Abstract
Carbon dioxide injection into sedimentary formations has been widely used in enhanced oil recovery (EOR) and geological-storage projects. Several field cases have shown an increase in water injectivity during CO2 Water-Alternating-Gas (WAG) projects. Although there is consensus that the rock-fluid interaction is the main mechanism, modeling this process is still challenging. Our main goal is to validate a physically based model on experimental observations and use the validated model to predict CO2 injectivity alteration based on geochemical reactions in carbonate rocks.
In this paper, we present a new method for CO2 reactive transport in porous media and its impact on injectivity. We hypothesize that if CO2 solubilizes in the connate water, then it induces a shift in chemical equilibrium that stimulates mineral dissolution. Consequently, porosity and permeability will increase, and cause alterations to well injectivity. We develop a predictive model to capture this phenomenon and validate the model against available data in the literature. We use UTCOMP-IPhreeqc, which is a fully coupled fluid-flow and geochemical simulator to account for rock/hydrocarbon/water interactions. In addition, we perform several experiments to test CO2/water slug sizes, mineralogy assembly, injected brine composition, and gravity segregation combined with the effect of heterogeneity.
Coreflood simulations using chemical equilibrium and kinetics indicate mineral dissolution at reservoir conditions. The results suggest that the intensity of rock dissolution depends on formation mineralogy and brine composition as carbonate systems work as buffers. Additionally, we show that prolonged CO2 and brine injection induces petrophysical alteration close to the injection region. Our field-scale heterogeneous reservoir simulations show that permeability alteration calculated based on Carman-Kozeny correlation and wormhole formulation had the same results. Furthermore, we observed that water injectivity increased by almost 20% during subsequent cycles of CO2-WAG. This finding is also supported by the Pre-Salt carbonate field data available in the literature. In the case of continuous CO2 injection, the carbonate dissolution was considerably less severe in comparison with WAG cases, but injectivity increased due to unfavorable CO2 mobility. With the inclusion of gravity segregation, we report that the injectivity doubles in magnitude. The simulations show more extensive dissolution at the upper layers of the reservoir, suggesting that preferential paths are the main cause of this phenomenon. The ideas presented in this paper can be utilized to improve history-matching of production data and consequently reduce the uncertainty inherent to CO2-EOR and carbon sequestration projects.