Diffusive titanium isotope fractionation in silicate melts

Abstract

We report the first study of titanium (Ti) isotope fractionation during diffusion in Fe-free basaltic melts using diffusion couple experiments. Ti is a high-field strength element with low diffusivity similar to Si. Understanding Ti isotope diffusion could provide insight into the behavior of other low-diffusivity elements in magmatic processes. For this study, we selected two experimental charges with the largest initial contrast in TiO₂ concentrations from previous diffusion couple experiments. We measured the ⁴⁹Ti/⁴⁷Ti isotope ratio profiles in these experiments using Secondary Ion Mass Spectrometry (SIMS). Our results show that SIMS measurements using Cameca IMS-1280 can achieve a precision of 0.05 ‰ to 0.1 ‰ (1SD internal error) for δ⁴⁹Ti at ∼3 wt% TiO₂ using multi-collector Faraday cup and electron multiplier (EM). However, the precision drops to 0.5‰ at 0.03 wt% TiO₂ using the EM. For an initial Ti concentration contrast of about 180 in a diffusion couple, the total variation in δ⁴⁹Ti across the diffusion couple profile is about 3.0 ‰. Diffusivities of different Ti isotopes are related by D₄₉/D₄₇ = (m₄₇/m₄₇)^β, where D₄₇ and D₄₉ are the diffusivities of ⁴⁷Ti and ⁴⁹Ti, m₄₇ and m₄₉ are their atomic masses, and β is an empirical parameter characterizing diffusive isotope fractionation. By fitting the isotope ratio profiles, we determined a β value of 0.0318 ± 0.0021 (1SD error) for the TiO₂-₋MgO interdiffusion couple at 1500 °C, and values of 0.019 to 0.031 for the SiO₂-TiO₂ interdiffusion couple. Hence, total Ti isotope ratio variation greater than 1‰ could be generated along a diffusion profile when the initial TiO₂ concentration contrast exceeds 15. Such high concentration contrast may be realized for basalt-rhyolite or basalt-komatiite magma mixing. This study expands the database of diffusion parameters for non-traditional isotopes and offers new insight into Ti isotope fractionation during magmatic process, with potential to further understand magma and rock evolution. For example, we schematically modeled Ti concentration and isotopes in Horoman peridotite massif, and were able to explain most observed data.