Isotopic constraints on the origin and nature of sulfur components within the ancient sub-continental lithospheric mantle

Estimating the sulfur isotope composition of Earth’s fertile mantle can provide insight into Earth’s constitution and the processes associated with establishing its highly siderophile element abundances. There is currently a disagreement in sulfur isotopic compositions between Ocean Island Basalts (OIB) (δ 34 S = 0) and Mid-Ocean Ridge Basalts (MORB) (δ 34 S ≈ −1) implying that their mantle sources are different. MORBs and OIBs are young, so it remains unknown whether sulfur isotopic composition differences stem from relatively ancient, perhaps primordial compositions inherited from Earth’s accretion and differentiation or, if they reflect the effects of constant recycling of crust to the convective upper mantle. Subcontinental lithospheric mantle (SCLM) keels represent Earth’s only ancient mantle reservoirs that have been preserved from mantle convection affecting MORBs and OIBs, and as such may retain important clues about the origin and evolution of its sulfur isotopic compositions. This study presents major and trace element compositions, and the isotope compositions of Sr, Nd, Os, and S from Kilbourne Hole, New Mexico peridotite xenoliths which confirm that KH xenoliths are associated with an ancient mantle reservoir (∼1.7 Ga). The average sulfur isotope composition of olivine and clinopyroxene hosted sulfides measured in-situ is δ 34 S = ∼−1‰, which contrasts the bulk S isotope measurements that extend to values of ∼+2‰. The difference in S isotope composition of in-situ vs. bulk measurements indicates that the phenocryst hosted sulfides likely record the composition of the ancient mantle, while the bulk S isotope composition records mixing with sulfur associated with subsequent metasomatism. Furthermore, the similarity in S isotope composition of individual KH sulfides and estimates for the mantle from MORB suggest that the process responsible for fractionating 34 S/ 32 S in Earth’s mantle was likely widespread and occurred prior to ∼1.7 Ga. Positive correlations among bulk δ 34 S and 87 Sr/ 86 Sr data, as well as bulk sulfide concentration and (Ce/Yb) N, illustrates that the bulk S reflects compositional influences from recycled continental crust materials. Similarly, variable Δ 33 S among the bulk KH measurements (−0.007 ± 0.008 to +0.038 ± 0.016) shows the compositional influence from mixing with recycled surface materials. The relationship among Δ 33 S and Δ 36 S values furthermore indicates that the metasomatic sulfur is of post-Archean age. Primary S isotope compositions can be preserved in the SCLM and can be discerned through a strategic bulk vs. in-situ S isotope measurement approach.