Aquifer disposal of CO2-rich greenhouse gases: Extension of the time scale of experiment for CO2-sequestering reactions by geochemical modelling

In previous work, Gunter et al. (1993), suggested water-rock reactions in deep aquifers in sedimentary basins could sequester injected-CO2-waste from industry, thereby reducing greenhouse gas emissions. Experiments, carried out at 105 degrees C and 90 bars CO2 pressure, to test the validity of this mineral-trapping of CO2 were unsuccessful due to sluggish kinetics of reaction. The most significant change recorded by the reaction products from these experiments was a large increase in alkalinity, which was attributed to very small amounts of water-mineral reaction. A computer model, PATHARC.94, was used to interpret this change in alkalinity and to predict the path and time necessary to reach equilibrium. Substantial trapping of CO2 by formation of siderite, calcite and aqueous bicarbonate ions was predicted to occur in 6 to 40 years. Potential errors as high as two orders of magnitude were estimated based on a thorough examination of the kinetic data used in the modelling. In order to achieve reasonable time estimates, ''reactive'' surface areas were approximated by 100 micron spherical grains in the computer model. This represents a smaller cumulative surface area than actually present in the experiment. When these results are extrapolated to the field, where the aquifers are at lower temperatures, Perkins and Gunter (1995a), concluded that CO2-trapping reactions are expected to take 100 s of years to complete. This is sufficient time for the trapping to occur as the residence time of a packet of fluid in a deep low-permeability aquifer in a sedimentary basin is measured in 10,000 s to 100,000 s of years.