Two-Phase Cement/CO2/Brine Interaction in Wellbore Environments

Abstract

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 126666, ’Computational Studies of Two-Phase Cement-CO2-Brine Interaction in Wellbore Environments,’ by J.W. Carey and P.C. Lichtner, Los Alamos National Laboratory, originally prepared for the 2009 SPE International Conference on CO2 Capture, Storage, and Utilization, San Diego, California, 2-4 November. Wellbore integrity is essential to ensuring long-term isolation of buoyant supercritical carbon dioxide (CO2) during geologic sequestration. The full-length paper summarizes recent progress in numerical simulations of cement/CO2/brine interactions with respect to migration of CO2 outside of casing. Using typical values for the hydrologic properties of cement, caprock (shale), and reservoir materials, the study shows that the capillary properties of good-quality cement will prevent flow of CO2 into and through cement. Introduction Wellbore integrity signifies isolation of fluids in the subsurface. In the geologic sequestration of CO2, this requires preventing the migration of buoyant, supercritical CO2, a fluid that is highly mobile (with a viscosity approximately 1/10 that of water) and that is partially immiscible with brine and oil. Aqueous CO2 forms by reaction of supercritical CO2 and brine and generates an acidic solution (pH in the range of 4.5 to 5 in the presence of calcite) that is chemically reactive with the Portland cement and steel casing that provide isolation in most wellbore systems. Wellbore failure in the context of geologic sequestration is leakage that occurs either through primary defects (e.g., along a microannulus at the cement/casing interface) or through secondary defects induced by chemical degradation of the wellbore system. One of the key goals in the evaluation of the long-term safety and security of CO2 sequestration is thus the determination of whether zonal isolation deteriorates or improves through chemical interactions with CO2, in addition to calculation of the flux of CO2 as a function of conditions in the wellbore environment. Chemical reaction of CO2 with Portland cement is complex and involves dissolution of portlandite, dis-solution of calcium silicate hydrate (the primary structural material in cement), and the precipitation of calcium carbonate (CaCO3). If the first two reactions predominate, the flux of CO2 will increase with time as the cement sheath deteriorates. On the other hand, if precipitation of CaCO3 is significant, CO2-leakage pathways may self-seal and ultimately limit the flux of CO2. The simulations were motivated by observations of cement performance at the CO2-enhanced-oil-recovery SACROC Unit in west Texas and by recent results from a natural CO2 reservoir. The SACROC well was 55 years old with 30 years of operation as a CO2 producer and injector. Samples of cement from 3 m above the reservoir/caprock interface showed reactions with CO2 at both the casing and formation interfaces. Preliminary 1D diffusion-based models were able to reproduce the observed mineralogical changes at SACROC. In particular, the model reproduced the complete carbonation of the cement along a 0.5-cm reaction front adjacent to the caprock and the presence of a siliceous deposit at the interface between the carbonated and uncarbonated cement.