A NEW APPROACH TO UNDERSTANDING ORE FORMATION PROCESSES: MULTIPLE SULFUR ISOTOPE ANALYSIS OF SULFIDE MINERALS OF THE BUSHVELD COMPLEX

When a mafic magma intrudes into wall rocks that record mass-independent fractionation of sulfur isotopes (older than ≈2.0 Ga), then multiple sulfur isotope analysis can be used to trace magma-wall rock interactions.  This technique can trace whether sulfur in the ore deposit is derived from surrounding wall rock.  The technique can also determine whether individual sulfide phases within a single hand sample have the same source of sulfur and whether they are equilibrium phases.  We demonstrate the applicability of this technique to the Platreef ore horizon of the Bushveld Complex, South Africa.

Solid-earth fractionation processes of ³³S/³²S, ³⁴S/³²S, and ³⁶S/³²S are mass-dependent, characterized by: D³³S ≈ 0.515 x d³⁴S and D³⁶S ≈ 1.92 x d³⁴S.  Deviation from mass-dependent isotopic fractionation is quantified as D³³S (≈ d³³S - 0.515 x d³⁴S) and D³⁶S (≈ d³⁶S - 1.92 x d³⁴S). Proxies for ‘primary’ igneous sulfur (meteorites, peridotite xenoliths) possess D³³S and D³⁶S values near 0 ‰.  Similarly, relatively pristine Bushveld igneous rocks have D³³S and D³⁶S values near 0 ‰ (D³³S values ≤ 0.2‰ and D³⁶S between -1 ‰ and 1 ‰).  On the other hand, sedimentary wall rocks surrounding the Platreef exhibit substantial non-zero D³³S and D³⁶S values (up to ≈ 5.0 ‰ and down to ≈ -3.6 ‰ respectively). As D³³S and D³⁶S are chemically-conservative tracers, non-zero D³³S and D³⁶S values in Platreef sulfide ore minerals would indicate material transfer of S from the surrounding wall rock.  Values of D³³S within the Platreef range up to 0.55 ‰ showing evidence for some interaction between the magma and the wall rock.

Sulfide minerals were separated to determine whether different phases had the same S source and whether they were in equilibrium.  Most samples contained chalcopyrite, pyrrhotite and pentlandite.  The pyrrhotite and pentlandite were inseparable.  For the analyses, chalcopyrite was separated from a pentlandite/pyrrhotite mixture.  Each pair of coexisting sulfide separates is consistent with single D³³S and D³⁶S values, suggesting that the phases within each sample share the same source of S.  In some samples d³⁴S_ccp > d³⁴S_pent/pyrr and in others the relationship is the reverse.  This suggests that at least some of the samples are not consistent with equilibrium fractionation of ³⁴S/³²S.