TY - JOUR
T1 - Lattice diffusion and solubility of argon in forsterite, enstatite, quartz and corundum
AU - Thomas, J. B.
AU - Cherniak, D. J.
AU - Watson, E. B.
N1 - Funding Information:
John Brady provided us with slabs of Linde forsterite used in many experiments. We thank Dave Jenkins at the Binghamton University for assistance and use of an internally-heated gas vessel. Dave Wark provided expertise during electron microprobe analyses at RPI. We are grateful to Zhenting Jiang for providing his expertise and access to the EBSD facility at Yale University. Discussions with Frank Richter regarding the significance of these results were particularly inspiring. Bernie Wood and Simon Kelley reviewed an earlier version of this manuscript and their remarks strengthened the current paper. Comments and suggestions by Steve Parman and an anonymous reviewer significantly improved this paper. This research was funded by the National Science Foundation under grant no. EAR-0337481 to EBW.
PY - 2008/7/30
Y1 - 2008/7/30
N2 - Argon diffusivities and solubilities in single, gem-quality crystals of forsterite, enstatite, quartz and corundum were determined from experiments conducted between 0.5 and ~ 6000 bars Ar pressure and temperatures from 425 °C to 1200 °C. Polished single-crystal slabs and specimens with natural facets were placed in open containers and exposed to an argon atmosphere either in a pressure vessel or in a gas-flow tube furnace at near-atmospheric pressure. Argon atoms from the pressure medium diffused into the crystals to produce near-surface concentration gradients, which were directly profiled using Rutherford backscattering spectrometry. The following Arrhenius relations were obtained for Ar diffusion:F o r s t e r i t e : DA r = 7.2 × 10- 20 exp (- 42 k J m o l- 1 / R T) m2 s- 1. E n s t a t i t e : DA r = 1.3 × 10- 20 exp (- 32 k J m o l- 1 / R T) m2 s- 1. Q u a r t z : DA r = 3.1 × 10- 19 exp (- 43 k J m o l- 1 / R T) m2 s- 1. C o r u n d u m : DA r = 1.2 × 10- 20 exp (- 33 k J m o l- 1 / R T) m2 s- 1. The reported activation energies probably represent apparent activation energies that are a combination of the effects of lattice diffusion of Ar atoms with trapping in point-defect vacancies. There are no discernible differences in Ar diffusion in different crystallographic directions, and diffusivities do not vary as a function of the intrinsic oxygen fugacity of the experimental vessels. In addition to diffusivities, the uptake gradients also yielded lattice solubilities of argon, as represented by the concentrations at the mineral surfaces. There are small differences between the solubilities of forsterite, quartz and corundum, but all solubilities are high at ~ 2000 ppm (by mass). The Ar solubilities for the iron-bearing minerals enstatite and San Carlos olivine are much higher than for the other minerals. The mean value for enstatite is 30-60% higher than the iron-free minerals, owing, perhaps, to incorporation of Ar in point defects that are more abundant in enstatite due to oxidation of minor FeO. Argon solubilities are independent of Ar pressure at PAr > ~ 1 bar, but slightly lower at PAr ≈ 0.5-1 bar. This behavior suggests that all sites capable of accommodating Ar atoms are filled at Ar pressures of ~ 1 bar. In all cases, Ar appears to be compatible in the minerals investigated, although the degree of compatibility will depend upon assumptions concerning the minimum Ar fugacity required to fully populate the available vacancies. A series of validation experiments and surface-sensitive analyses were conducted to confirm that the near-surface regions of crystals used in the experiments were crystalline, unreacted and free of structural damage that could potentially influence diffusion and solubility results.
AB - Argon diffusivities and solubilities in single, gem-quality crystals of forsterite, enstatite, quartz and corundum were determined from experiments conducted between 0.5 and ~ 6000 bars Ar pressure and temperatures from 425 °C to 1200 °C. Polished single-crystal slabs and specimens with natural facets were placed in open containers and exposed to an argon atmosphere either in a pressure vessel or in a gas-flow tube furnace at near-atmospheric pressure. Argon atoms from the pressure medium diffused into the crystals to produce near-surface concentration gradients, which were directly profiled using Rutherford backscattering spectrometry. The following Arrhenius relations were obtained for Ar diffusion:F o r s t e r i t e : DA r = 7.2 × 10- 20 exp (- 42 k J m o l- 1 / R T) m2 s- 1. E n s t a t i t e : DA r = 1.3 × 10- 20 exp (- 32 k J m o l- 1 / R T) m2 s- 1. Q u a r t z : DA r = 3.1 × 10- 19 exp (- 43 k J m o l- 1 / R T) m2 s- 1. C o r u n d u m : DA r = 1.2 × 10- 20 exp (- 33 k J m o l- 1 / R T) m2 s- 1. The reported activation energies probably represent apparent activation energies that are a combination of the effects of lattice diffusion of Ar atoms with trapping in point-defect vacancies. There are no discernible differences in Ar diffusion in different crystallographic directions, and diffusivities do not vary as a function of the intrinsic oxygen fugacity of the experimental vessels. In addition to diffusivities, the uptake gradients also yielded lattice solubilities of argon, as represented by the concentrations at the mineral surfaces. There are small differences between the solubilities of forsterite, quartz and corundum, but all solubilities are high at ~ 2000 ppm (by mass). The Ar solubilities for the iron-bearing minerals enstatite and San Carlos olivine are much higher than for the other minerals. The mean value for enstatite is 30-60% higher than the iron-free minerals, owing, perhaps, to incorporation of Ar in point defects that are more abundant in enstatite due to oxidation of minor FeO. Argon solubilities are independent of Ar pressure at PAr > ~ 1 bar, but slightly lower at PAr ≈ 0.5-1 bar. This behavior suggests that all sites capable of accommodating Ar atoms are filled at Ar pressures of ~ 1 bar. In all cases, Ar appears to be compatible in the minerals investigated, although the degree of compatibility will depend upon assumptions concerning the minimum Ar fugacity required to fully populate the available vacancies. A series of validation experiments and surface-sensitive analyses were conducted to confirm that the near-surface regions of crystals used in the experiments were crystalline, unreacted and free of structural damage that could potentially influence diffusion and solubility results.
KW - Argon
KW - Diffusion
KW - Partition coefficient
KW - Rutherford backscattering spectrometry
KW - Solubility
UR - http://www.scopus.com/inward/record.url?scp=45449099051&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=45449099051&partnerID=8YFLogxK
U2 - 10.1016/j.chemgeo.2008.03.007
DO - 10.1016/j.chemgeo.2008.03.007
M3 - Article
AN - SCOPUS:45449099051
SN - 0009-2541
VL - 253
SP - 1
EP - 22
JO - Chemical Geology
JF - Chemical Geology
IS - 1-2
ER -