The brine-CO2 partition coefficients (DiV/L = CiV/CiL) calculated from the experimental data ranged from 4 ¡Á 10-4 ?1 ¡Á 10-3 for Fe, 8 ¡Á 10-5 ?7 ¡Á 10-4 for Cu, 4 ¡Á 10-6 ?2 ¡Á 10-4 for Zn, and 6 ¡Á 10-6 ?5 ¡Á 10-4 for Na. The fractionation of these elements into the CO2 did not cause a measureable decrease of their concentrations in the brine. The total element concentrations in the CO2 samples ranged from 0.2 to 0.6 mg/kg (ppm) Fe, 0.1 to 0.6 mg/kg Cu, 0.004 to 0.4 mg/kg Zn and 0.4 to 39 mg/kg Na, with the Na concentrations generally displaying a positive correlation to CO2 density. The values of DiV/L in a typical CO2 storage reservoir at 7.0-28.5 MPa and 35-98¡ãC, in which the density of CO2 is ~ 0.3?.7 g/cm3 (as compared to 0.1-0.6 g/cm3 in this study), could be expected to be approximately equivalent to those determined in this study.
Considering the metal concentrations typical of brines in CO2 storage aquifers that have reacted with CO2 and sandstone (20?00 mg/kg Fe, 0.3? mg/kg Cu, 3? mg/kg Zn), these results suggest that a plume of injected CO2 could contain up to 0.1 mg/kg Fe, 0.3 ¦Ìg/kg (ppb) Cu and 1 ¦Ìg/kg Zn, in addition to 16 mg/kg Na. In a Sleipner (North Sea)-sized reservoir used to store 14 Mt of CO2, this would lead to the mobilization of about 1 t Fe, 5-10 kg Cu and Zn, and 200 t Na. In terms of long-term CO2 storage, the potential consequences of these results include the precipitation of carbonate minerals in shallower, more distal regions of the aquifer and the transferral of metals to adjacent aquifer systems.