The Function of Gas-Water Relative Permeability Hysteresis in the Sequestration of Carbon Dioxide in Saline Formations

Abstract

Abstract The injection of carbon dioxide (CO2) into saline formations for the purpose of limiting greenhouse gas emissions has been proposed as an alternative to atmospheric venting of carbon dioxide. To date, research of the subsurface interactions of injected CO2 in saline formations have focused on two main areas: dissolution of CO2 into saline formation waters and the geochemical/mineralisation effects of the injected CO2 on the formation rock matrix. Both of these topics investigate forms of potential permanent sequestration or trapping of CO2. It is proposed that trapping of CO2 as a residual phase due to imbibition of water is another form of permanent sequestration of CO2. This paper will outline the role that gas-water relative permeability hysteresis can have towards the success of storage for a geological CO2 sequestration project. This process is similar to the experience of aquifer encroachment on depleted gas reservoirs where water traps gas along its path. During the injection phase of a geological CO2 sequestration project, the movement of CO2 is dominated by a drainage relative permeability state, as CO2 displaces brine, the wetting phase in the formation. After injection, the governing force on the injected fluid is gravity. CO2, at most reservoir conditions, is more buoyant than brine and thus injected CO2 will tend to migrate up-dip along the top seal. The migration of the CO2 plume, over hundreds of years, is dominated by both drainage and imbibition relative permeability effects. At the head of the migrating plume, drainage relative permeability is prominent as water drains away from the migrating plume. At the tail of the migrating plume, imbibition relative permeability dominates as water imbibes behind the migrating plume trapping CO2 as a residual phase. Thus the amount of mobile CO2 diminishes over time, as more of the plume dissolves and is trapped as a residual phase. A reservoir simulation study was conducted to investigate gas trapping in a geologic sequestration project. Many subsurface parameters were varied, including the effects of gas-water relative permeability hysteresis. The results of these parametric changes were monitored with regard to the effects on migration, pressure build up at the geological seal and the fraction of CO2 in the formation that is either mobile, dissolved or trapped as a residual phase. It is demonstrated that the factors determining the long term outcome and hence success of a sequestration project includes: the permeability field, the degree of dissolution and the degree of gas trapping as a residual phase. Trapping CO2 as a residual phase is shown to have a constructive impact on the success of a geologic sequestration project. Introduction The atmospheric emission of greenhouse gases, such as carbon dioxide (CO2), has been linked to alterations in the earth's climate1. Geologic sequestration of CO2 by injection into saline formations has been suggested as a viable alternative to venting gaseous CO2 into the atmosphere2,3. Much progress has been made into understanding the subsurface interactions and behaviour of injected carbon dioxide into saline formations. To date, research of the subsurface interactions of injected CO2 have discussed two mechanisms of permenant storage of CO2: dissolution of CO2 into saline formation waters and the geochemical/mineralisation effects of the injected CO2 on the formation rock matrix2. There remains an additional mechanism that is able to permenantly trap CO2 in a saline formation which is worthy of further investigation, that is the application of gas-water relative permeability hysteresis in trapping injected CO2 as residual phase. This phenomenon is well known in the petroleum industry during the application of CO2 flooding and WAG (Water-Alternate-Gas) injection strategies in enhanced oil recovery projects. This mechanism has had demonstrated impacts on the implementation and outcomes of EOR projects.