In order for carbon capture and storage (CCS) to have a significant impact on climate change, very large amounts of captured CO2 need to be injected and stored underground.  This requires quantitative predictions of the behavior and ultimate fate of the injected CO2 as well as any displaced resident fluids such as brines.  Part of these quantitative predictions involves an analysis of the probability of fluid leakage out of the injection reservoir.  We have developed a range of computational models to make these kinds of predictions, including estimates of leakage along old wells.  When coupled with recent field measurements involving physical testing in a set of old wells, we can begin to make appropriate leakage estimates for different injection formations.  This will be demonstrated for one particular field location in Alberta, Canada.  In addition to calculation of fluid leakage, our models also predict the fraction of CO2 trapped by local capillary forces within the pores, and the fraction of CO2 that dissolved into the resident brine.  This provides a prediction of long-term storage security.  Finally, our models also predict pressure increases in the subsurface and, based on EPA guidance, can indicate the areas of review associated with injection projects.  Ongoing analysis of the Basal Aquifer formation, which spans western Canada and the northern plains states of the US, shows how this kind of analysis can guide in well design and placement.  Results for injections into different formations will be presented, and comments about how our analysis tools may be used for other kinds of systems, including shale gas, will be discussed.

Geological carbon storage: Models for leakage risk and large-scale deployment

246th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.