GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 189-13
Presentation Time: 9:00 AM-6:30 PM

DETERMINATION OF FERRIC-FERROUS RATIO IN PYROXENES WITH X-RAY ABSORPTION SPECTROSCOPY


STEVEN, Cody J., Geological Sciences, University of Idaho, Moscow, ID 83844, DYAR, M. Darby, Dept. of Astronomy, Mount Holyoke College, South Hadley, MA 01075 and MCCANTA, Molly C., Earth and Planetary Sciences, University of Tennessee, 1621 Cumberland Ave, Knoxville, TN 37996

Iron is the most abundant multivalent element in the Earth’s crust, and is commonly used as an indicator of oxygen fugacity in geologic systems. Oxygen fugacity is a principle component of phase equilibria of a chemical system, and the redox state of sample can be used to ascribe provenance and interpret the petrogenesis of a rock. Despite its importance, few methods are sensitive to Fe valence state, and the predominant method, Mossbauer spectroscopy, requires a bulk sample for analysis. Because of the limitations of Mossbauer spectroscopy, calibrations using X-ray absorption spectroscopy (XAS) continue to develop, because its 1×1 mm spot size permits analyses on the same scales as other probes. Multivariate models that utilize full spectral ranges of XAS have proven to accurately predict redox ratios in spite of the polarization of the synchrotron beam, even in anisotropic minerals. Though these models have been established for garnets, amphiboles, and glasses, several other geologically significant mineral groups remain to be calibrated. In particular, pyroxenes are widely used in thermobarometry and occur in a wide variety of terrestrial and extraterrestrial environments. Further, pyroxenes can be chemically zoned or exsolved, so their redox ratios provide insights into the behavior of iron partitioning under the crystallization or recrystallization conditions of the pyroxene. Fe-XAS data collected from a set of 70 pyroxenes in this study and predicted using multivariate analysis show a strong correlation with true ferric-ferrous ratio from Mossbauer spectroscopy. The resultant prediction model is far more accurate than that determined from the pre-edge region alone.