Do concentrations of Mn, Eu and Ce in apatite reliably record oxygen fugacity in magmas?

Apatite is a common accessory mineral in igneous rocks, and its ability to readily accommodate a wide range of trace elements means that it may provide a useful record of magmatic composition and magmatic processes. Relative abundances of redox sensitive elements in apatite could also provide a much-needed probe of magmatic oxygen fugacity, fO(2). Here, examination of a large data set of apatite and whole rock compositions is used to assess the recently proposed, and increasingly used, Mn-in-apatite oxybarometer. Correlation of apatite-whole rock Mn partition coefficients with whole rock SiO2, and calculated melt polymerisation and aluminosity, support a model where apatite Mn content is largely dependent onmelt structure. In more evolved systems, a decrease in availability of non-bridging oxygens in silicate melts drives Mn from being incompatible, to an increasingly compatible element in apatite. Magmatic fO(2) calculated from Ce concentrations in zircon also shows no discernible correlation with apatite Mn content. Mn content of apatite clearly does not record magmatic fO(2), although it may be useful in indicating the extent of melt evolution. In contrast, concentrations of other trace elements in apatite have a much weaker, or no discernible dependence on melt structure and composition, supporting the assertion that apatite records key aspects of magmatic composition. Eu and Ce anomalies in chondrite-normalised Rare Earth Element (REE) data provide an alternative means of using apatite to probe magmatic fO(2). A new method for more reliably comparing the extent of Eu and Ce anomalies in apatite REE data is used to assess controls on Eu and Ce content. Apatite Eu content shows a strong dependence on whole rock composition, indicating that Eu anomalies often reflect feldspar crystallisation. Ce anomalies in apatite are much smaller and vary from weakly positive to weakly negative. Predicted strong, negative Ce anomalies in apatite data are not observed, implying smaller differences in Ce3+ and Ce4+ compatibility in apatite than in minerals such as zircon. The cause of small positive Ce anomalies in apatite data is unclear. Comparison of apatite and zircon data suggests that fO(2) may have some control on apatite Ce content. However, the narrow range of Ce anomalies in apatite data, and lack of an identifiable control of fO(2), limit the extent to which apatite Ce recordsmagmatic fO(2). As such, REE contents of apatite also cannot currently be used to reliable indicate trends in magmatic fO(2). (C) 2020 Published by Elsevier B.V.