Paper No. 18
Presentation Time: 9:00 AM-6:30 PM


PICCOLI, Philip, Department of Geology, University of Maryland, College Park, College Park, MD 20742 and WEBSTER, James D., American Museum of Natural History, Department of Earth and Planetary Sciences, Central Park West at 79th St., New York, NY 10024,

Halogens play an important role in a variety of magmatic processes including the transport and concentration of metals, magma rheology, and the explosivity of volcanic eruptions. The primary halogen-bearing phases in magmatic systems are biotite, amphibole and apatite: all have been used to evaluate halogen budgets. In some cases, apatite is more appropriate for modeling halogens for the following reasons: 1) it is nearly ubiquitous in magmatic systems over a wide range of PTXconditions, 2) modeling of activity composition relationships is comparatively simple, and 3) it is though to be more resistant to sub-solidus re-equilibration in many environments. What we describe here is an expansion of a model to use the composition of apatite as an estimator of melt and aqueous fluid Cl and F (Piccoli and Candela, 2004). This work represents two significant changes to existing modeling efforts: modification of the algorithms to allow for the calculations to be used to model systems at lower ASIs (peralkaline), and an evaluation of the non-ideal mixing effect of along the F-Cl end-member join. Mixing of apatite end-members has often been through to be ideal, however, Hovis and Harlov (2010) found that mixing along the F-Cl join is non-ideal (Hex maximum of 8.3kJ/mol) and asymmetric with respect to composition. Data on the mixing relationship F-OH and Cl-OH, but don’t exist. As a test of the appropriateness of the model, we compare output of the model to halogens in Augustine melt inclusions as a test of the accuracy of the predictions.

Lavas from the 2006 Augustine eruption contain apatite containing 2.16 wt% F and 1.19 wt% Cl (mole fraction of FAp+CAp range from 0.54 to 0.94). In equilibrium with melt at 100 MPa and 800oC, those apatites would be in equilibrium with the system at log f(HCl)/f(H2O) of -2.45 and f(HF)/f(H2O) of - 5.2. These values would correspond to Cl and F concentrations of 3650 and 40 ppm in the melt. Melt inclusions from the 2006 eruptions have Cl and F concentrations of 3300-3800 ppm and ~200-300 ppm, respectively. If melt inclusions are representative of the melt composition at the time of the bulk of apatite crystallization, apatite can be used as an accurate predictor of melt Cl but underestimates melt F.