To understand possible volcanogenic fluxes of CO2 to the Martian atmosphere, we investigated experimentally carbonate solubility in a synthetic melt based on the Adirondack-class Humphrey basalt at 1-2.5GPa and 1400-1625°C. Starting materials included both oxidized and reduced compositions, allowing a test of the effect of iron oxidation state on CO2 solubility. CO2 contents in experimental glasses were determined using Fourier transform infrared spectroscopy (FTIR) and Fe3+/FeT was measured by Mössbauer spectroscopy. The CO2 contents of glasses show no dependence on Fe3+/FeT and range from 0.34 to 2.12wt.%. For Humphrey basalt, analysis of glasses with gravimetrically-determined CO2 contents allowed calibration of an integrated molar absorptivity of 81,500±1500Lmol-1cm-2 for the integrated area under the carbonate doublet at 1430 and 1520cm-1. The experimentally determined CO2 solubilities allow calibration of the thermodynamic parameters governing dissolution of CO2 vapor as carbonate in silicate melt, KII, (Stolper and Holloway, 1988) as follows: lnKII0=-15.42±0.20, ΔV0=20.85±0.91cm3mol-1, and ΔH0=-17.96±10.2kJmol-1. This relation, combined with the known thermodynamics of graphite oxidation, facilitates calculation of the CO2 dissolved in magmas derived from graphite-saturated Martian basalt source regions as a function of P, T, and fO2. For the source region for Humphrey, constrained by phase equilibria to be near 1350°C and 1.2GPa, the resulting CO2 contents are 51ppm at the iron-wüstite buffer (IW), and 510ppm at one order of magnitude above IW (IW+1). However, solubilities are expected to be greater for depolymerized partial melts similar to primitive shergottite Yamato 980459 (Y 980459). This, combined with hotter source temperatures (1540°C and 1.2GPa) could allow hot plume-like magmas similar to Y 980459 to dissolve 240ppm CO2 at IW and 0.24wt.% of CO2 at IW+1. For expected magmatic fluxes over the last 4.5Ga of Martian history, magmas similar to Humphrey would only produce 0.03 and 0.26bars from sources at IW and IW+1, respectively. On the other hand, more primitive magmas like Y 980459 could plausibly produce 0.12 and 1.2bars at IW and IW+1, respectively. Thus, if typical Martian volcanic activity was reduced and the melting conditions cool, then degassing of CO2 to the atmosphere may not be sufficient to create greenhouse conditions required by observations of liquid surface water. However, if a significant fraction of Martian magmas derive from hot and primitive sources, as may have been true during the formation of Tharsis in the late Noachian, that are also slightly oxidized (IW+1.2), then significant contribution of volcanogenic CO2 to an early Martian greenhouse is plausible.
Bibliographical noteFunding Information:
The authors thank Justin Filiberto, Bruno Scaillet, Simon Kohn, and the associate editor F.J. Ryerson for their comments, which greatly improved the quality of this manuscript. This work is supported by NASA Mars Fundamental Research Program Grants 07-MFRP07-0038 and 10-MFRP10-0069 . The authors also are grateful to NSF for the Grant, EAR0930034 that enabled the purchase of the Bruker FTIR. Electron microprobe analyses were carried out at the Electron Microprobe Laboratory, Department of Earth Sciences, UMN. The original starting material was generously provided by Justin Filiberto. Glass thickness measurements were carried out in the Institute of Technology Characterization Facility, UMN, which receives partial support from NSF through the NNIN program. Mössbauer analyses were conducted by Thelma Berquó at the Institute for Rock Magnetism, Department of Earth Sciences, UMN.