Fingerprinting crustal anatexis with apatite trace element, halogen, and Sr isotope data

Abstract

The geochemistry of apatite is a valuable tool for granite petrogenetic study. Here we explore the capability of apatite for fingerprinting anatectic reactions during crustal melting. We conducted a detailed petrographic, trace element, volatile, and Sr isotopic study of the apatites from four groups of leucosomes and leucogranites that formed by amphibole dehydration melting (group I), hydration melting (group II), muscovite dehydration melting (group III), and feldspar accumulation (Group IV) from the central Himalayan orogen. The coexistence of primary CO2-rich fluid and multiphase solid inclusions in apatite indicates fluid–melt immiscibility and suggests that C–O–H fluids were present during hydration melting. We observe marked differences in major and trace element, as well as halogen (F and Cl) contents in apatite from different groups, corresponding to different melting reactions. Apatite REE, Y, Mn, F, and F/Cl exhibit a decrease, whereas Sr/Y and Cl increase, from muscovite dehydration melting to hydration melting, and to amphibole dehydration melting. Amphibole dehydration melting results in similar CaO, Sr, and Eu/Eu* in apatite to hydration melting, but the latter leads to lower (La/Yb)N in apatite. These geochemical differences in apatite can be explained by different minerals participating in the melting reactions and varying melt polymerization resulting from the different melting reactions. The degree of melt polymerization decreases in the sequence from muscovite dehydration melting to hydration melting, and to amphibole dehydration melting. A higher degree of melt polymerization results in higher values for trace element partition coefficients between apatite and melt, and therefore the enhanced incorporation of REE, Y and Mn in apatite. Higher Sr and Eu/Eu* in apatite reflect the consumption of a larger amount of plagioclase in the melting reaction. The F and Cl contents in partial melts were calculated based on apatite composition, which exhibit an increasing trend from hydration melting to muscovite hydration melting, and to amphibole hydration melting. Such a trend can be explained by different proportions of hydrous minerals (muscovite and amphibole) participating in the melting reactions, with amphibole responsible for the high Cl content in melt and apatite. In addition, the apatite Sr isotopic compositions record the Sr isotopic disequilibrium during partial melting and exhibit distinct evolutional trends resulting from different anatectic reactions. Muscovite dehydration melting forms melt with higher 87Sr/86Sr than the restite due to radiogenic Sr released by muscovite. Apatite therefore records progressively decreasing 87Sr/86Sr during melt evolution before the isotopic homogenization between the partial melt and the restite. In contrast, apatites crystallized from the melt of hydration melting show an evolutional trend of increasing 87Sr/86Sr. In summary, the elemental and isotopic compositions of apatite can be related to different types of melting reactions, and can therefore be used as an indicator of the mechanisms of melt formation during crustal anatexis. We demonstrate such an application through the case study of group IV apatite from feldspar accumulation, and also explore its utility in granite petrogenetic studies.