In situ sulfur isotope analysis of sulfide minerals by SIMS: Precision and accuracy, with application to thermometry of ∼ 3.5 Ga Pilbara cherts

Abstract

Secondary ion mass spectrometry (SIMS) measurement of sulfur isotope ratios is a potentially powerful technique for in situ studies in many areas of Earth and planetary science. Tests were performed to evaluate the accuracy and precision of sulfur isotope analysis by SIMS in a set of seven well-characterized, isotopically homogeneous natural sulfide standards. The spot-to-spot and grain-to-grain precision for δ34S is ±0.3‰ for chalcopyrite and pyrrhotite, and ±0.2‰ for pyrite (2SD) using a 1.6nA primary beam that was focused to 10µm diameter with a Gaussian-beam density distribution. Likewise, multiple δ34S measurements within single grains of sphalerite are within ±0.3‰. However, between individual sphalerite grains, δ34S varies by up to 3.4‰ and the grain-to-grain precision is poor (±1.7‰, n=20). Measured values of δ34S correspond with analysis pit microstructures, ranging from smooth surfaces for grains with high δ34S values, to pronounced ripples and terraces in analysis pits from grains featuring low δ34S values. Electron backscatter diffraction (EBSD) shows that individual sphalerite grains are single crystals, whereas crystal orientation varies from grain-to-grain. The 3.4‰ variation in measured δ34S between individual grains of sphalerite is attributed to changes in instrumental bias caused by different crystal orientations with respect to the incident primary Cs+ beam. High δ34S values in sphalerite correlate to when the Cs+ beam is parallel to the set of directions <uuw>, from [111] to [110], which are preferred directions for channeling and focusing in diamond-centered cubic crystals. Crystal orientation effects on instrumental bias were further detected in galena. However, as a result of the perfect cleavage along {100} crushed chips of galena are typically cube-shaped and likely to be preferentially oriented, thus crystal orientation effects on instrumental bias may be obscured. Test were made to improve the analytical precision of δ34S in sphalerite, and the best results were achieved by either reducing the depth of the analysis pits using a Köhler illuminated primary beam, or by lowering the total impact energy from 20keV to 13keV. The resulting grain-to-grain precision in δ34S improves from ±1.7‰ to better than 0.6‰ (2SD) in both procedures. With careful use of appropriate analytical conditions, the accuracy of SIMS analysis for δ34S approaches ±0.3‰ (2SD) for chalcopyrite, pyrite and pyrrhotite and ±0.6‰ for sphalerite. Measurements of δ34S in sub-20µm grains of pyrite and sphalerite in ∼3.5Ga cherts from the Pilbara craton, Western Australia show that this analytical technique is suitable for in situ sulfur isotope thermometry with ±50°C accuracy in appropriate samples, however, sulfides are not isotopically equilibrated in analyzed samples.