nai

Project Progress

Since the pioneering discovery reported by Farquhar et al. in 2000 that the rarer isotopes of sulfur (³³S and ³⁶S) show mass independent fractionation (S-MIF; Δ³³S or Δ³⁶S ≠ 0) in rocks older than ~2.4 Ga, a large body of work has emerged that concerns the degree to which these isotopes record information about atmospheric redox state and microbial metabolism. Since the earliest studies, the range of S-MIF observed in the rock record has expanded significantly, and consistent trends have emerged. One such observation is that Archean sulfur four isotope data tend to plot on a “reference array” with a Δ³⁶S/Δ³³S slope of ~ –0.9. Recently, a correlation has been observed between departures from this reference array and extremely low δ¹³Corg values in Neoarchean drill core samples, interpreted to represent the intermittent influence of an organic haze on atmospheric production of S-MIF. Laboratory culture experiments supported by models have revealed that mass dependent sulfur isotope fractionation effects in metabolic networks combine to generate small deviations in Δ³³S and Δ³⁶S that can be used to distinguish the variable expression of different sulfur metabolisms in the rock record.

This year, techniques were developed at WiscSIMS for the simultaneous measurement of all four stable isotopes of sulfur in order to evaluate whether in situ measurements at the micrometer scale could be used effectively to address some of these emerging paleoenvironmental and biogeochemical questions. In an initial proof of concept study, 33 sulfur four isotope analyses of detrital pyrite in a glacial diamictite from the ~2.4 Ga Meteorite Bore Member of Western Australia, and 58 such analyses of pyrite in the ~2.5 Ga Mount McRae Shale, the ~2.6 Ga Carrawine Dolomite, and the ~2.6 Ga Jeerinah Shale were conducted. Using a ~25 × 20 micrometer spot, average external precision for these analyses (2 SD; based on 116 analyses of the UWPy-1 Balmat pyrite standard) was 0.25‰, 0.05‰, and 0.85‰ for δ³⁴S, Δ³³S, and Δ³⁶S, respectively. A regression including all 95 sample analyses yields a line with Δ³⁶S/Δ³³S slope –0.87 and r2 = 0.93. Slight deviations within this overall trend are resolved, and these may be related to the atmospheric and/or biological effects discussed above.