Apatite as a Tool for Tracking Magmatic CO2 Contents
Abstract:
We present the results of high-temperature, high-pressure experiments investigating the partitioning behaviour of CO2 between apatite and basaltic melt. Experiments were run in piston cylinder apparatus at 1 GPa and 1250 °C, with a slow initial cooling ramp employed to facilitate crystal growth. Each charge contained the starting basaltic powder doped with Ca-phosphate and variable proportions of H2O, CO2, and F. Run products are glass-rich charges containing 15–25 vol% large, euhedral apatite crystals (± cpx and minor biotite). Experimental apatites and glasses have been characterised by BSE imaging, electron microprobe, and ion microprobe.
Apatites range in composition from near-endmember fluorapatite (3.0 wt% F), to near-endmember hydroxyapatite (1.7 wt% H2O), to carbon-rich apatite containing up to 1.6 wt% CO2. Apatite compositions are stoichiometric if all anions (F–, OH–, and CO32—) lie in the channel site, suggesting an “A-type” substitution under these conditions (i.e. CO32— + [] = 2X—, where X is another channel anion and [] is a vacancy; e.g. Fleet et al. 2004). Importantly, CO2 partitions readily into apatite at all fluid compositions considered here. CO2 is also more compatible in apatite than water at our run conditions, with calculated H2O–CO2 exchange coefficients close to or greater than 1. Our results indicate that when channel ions are primarily occupied by H2O and CO2 (i.e. F- and Cl-poor magmatic systems), apatite can preserve higher absolute CO2 concentrations than the surrounding melt. In this way, apatite may prove more sensitive than other direct records of pre-eruptive volatile contents, such as phenocryst-hosted melt inclusions. We suggest that apatite has the potential to be a useful tool for tracking magmatic CO2 abundances.
