What Are the Time Scales for Carbonate Mineral Sequestration in the Subsurface?
Abstract
Mineral trapping of CO2 in the subsurface is acknowledged to be the most secure form of sequestration, but some studies have suggested that the process is extremely slow, perhaps on the order of 10,000 years or more. But what are the arguments for these long time scales based on? Certainly part of it has to do with the slow dissolution rates of silicates needed to provide a source of cations (Ca2+, Mg2+, and Fe2+) and alkalinity for carbonate precipitation. Rates of dissolution for many silicates are very slow (e.g., albitic plagioclase and chlorite), while other silicate minerals (anorthitic feldspar, olivine) dissolve appreciably faster. Determining which mineral is rate-limiting in the case of the faster dissolving silicates (is it the dissolving silicate or precipitating carbonate?), however, is not always straightforward without a careful analysis of dissolution and precipitation as a coupled process. We use micro-continuum modeling to address the issue of time scales for carbonate mineral trapping of CO2. Two sets of simulations have been carried out: 1) largely generic simulations at the centimeter scale that include heterogeneous distributions of both residually trapped scCO2 and reactive mineral phases (Ca-bearing plagioclase and chlorite), and 2) flow plus diffusion and pure diffusion simulations of the reactivity of a reservoir sandstone from a scCO2 pilot injection site at Cranfield, Mississippi. The first set of simulations are designed specifically to examine the assumption that the reservoir within which carbonate mineral precipitation might occur is well-mixed and therefore characterized everywhere by the low pH values typical of brine in equilibrium with supercritical CO2. Our analysis of physically and chemically heterogeneous subsurface materials, especially during the residual trapping stage, suggest that local chemical microenvironments can develop in which pH, alkalinity, and cation concentrations rise sufficiently high that substantial carbonate can precipitate. The second set of simulations make use of initial porosity (both connected and unconnected) and mineral distributions mapped in a sample from the pilot injection site at Cranfield, Mississippi (Landrot et al., 2012). For the calculation of flow, various porosity-permeability relationships are applied at the micron scale within the micro-continuum model. Preliminary modeling both suggest that carbonate precipitation can be significant on the time scale of tens of years, although the results do scale with the reactivity of the dissolving silicate phase.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.V34A..03S
- Keywords:
-
- 1009 GEOCHEMISTRY Geochemical modeling