Evaluation of cosmogenic 3He and 21Ne production rates in olivine and pyroxene from two Pleistocene basalt flows,western Grand Canyon,AZ, USA

Combining cosmogenic ³He and ²¹Ne (³He_c and ²¹Ne_c) measurements on both pyroxene and olivine from the Pleistocene Bar Ten flows (85–107 ka) greatly increases our ability to evaluate the accuracy of ³He_c and ²¹Ne_c production rates and, therefore, ³He_c and ²¹Ne_c surface exposure ages. Comparison of ³He_c and ²¹Ne_c age-pairs yielded by experimentally determined production rates and composition-based model calculations indicates that the former give more accurate surface exposure ages than the latter in this study. However, experimental production rates should be adjusted to the composition of the minerals being analyzed to obtain the best agreement between ³He_c and ²¹Ne_c ages for any given sample. ²¹Ne_c/³He_c values are 0.400 ± 0.029 and 0.204 ± 0.014 for olivine and pyroxene, respectively, in Bar Ten lava flows, in agreement with previously published values, and indicate that ²¹Ne_c/³He_c in olivine and pyroxene is not affected by erosion and remains constant with latitude, elevation, and time (up to 10 Myr). Samples with ²¹Ne_c/³He_c that do not agree with these values may indicate the presence of non-cosmogenic helium and/or neon. The neon three-isotope diagram can also indicate whether or not all excess neon in mineral separates comes from cosmogenic sources. An error-weighted regression for olivine defines a spallation line y = (1.033 ± 0.031)x + (0.09876 ± 0.00033)], which is indistinguishable from that for pyroxene (Schäfer et al., 1999). We have derived a production rate of 25 ± 8 at/g/yr for ²¹Ne_c in clinopyroxene (En_43–44) based on the ⁴⁰Ar/³⁹Ar age of the upper Bar Ten flow. Our study indicates that the production rate of ²¹Ne_c in olivine may be slightly higher than previously determined. Cosmogenic ³He and ²¹Ne remain extremely useful, particularly when paired, in determining accurate eruption ages of young olivine- and pyroxene-rich basaltic lava flows.