Storage of CO2 has become an important topic in the past few years. This is because excessive CO2 in the atmosphere is considered as the main factor responsible for climate change and green house effect. The emission of CO2 gas is mainly caused by human activities from different sources such as electric power plant, petroleum refinery, transportation and natural gas consumption. Therefore, the increase of CO2 gas in the atmosphere needs be reduced. Currently, CO2 storage in deep saline aquifers is an attractive approach that is being studied extensively. Saline aquifers have large volumes that can handle a large amount of CO2 injection. Furthermore, CO2 can be easily dissolved in brine. However, the above storage may not be secure if the stress and deformation of the aquifer are not analyzed thoroughly. The main focus of this work is using the coupled code between reservoir flow and geomechanics to study when and where gas leakages may occur through the caprock under different scenarios such as injection methods and orientation of injecting wells. A modified Barton-Bandis model is implemented to compute permeability of fractures that occur in the caprock when its tensile strength is overcome by applied stresses, which then allows free CO2 to escape from the aquifer. Examples are presented to illustrate the workflow in geomechanical risk mitigation of CO2 storage.


The excess of CO2 amount in the atmosphere is thought to be a main factor to cause green house effect and global warming. One of proposals to reduce the CO2 emission is to capture and store it in deep saline aquifers. However, this storage is secure only when there is no leakage of CO2 from the aquifer. The potential for leakage depends on many factors such as formation characteristics, caprock integrity, trapping mechanisms, rate of storage and chemical/mineral reactions, etc. The CO2 storage in aquifers has been vastly studied by many researchers. Kumar and Bryant [1] studied injection strategies to maximize CO2 trapping and minimize the leakage potential. According to their work, the leakage potential is subject to three response variables such as fraction of mobile CO2 in the aquifer, the maximum lateral extent of the plume from the injector and time for the plume to reach the caprock. With the help of semi-analytical model that they developed, for a desired injection rate, the perforation interval can be optimized to provide uniform flux distribution and to keep the pressure in the aquifer below the fracture pressure. Burton and Bryant [2] used the approach of dissolving CO2 in brine prior to injection into a deep formation. In their process, brine produced from a well completed in the aquifer is used to dissolve CO2 and then the CO2-saturated brine is injected back to the aquifer through an injecting well located far away from the producing well. The CO2- saturated brine is denser than the aquifer brine, which eliminates the risk of buoyancy-driven leakage that might happen when only CO2 is injected alone.

This content is only available via PDF.
You can access this article if you purchase or spend a download.