MODELING THE LONG-TERM ISOLATION PERFORMANCE OF NATURAL AND ENGINEERED GEOLOGIC CO_2 STORAGE SITES

摘要

Long-term cap rock integrity represents the single most important constraint on the long-term isolation performance of natural and engineered geologic CO_2 storage sites. CO_2 influx that forms natural accumulations and CO_2 injection for EOR/sequestration or saline-aquifer disposal both lead to concomitant geochemical alteration and geomechanical deformation of the cap rock, enhancing or degrading its seal integrity depending on the relative effectiveness of these interdependent processes. This evolution of cap-rock permeability can be assessed through reactive transport modeling, an advanced computational method based on mathematical models of the coupled physical and chemical processes catalyzed by the influx event. Using our reactive transport simulator (NUFT), supporting geochemical databases and software (SUPCRT92), and distinct-element geomechanical model (LDEC), we have shown that influx-triggered mineral dissolution/precipitation reactions within typical shale cap rocks continuously reduce microfrac apertures, while pressure and effective-stress evolution first rapidly increase then slowly constrict them. For a given shale composition, the extent of geochemical enhancement is nearly independent of key reservoir properties (permeability and lateral continuity) that distinguish saline aquifer and EOR/sequestration settings and CO_2 influx parameters (rate, focality, and duration) that distinguish engineered disposal sites and natural accumulations, because these characteristics and parameters have negligible impact on mineral reaction rates. In contrast, the extent of geomechanical degradation is highly dependent on these reservoir properties and influx parameters, because they effectively dictate magnitude of the pressure perturbation. Specifically, initial geomechanical degradation has been shown inversely proportional to reservoir permeability and lateral continuity and proportional to influx rate. As a result, while the extent of geochemical alteration is nearly independent of filling mode, that of geomechanical deformation is significantly more pronounced during engineered injection. This discrepancy limits the extent to which natural CO_2 reservoirs and engineered disposal sites can be considered analogous, and further suggests that the secure cap rock of a given natural CO_2 accumulation may be incapable of providing an effective seal in the context of an engineered injection. A new conceptual framework reveals that ultimate counterbalancing of opposing geochemical and geomechanical effects is feasible, which suggests that shale cap rocks may evolve into effective hydrodynamic seals-in both natural and engineered storage sites-as a function of progressive geochemical alteration that attends some degree of initial CO_2 leakage.