Comparison of chemical procedures for Re-isotopic measurements by N-TIMS

Precise and accurate measurements of the isotopic composition of rhenium (Re) in geologic samples are of paramount importance for the successful application of the Re-Os geochronometer. Two methods for chemical isolation of Re are used in the geochronology community - column chemistry using anion exchange resin and solvent-solvent extraction using a sodium hydroxide-acetone mixture. Despite their ubiquitous application, a systematic comparison of these two methods is missing. Here we compare Re-isotopic data on a variety of samples (sulfides, hydrocarbons, shales, and in-house standard solutions) that were processed by both chemical methods. Measured 185Re/187Re ratios for both methods have similar precision and overlap within the propagated long-term analytical reproducibility of ~1‰ for samples that yield moderate to strong intensities (>50 mV on the weaker Re mass). In contrast, measured 185Re/187Re ratios for most samples that yield less intense Re signals (<50 mV on the weaker Re mass) disagree by ~3-6‰ between the two methods, exceeding the propagated analytical uncertainty of ~3‰ for low-intensity measurements. A systematic investigation of samples with natural Re isotopic composition (shale, molybdenite, and standard solutions) shows that use of anion exchange resin (column chemistry) may cause elevated 185Re/187Re ratios. This effect is more pronounced for low-Re shales (<1 ng total Re processed) and becomes nearly or completely insignificant for high-Re samples measured over a wide range of signal intensities. Standard solutions processed through column chemistry, however, preserve their natural Re isotopic composition. Together, these data suggest that complex sample matrices leading to sample-derived impurities passed through column chemistry may cause isobaric interferences that deviate measured 185Re/187Re ratios in geologic samples. In contrast, the acetone-NaOH method produces more consistent Re isotopic results across the same range of sample types, Re loads and signal intensities. Thus, we recommend use of the acetone-NaOH solvent-solvent extraction procedure for low-Re samples, and provide a practical way to lower the typically higher Re blank supplied by NaOH.