Experimental determination of the solubility of carbon dioxide in molten basalt at low pressure

We report the first measurements of CO_2 solubility in molten basalt at pressures comparable to those at which submarine basalts erupt. A basalt from the Juan de Fuca ridge was equilibrated with CO_2-rich vapor at 1200°C, 100–1500 bar for up to four hours. After quenching, the glass was analyzed for dissolved carbonate ions by infrared spectroscopy. No forms of dissolved CO_2 other than carbonate were detected. CO_2 solubility is roughly a linear function of pressure at these low pressures. The experimentally determined solubility differs from previous estimates based on CO_2 concentrations of submarine glasses, on CO_2 solubilities in basaltic liquids at significantly higher pressures, and on CO_2 concentrations of glasses equilibrated with H_2O-CO_2 vapor. Our results are compatible with those obtained previously at higher pressures on a molten Kilauea tholeiite only if there is a significant positive dependence of carbonate solubility on temperature. CO_2 contents of mid-ocean ridge glasses measured by infrared spectroscopy are generally higher than would be expected based on solubilities at the hydrostatic pressures for the water depths from which the glasses were recovered, but the lowest dissolved CO_2 contents agree with the experimentally determined solubilities. We propose that submarine glasses with low CO_2 contents were quenched from magmas that were able to degas because they rose slowly from depth. The common occurrence of glasses with dissolved CO_2 contents in excess of the experimentally determined solubility suggests they were quenched from magmas that ascended too rapidly to degas fully. In conjunction with our solubility data, the highest CO_2 contents allow minimum estimates of depths to magma chambers. Depths of ⩾ 2.3 km beneath the ridge are indicated for the East Pacific Rise at 21°N, in agreement with geophysical constraints