DOE OSTI2014
Ultrasonic P- and S-wave velocities were measured over a range of confining pressures while injecting CO2 and brine into the samples. Pore fluid pressure was also varied and monitored together with porosity during injection. Effective medium models were developed to understand the mechanisms and impact of observed changes and to provide the means for implementation of the interpretation methodologies in the field. Ultrasonic P- and S-wave velocities in carbonate rocks show as much as 20-50% decrease after injection of the reactive CO2-brine mixture; the changes were caused by permanent changes to the rock elastic frame associated with dissolution of mineral. Velocity decreases were observed under both dry and fluid-saturated conditions, and the amount of change was correlated with the initial pore fabrics. Scanning Electron Microscope images of carbonate rock microstructures were taken before and after injection of CO2-rich water. The images reveal enlargement of the pores, dissolution of micrite (micron-scale calcite crystals), and pitting of grain surfaces caused by the fluid- solid chemical reactivity. The magnitude of the changes correlates with the rock microtexture – tight, high surface area samples showed the largest changes in permeability and smallest changes in porosity and elastic stiffness compared to those in rocks with looser texture and larger intergranular pore space. Changes to the pore space also occurred from flow of fine particles with the injected fluid. Carbonates with grain-coating materials, such as residual oil, experienced very little permanent change during injection. In the tight micrite/spar cement component, dissolution is controlled by diffusion: the mass transfer of products and reactants is thus slow and the fluid is expected to be close to thermodynamical equilibrium with the calcite, leading to very little dissolution, or even precipitation. In the microporous rounded micrite and macropores, dissolution is controlled by advection: because of an efficient mass transfer of reactants and products, the fluid remains acidic, far from thermodynamical equilibrium and the dissolution of calcite is important. These conclusions are consistent with the lab observations. Sandstones from the Tuscaloosa formation in Mississippi were also subjected to injection under representative in situ stress and pore pressure conditions. Again, both P- and S-wave velocities decreased with injection. Time-lapse SEM images indicated permanent changes induced in the sandstone microstructure by chamosite dissolution upon injection of CO2-rich brine. After injection, the sandstone showed an overall cleaner microstructure. Two main changes are involved: (a) clay dissolution between grains and at the grain contact and (b) rearrangement of grains due to compaction under pressure Theoretical and empirical models were developed to quantify the elastic changes associated with injection. Permanent changes to the rock frame resulted in seismic velocity-porosity trends that mimic natural diagenetic changes. Hence, when laboratory measurments are not available for a candidate site, these trends can be estimated from depth trends in well logs. New theoretical equations were developed to predict the changes in elastic moduli upon substitution of pore-filling material. These equations reduce to Gassmann’s equations for the case of constant frame properties, low seismic frequencies, and fluid changes in the pore space. The new models also predict the change dissolution or precipitation of mineral, which cannot be described with the conventional Gassmann theory.