Modeling of CO2 Injection to Shale Gas Reservoir with Organic/Inorganic Network

Abstract

Though studies on horizontal well and hydraulic fracture have been performed extensively, the effects of organic matter on shale gas production are not well known. In this study, an improved numerical model, which reflects the organic and inorganic matters in the shale gas reservoirs, is developed. To confirm the improved accuracy of the developed model, four numerical models are built and compared to match the field data of Eagle Ford shale reservoir. These models are classified with (1) bulk model that reservoir data is applied identically and (2) OM (organic matters)/IOM (inorganic matters) model which is distinguished into the organic and inorganic sections. Organic matters in OM/IOM model are randomly distributed by total organic carbon (TOC). Bulk and OM/IOM models are sub-classified with geomechanical effects into four models, such as Bulk, Bulk-geo, OM/IOM, and OM/IOM-geo models. Since the organic matter is much softer than minerals, the elastic properties of OM/IOM-geo model are also divided into organic matter and minerals. History matching of four models is conducted to verify the effects of organic matter. The history matching error of OM/IOM model is less than bulk model and the models, which reflects the geomechanical effects, have less error than that do not; 4% for bulk-model, 2.92% for bulk-geo model, 3.11% for OM/IOM model, and 2.41% for OM/IOM-geo model. Since the organic matter is scattered in actual shale gas reservoirs, OM/IOM model, which can describe it, should be considered. Based on these results, numerical modeling for CO2 injection was carried out. Bulk and OM/IOM models are used to validate the difference of CO2 EGR efficiency. The cumulative CH4 production of bulk and OM/IOM models increases by up to 6% and 2% by CO2 flooding. The difference in EGR efficiency of bulk and OM/IOM models is because of the difference in CO2 breakthrough time. Since CO2 EGR studies with bulk model can overestimate the effect, the OM/IOM model should be used when evaluating the potential for CO2 injection.