No measurable calcium isotopic fractionation during crystallization of Kilauea Iki lava lake, Hawaii, and its implication to calcium isotopic fractionation during mantle partial melting

Large Ca isotopic variation has been observed in terrestrial igneous rocks, with δ^44/40Ca_{SRM 915a} ranging from -0.08 to 1.68. Several processes have been proposed to fractionate Ca isotopes in igneous rocks. In order to investigate possible Ca isotopic fractionation during basaltic magma differentiation, we measured Ca isotopic compositions of lavas recovered from Kilauea Iki lava lake at Hawaii. This set of lavas record the whole crystal fractionation history of tholeiitic magma, ranging from olivine accumulation/fractionation to multiple phase crystallization including plagioclase and clinopyroxene. Our results show no detectable Ca isotopic variation in all measured Kilauea lavas, with an average δ^44/40Ca of 0.80±0.08 (2SD, n = 19). Using this observation and inter-mineral Ca isotopic fractionation factors, a Monte Carlo approach is used to estimate the mineral-melt ⁴⁴Ca/⁴⁰Ca fractionation factors. We found that Ca isotopic fractionation between clinopyroxene and melt is small, with Δ^44/40Ca^cpx-melt = 0.04±0.03 at 1200 °C. We used this estimated Δ^44/40Ca^cpx-melt and inter-mineral Ca isotopic fractionation factors to investigate Ca isotopic effects during mantle partial melting under 1-2 GPa. Our simulations show that the largest ⁴⁴Ca/⁴⁰Ca effect, up to +0.3‰, is achieved in large degree melting residues during fractional and dynamic melting. In contrast, partial melts show negligible ⁴⁴Ca/⁴⁰Ca isotopic effect, <0.07‰. Consequently, the observed large Ca isotopic variation in basalts most likely reflects source heterogeneity.