Diffusion of helium in zircon and apatite

Diffusion of helium has been characterized in natural zircon and apatite. Polished slabs of zircon and apatite, oriented either normal or parallel to c were implanted with 100 keV He-3 at a dose of 5 x 10(15 3) He/cm(2). Diffusion experiments on implanted zircon and apatite were run in Pt capsules in 1-arm furnaces. He-3 distributions following experiments were measured with Nuclear Reaction Analysis using the reaction He-3(d,p)He-4. For diffusion in zircon we obtain the following Arrhenius relations: D-Ic = 2.3 x 10(-7) exp(-146 +/- 11 kJmol(-1)/RT)m(2)s(-1). D-Ic = 1.7 x 10(-5) exp(-148 +/- 17kJmol(-1)/RT)m(2)s(-1). Although activation energies for diffusion normal and parallel to c are comparable, there is marked diffusional anisotropy, with diffusion parallel to c nearly 2 orders of magnitude faster than transport normal to c. These diffusivities bracket the range of values determined for He diffusion in zircon in bulk-release experiments, although the role of anisotropy could not be directly evaluated in those measurements. In apatite, the following Arrhenius relation was obtained over the temperature range of 148-449 degrees C for diffusion normal to c: D = 2.10 x 10(-6) exp(-117 +/- 6kJmol(-1)/RT)m(2)s(-1). In contrast to zircon, apatite shows little evidence of anisotropy. He diffusivities obtained in this study fall about an order of magnitude lower than diffusivities measured through bulk release of He through step-heating, and within an order of magnitude of determinations where ion implantation was used to introduce helium and He distributions measured with elastic recoil detection. Since the diffusion of He in zircon exhibits such pronounced anisotropy, helium diffusional loss and closure cannot be modeled with simple spherical geometries and the assumption of isotropic diffusion. A finite-element code (CYLMOD) has recently been created to Simulate diffusion in cylindrical geometry with differing radial and axial diffusion coefficients. We present some applications of the code in evaluating helium lost from zircon grains as a function of grain size and length to diameter ratios, and consider the effects of shape anisotropy, where diffusion is isotropic (as in the case of apatite) but shapes of crystal grains or fragments may depart significantly from spherical geometry. (C) 2009 Elsevier B.V. All rights reserved.