Macroscale NTIMS and microscale LA-MC-ICP-MS Re-Os isotopic analysis of molybdenite: Testing spatial restrictions for reliable Re-Os age determinations, and implications for the decoupling of Re and Os within molybdenite

We present a detailed study of Re-Os age determinations for eight natural molybdenite samples of like polytype (2H), spanning a range of age, natural grain size and deposit type. The focus of the study is to critically evaluate the effects of sampling, sample preparation and aliquant size on the accuracy and reproducibility of Re-Os ages for these molybdenite samples. We find that for some molybdenite samples, analysis of small sample aliquants (<20 mg) may not yield accurate or reproducible Re-Os ages, whereas analysis of larger aliquants from the same mineral separate do yield reproducible Re-Os dates. Such an observation is best explained if Re and Os-187 are internally decoupled within molybdenite grains. This finding is supported from spot analyses by laser ablation MC-ICP-MS analyses presented here and is consistent with previously published observations. The degree of decoupling between Re and Os-187 appears to increase both as a function of increasing grain size, and increasing age of molybdenite. From detailed dating of individual molybdenite mineral separates, we provide approximate minimum aliquant amounts required for reproducible Re-Os age dating, as a function of molybdenite age and grain size. Geologically younger, naturally fine-grained molybdenite samples appear to show little Re and Os-187 decoupling, and reproducible ages can be determined from some samples with as little as 1 mg of aliquant. Geologically old, and coarse-grained molybdenite samples may require as much as 40 mg of aliquant from a much larger mineral separate to overcome Re and Os-187 decoupling. The mechanism(s) of Re and Os-187 decoupling within molybdenite is not constrained by this results of this study, but the observation that the degree of decoupling increases with grain size (distance) and age (time/geologic history) may suggest primary diffusive control. Assuming that Re and Os-187 decoupling in molybdenite results primarily from diffusion of Os-187, apparent diffusion coefficients are calculated (D = x(2)/t). Estimates of D for Os made in this way range from 2.8 x 10(-26) to 2.1 x 10(-21) m(2)/s, which are broadly similar to experimentally derived diffusion coefficients for Os in Fe-sulfide minerals and for Re in molybdenite at temperatures <500degreesC. Copyright (C) 2004 Elsevier Ltd.