Chloride partitioning and solubility in hydrous phonolites from Erebus volcano: A contribution towards a multi-component degassing model

Abstract

International audienceWe present results from a series of experiments on the partitioning of chlorine between a hydrous fluid phase and a phonolitic melt that we then use to model specific aspects of volcanic degassing. Experiments were performed from 250 to 10 MPa on a phonolite from Erebus lava lake, Antarctica, at 1000 °C near the QFM solid buffer. Only one of 48 experimental samples shows coexistence of low-density vapour and high-density brine in the fluid phase while 35 samples are unambiguously in the vapour field. Large uncertainties in the H 2 O–NaCl phase diagram at the P–T considered do not allow us to assign reliable phase behaviour to the remaining samples. We select a vapour-dominated subset of our dataset to estab-lish an empirical HCl solubility law valid at low pressures and salinities. This law is incorporated into a thermodynamical model of degassing, which we use to compute equilibrium temperatures from gas measurements made at Erebus in 2010. The quiescent lake activity features cyclic temperature variations between 1000 and 1050 °C, which is consistent with thermal convection within the shallow plumbing system. Backward tracking of representative gas compositions and temperatures shows the evolution with pressure of the molar ratio of SO 2 /HCl in the gas, a quantity that is often measured in volcanic plumes. Model outputs suggest that this ratio can vary by up to an order of magnitude when the pressure changes from atmospheric to 100 MPa, depending on degassing style (coupled vs. decoupled ascent of gas and melt). Ó 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/)

Experimentally derived F, Cl, and Br fluid/melt partitioning of intermediate to silicic melts in shallow magmatic systems

“…The propagated errors are plotted in figures 4-8. Previous studies have shown that these propagated errors often overestimate the absolute error and thus can be considered maximum values (e.g., Alletti et al, 2014;Cadoux et al, 2018). Finally, Quizapu and Kelud both contain trace amounts of apatite (<1 vol%) and Quizapu contains ~2 vol% hornblende Drake 1992 Ruprecht et al, 2012;Jeffery et al, 2014;Cassidy et al, 2016;2019;First et al 2021).…”

Cyclic degassing of Erebus volcano, Antarctica

“…The D-Compress thermodynamic model is based on experimentally determined solubility laws for key species (H 2 O, CO 2 , H 2 , SO 2 , H 2 S) and redox equilibrium equations for gaseous carbon-, oxygen-, hydrogen-, and sulphur-bearing species (Burgisser and Scaillet 2007;Burgisser et al 2012;Alletti et al 2014). Given relative proportions of these volatiles as gas ratios, CO 2 /H 2 O, CO 2 /CO, OCS/SO 2 , CO 2 /SO 2 , and optionally CO 2 /HCl, and assuming that the gases are at equilibrium, D-Compress calculates equilibrium temperature at a given pressure.…”

“…However, interpreting degassing history is complicated since the composition of the gas phase changes as magma ascends. With an increasing body of experimental work to understand the solubility and redox relationships of different volatile species, thermodynamic models for degassing that incorporate empirical solubility laws are becoming available (Burgisser et al 2012(Burgisser et al , 2015Alletti et al 2014).…”