different coexisting sulfide textures present within the globular disseminated ores of the Norilsk Ni-Cu-Co-PGE mineralised intrusions. The aim is to gain a better understanding of the genetic processes at play for these different textures and for the ore bodies as a whole. Two specific techniques were used: in-situ laser ablation ICP-MS, and in-situ S isotopic analyses. Both techniques provide information on specific genetic processes: (1) S assimilation and crustal contamination and (2) interaction between the sulfide melt and the silicate melt within the magmatic plumbing system. S isotopic compositions of the sulfides allow interpretations on the potential source of the S, and of re-homogenisation of this S isotopic signature of the sulfides during interaction with the silicate melt. On the other hand, trace element compositions of the main sulfide minerals, especially Pd and Se contents of pentlandites, can be proxies for sulfide melt composition and hence can provide indications of relative variation in the efficiency of sulfide-silicate element partitioning at the scale of individual blebs. These two datasets were also combined with a 3D characterisation of the sulfide textures within representative samples. This 3D characterisation allows the identification of three textural types: large globules, interstitial frameworks connected to globules, and isolated disconnected blebs. In most cases, sulfides of all textural types within individual samples present similar S isotopic compositions and similar Pd in pentlandite contents, indicative of a common original sulfide melt that experienced different degrees of post-accumulation percolation. However, in one of the samples (VZU3C), for which we used 3D micro CT scans to identify completely isolated sulfide droplets (definitely not interconnected with either the disseminated or the large sulfide blebs), we can observe a slight difference in the Pd in pentlandite and δ34S composition, with the fully isolated droplets being slightly more enriched in Pd and having slightly lighter δ34S values. This would be compatible with a slightly higher R factor: small droplets would be easier to transport within the silicate melt and have a relatively larger specific surface area to exchange with the silicate melt. There are two clear observations within this dataset. First, the heavy δ34S signature of the sulfides, reflecting the assimilation by the melt of large amounts of heavy crustal S, with values that are homogeneous within each sample, but vary widely between intrusions (varying between 4.7 ‰ and 17 ‰ δ34S), and even between samples from the same intrusion (12‰ to 17‰ δ34S in samples from the Kharaelakh intrusion). Secondly, the pentlandite is enriched in Pd, a result of high R factors coupled with the magmatic-peritectic origin of much of the pentlandite. However, one of the interesting and puzzling observations within this dataset is the lack of correlation between the increase in Pd contents in pentlandite and the decrease in δ34S compositions as would be expected if; 1) increase in Pd tenors is due to an increased R factor (silicate/sulfide liquid mass ratio), and 2) diffusion of Pd and of the various S isotopes between the silicate melt and the liquid sulfide droplets occurs at a similar rate. This might suggest that there are other factors at play, such as contrasting diffusivities for Pd and S, or S isotopic fractionation during interaction with a volatile phase.