Experimental Study of Noble Gas Partitioning and Diffusion in Common Crustal Minerals

Recent study [e.g. 1,2,3] has shown that the partitioning of noble gases amongst system phases (including minerals, fluids, and grain boundaries) is fundamental to thermochronologic models and interpretations of the commonly used K/Ar and U-Th/He decay systems. However, little direct data on noble gas partitioning between common minerals of the crust exists. We have investigated two experimental methods to measure noble gas partitioning between quartz, K-feldspar, plagioclase feldspar, and phlogopite. First, piston cylinder experiments employed an irradiated fluorite powder as a solid source of Ar-37 and He-4 within which multiple minerals were packed and allowed to exchange noble gases over experimental durations of 1-2 weeks and conditions of 925 C and 1.5 GPa. This Ar-37 source was used to avoid analytical artifacts of atmospheric Ar surface contamination. Results from the piston cylinder experiment show marked diffusive loss from the doped fluorite, but virtually no diffusive uptake of Ar-37 into any of the mineral phases in the capsule. The lack of measurable Ar-37 in the minerals suggests that Ar-37 is heavily partitioned into the grain boundary network, or that Ar-37 has leaked from the capsule. At the same time, piston cylinder results also indicate near uniform uptake of more rapidly diffusing He-4 into all minerals, which would suggest that the partition coefficient of He-4 between all of the silicate mineral phases present in the experiments is roughly equal to 1. Additional experiments are being conducted to reproduce this preliminary result. Second, minerals were exposed to pressurized argon gas in cold seal experiments over a period of 20 days at temperatures ranging from 825 to 485 C. Diffusive uptake profiles measured by UV-laser depth profiling and, in the case of quartz, also by Rutherford Backscattering, are observed for Ar in phlogopite, quartz, and feldspars. To our knowledge, this is the first time that a combined RBS/UV-laser depth profile has been reported for a single sample. Theoretically, equilibrium surface concentrations extrapolated from simple diffusion profiles in each mineral may be ratio-ed to extract mineral-mineral partition coefficients. Apparently anomalous high surface Ar concentrations were found in all minerals during UV-laser analysis, corroborating observations from previous studies [cf. 4]. RBS analysis of quartz confirmed the high concentrations but revealed quite different diffusive behavior in the near surface ~200nm region versus deeper (~100 microns) diffusive penetration (measured by UV-laser). This suggests the possibility of multiple, discrete diffusive pathways thus complicating partitioning interpretations. Additional cold seal experiments will involve the use of He-3 and Ne-22 wherein further assessment of near surface versus deeper diffusion regimes should be possible in all minerals. [1] Kelley SP (2002) Chem. Geol. 188, p. 1-22; [2] Baxter EF (2003) EPSL 216, p. 619-634; [3] Watson EB & Cherniak DJ (2003) GCA 67, p. 2043-2062; [4] Wartho JA et al. (1999) EPSL 170, p. 141-153.