A multi-method, multi-scale theoretical study of He and Ne diffusion in zircon

The quantification of He and Ne diffusion behavior in crystals rich in U and Th such as zircon is key for the interpretation of (U-Th)/He-4 and (U-Th)/Ne-21 thermochronometric ages. Multiple parameters such as chemical substitution, channel obstruction and damage can modify the diffusivity compared to a pristine structure. To investigate the impact of these parameters, we have conducted a theoretical diffusion study combining a series of methods and approaches to address the problem across the necessary range of scales (atomic to crystal size). First, using quantum calculation, we determine the different He and Ne insertion sites, insertion energies and diffusion pathways at the atomic scale for an ideal pristine zircon structure (i.e. damage free). These results serve as input for a 3D random walk simulation of atomic trajectories that provides diffusion coefficients for damage-free zircon crystals. Second, as natural zircon crystals are not perfect, we model the impact of different types of damage and diffusion pathway obstruction at the atomic level on He and Ne diffusion in 3D. The calculated He and Ne diffusion coefficients for pure ZrSiO4 exhibit strongly anisotropic behavior and very high diffusivity along the taxis, and with 3D, closure temperatures of -197 degrees C and -202 degrees C respectively. The results for He are comparable to previous DFT studies but strongly different from experimental diffusion results; results for Ne are similar in this respect. Modelling the impact of different types of damage (vacancies, recoil, fission, voids or fluid inclusions) and obstruction on He and Ne diffusion reveals important implications for the (U-Th)/He and (U-Th)/Ne thermochronometers. First, obstruction alone does not significantly modify He and Ne diffusion except to reduce anisotropy. Second, trapping is the primary mechanism altering He and Ne diffusion even at low dose, and we predict the maximal trapping energies for He and Ne to be 164 and 320 kJ/mol, similar to values inferred from experimental data. We also propose that the closure temperature increases non-linearly with damage, with effective trapping energy increasing with dose until a threshold, possibly corresponding to a percolation transition, after which retentivity decreases. Based on field data sets we also anticipate a value for this threshold of around similar to 2-5 x 10(17) alpha/g, lower than previously proposed. We show Ne to be highly blocked by damage and predict similar diffusion behavior to He, but with higher retentivity. We demonstrate the importance of investigating rare gas diffusion at the atomic level for comparison with experimental data, in order to build a predictive diffusion law at different scales. (C) 2019 Elsevier Ltd. All rights reserved.