Back to search

FRINATEK-Fri prosj.st. mat.,naturv.,tek

Mechanisms for CO2 storage in the presence of residual oil

Alternative title: Mekanismer for lagring av CO2 i nærvær av gjenværende olje

Awarded: NOK 11.3 mill.

CO2 storage in mature oil reservoirs is an attractive strategy because infrastructure exists, and the geology is known from the oil production phase. Such reservoirs often contain residual oil trapped as disconnected ganglia in the pore space after water injection. Thus, a CO2 storage strategy requires knowledge of how water and this residual oil influence the CO2 flow pattern and storage mechanisms, and conversely how CO2 invasion impacts the behavior of the residing fluids. CO2 dissolution in oil and water will change the fluid properties, lead to oil swelling, and potentially alter the wetting state of the porous rock. These processes can serve both oil recovery and CO2 storage. Cyclic injection, which often is a realistic storage option when CO2 availability is low, can optimize residual trapping, but reservoir simulation is challenging because flow mechanisms are complex. Little is known about how the behavior changes when residual oil ganglia are present. This project will bring forth advanced pore-scale models for three-phase flow to explore how CO2 dissolution in the presence of water and oil affects residual CO2 trapping, oil trapping and mobilization, capillary pressure, relative permeability, and hysteresis behavior over multiple CO2/water invasion cycles in porous rock. The models will be validated against a wide range of advanced CO2/oil/water pore-scale experiments provided by our international research partners. The validated pore-scale simulators will be released as open-source so that users can make their own calculations with the aim at reducing the number of required lab measurements for CCUS operations. The project will also bring forth suitable macroscale three-phase flow models that capture the effective three-phase flow behavior observed at pore scale. This is a first necessary step towards reliable simulation of large-scale CCUS operations in mature oil reservoirs. In the reporting period, we have continued the work on the development of two pore scale models for three-phase flow (Task 1 and 2). The first model is currently developed for incompressible and immiscible three-phase flow, coupled with a locally conservative level set approach for interface tracking. We have recently implemented a dynamic contact angle in the code and are investigating methods that can improve the accuracy of the numerical description. As part of this activity, we have also started to extend a method for diffusion-driven coarsening of gas bubble populations to allow gas mixtures. This will allow investigations of the behavior of trapped ganglia containing CO2 and other fluid components. The second model, which is based on the lattice-Boltzmann method, has been extended to allow for partial dissolution of phases in two-phase systems. Current work in this activity focuses on numerical testing and preparation of a manuscript for submission to a journal. Previously in the project, we have used the three-phase level set method to simulate gas-water invasion cycles in the presence of oil for comparison with two data sets provided by our international research partners (ICL and OSU) (Task 6). The first set explores three-phase gas and oil trapping on sand packs. The results agree well with data and show that the total amount of trapping is higher in three-phase systems and that more gas and less oil get trapped with higher initial gas saturation. The other simulation set explores spatial and temporal behavior of contact angles during rate-controlled displacement on sphere packs. In the reporting period, we have also simulated an extensive set of CO2-water invasion cycles from different initial saturations on a sandstone pore geometry to explore differences in hysteresis behavior of connected phases and disconnected ganglia. The results show hysteresis between CO2 and water invasion of the number of oil and CO2 ganglia and the disconnected phase saturations. We have also performed simulations of diffusion-driven coarsening of trapped CO2 ganglia in porous rock. From all these simulation activities we are preparing four manuscripts that we will submit to journals spring 2024. The PhD student has started to study the level set codes and the discrete-domain model for hysteresis. He has extended the three-phase discrete-domain model to allow compartments with different pore volumes and will start using data from pore-scale simulations and experiments to generate energy landscapes for the model (Task 8). We have also set up a compositional model for oil and CO2 and a black-oil model for oil and water, using an open-source reservoir simulator (Task 9). These setups were validated against other simulators and will be extended to three phases. In the reporting period, we have started to develop multiphase flow formulations based on separate saturations for the connected and disconnected phases, with the aim at utilizing data from pore-scale simulations in the model.

CO2 storage in mature hydrocarbon reservoirs is an attractive strategy because infrastructure exists, and the geology is known from the oil production phase. However, such a strategy requires knowledge of how water and residual oil influence the CO2 invasion pattern and storage mechanisms, and conversely how CO2 invasion impacts the behavior of residual oil ganglia and water. The presence of residual oil in such reservoirs introduces CO2 dissolution in oil and water that can serve both recovery and storage. Cyclic injection, which often is a realistic option due to low CO2 availability, can optimize residual trapping, but reservoir simulation is challenging because flow properties and hysteresis are complex. Little is also known about how this behavior changes when residual oil ganglia are present, and CO2 dissolution alters the oil properties (like viscosity and density), which lead to oil swelling and potentially changed wetting state of the porous rock. This project will bring forth advanced pore-scale models for three-phase flow to investigate how CO2 dissolution in the presence of water and oil affects residual CO2 trapping, oil trapping and mobilization, capillary pressure, relative permeability, and hysteresis behavior over multiple CO2/water invasion cycles in porous rock. The models will be validated against a wide range of advanced CO2/oil/water pore-scale experiments provided by our international research partners. The validated pore-scale simulators will be released as open-source so that users can make their own calculations with the aim at reducing the number of required lab measurements for CCUS operations. The project will also bring forth suitable macroscale three-phase flow models that capture the effective three-phase flow behavior observed at pore scale, including fluid-ganglia dynamics. This is a first necessary step towards reliable simulation of large-scale CCUS operations in mature hydrocarbon reservoirs.

Funding scheme:

FRINATEK-Fri prosj.st. mat.,naturv.,tek