地化学・貯留層シミュレーションによる二酸化炭素の地化学トラッピングの検討：東京湾岸モデル

Abstract

Two dimensional numerical simulation was performed using reactive geochemical transport code TOUGHREACT based on the geology of the Tokyo Bay area to understand long-term geochemical evolution of aqueous and mineral phases after CO2 injection and to elucidate key issues affecting CO2 mineral trapping. The injected CO2 migrates upward in the reservoir (the Umegase Formation) towards the cap rock (the Kokumoto Formation) driven by buoyant force. In the CO2 plume, water pH drops from 7.8 (initial) to 4.9 by dissolution and dissociation of CO2. The acidified water containing CO2 and related species induces mineral dissolution and promotes complexing of dissolved ions, which increase the water density and cause its downward movement to the bottom of the reservoir. This process gives rise to convective mixing which further promotes CO2 solubility trapping. In the peripheral zone of the CO2 plume, the acidified water comes in contact with intact reservoir water, giving rise to the precipitation of calcite (and siderite, in lesser amount). The precipitation of calcite and other carbonates might form a low permeability barrier surrounding the CO2 plume, and is expected to increase the storage security with time. Among newly formed carbonates, calcite becomes the most dominant in a long period after injection cease. Dawsonite, which is occasionally considered to be important in trapping CO2, is found to have limited stability that requires high CO2 partial pressure at the conditions of saline aquifer storage. Our study also finds that the formation of dawsonite is in a close relation to the dissolution of plagioclase, and that its precipitation is strongly suppressed by the formation of kaolinite. After reviewing rates of dissolution of carbonates and feldspars, we conclude that the precipitation of dawsonite is controlled by plagioclase dissolution kinetics which determines the supply of Al as a major constituent of dawsonite. Plagioclase is a dominant clastic mineral in sandstones in geologically young formations of the Japanese Islands. Its dissolution could be a key issue in geochemical modeling of CO2 aquifer storage.