Sulfur partition coefficient between apatite and rhyolite: the role of bulk S content

Abstract

Experimental studies have been performed to constrain sulfur behavior during apatite crystallization and to determine sulfur partition coefficient between apatite and melt (Kd S apatite/melt ) at oxidizing conditions. Crystallization experiments have been conducted with a hydrous rhyolitic melt and different bulk sulfur contents (0.15 to 2 wt.% S) at 900 and 1,000°C, 200 MPa and Δlog $$f_{{{\rm O}_{2} }} $$ =NNO+3.6. The sulfur content in the glass increases with increasing amount of added S. Anhydrite crystallizes for S added = 0.75 wt.% (0.10 and 0.13 wt.% SO3 in glass at 900 and 1,000°C, respectively). The amount of anhydrite increases and the amount of apatite decreases with increasing amount of added sulfur. The sulfur exchange reaction in apatite is influenced by the bulk composition of the melt (e.g., P content). However, changing melt composition has only little effect on Kd S apatite/melt for the investigated rhyolitic composition. The Kd S apatite/melt does not depend directly on temperature, decreases from 14.2 to 2.7 with increasing S content in glass from SO3=0.03 to 0.19 wt.%, respectively, and can be predicted from the following equation: ln Kd = −0.0025×S in melt (in ppm)+2.9178. The combination of experimental data obtained for rhyolitic and andesitic melts reveals that the sulfur partition coefficient tends toward a value of 2 for high-sulfur content in the glass (>0.2 wt.% SO3). Using S in apatite as proxy for determining S content in melt is promising but additional experimental data are needed to clarify the individual effects of T, $$f_{{{\rm S}_{2} }} $$ , and P and Ca content in the melt on S partitioning.