|Title||Titanium isotopic fractionation in Kilauea Iki lava lake driven by oxide crystallization|
|Publication Type||Journal Article|
|Year of Publication||2019|
|Authors||Johnson A.C, Aarons S.M, Dauphas N., Nie N.X, Zeng H., Helz R.T, Romaniello S.J, Anbar A.D|
|Type of Article||Article|
|Keywords||Basalt; continental-crust; coordination chemistry; differentiation; diffusion; Geochemistry & Geophysics; identification; Magma differentiation; magmatic differentiation; magnesium; mantle; melts; silicate-glasses; stable isotopes|
Recent work has demonstrated that titanium (Ti) isotopes undergo mass-dependent isotope fractionation during magmatic differentiation, leaving evolved silicic melts preferentially enriched in heavy Ti isotopes. Preferential incorporation of light Ti isotopes in crystallizing Fe-Ti oxides is thought to be the mechanism responsible for this fractionation in magmatic rocks. To test this hypothesis, we present Ti isotope measurements of Fe-Ti oxide mineral separates of Kilauea Iki lava lake samples. We find that the Ti in Fe-Ti oxides is isotopically light while Ti in the residual melt and minerals is isotopically heavy. This result is consistent with the results of density functional theory (DFT) calculations in other studies, which show progressive heavy isotope enrichment for Ti from 6-fold, 5-fold, through 4-fold coordinated minerals. We therefore conclude that Ti isotopes in silicate melts undergo isotope fractionation during the crystallization of Fe-Ti oxides because Ti in oxides is primarily in 6-fold coordination whereas Ti in silicate melts is in 5- or 4-fold coordination (Ti in more evolved magmas tends to be in lower coordination). Based on our mineral separate results, we estimate the fractionation factor at 1000 degrees C between silicate and oxide Delta Ti-49(silicate-oxide) to be 0.39 +/- 0.06%. This result is consistent with the fractionation factors inferred in previous studies based on Ti isotopic analyses and modeling of bulk rock measurements. We use this fractionation factor and the fractionation factors proposed by previous workers in Rhyolite MELTS to model the delta Ti-49 evolution of plume lavas. We find the model to generally predict the fractionations observed in Kilauea Iki, as well as the fractionations previously observed in volcanics from Hekla, Iceland and Afar, East Africa. (C) 2019 Elsevier Ltd. All rights reserved.