OR/13/004 The role of interfaces in CO2 migration

Rochelle, C A, Purser, G, Milodowski, A E, Noy, D J, Wagner, D, Butcher, A, and Harrington, J F. 2013. CO2 migration and reaction in cementitious repositories: A summary of work conducted as part of the FORGE project. British Geological Survey Internal Report, OR/13/004.

Within a multi-barrier concept repository there will be many different interfaces between the repository components. It was not the aim of this study to investigate these interfaces, so only a few comments can be made about these. That said, our tests did generate interfaces between regions of partially-carbonated and fully-carbonation cements as a result of CO₂-cement chemical reactions. This resulted in shrinkage cracks and reaction fronts, which potentially could act as foci for fluid flow. Shrinkage could have resulted from water loss during carbonation, or actual solids volume decreases. There are indications that this localized zone of shrinkage and porosity enhancement migrates over time, with secondary carbonate precipitation progressively filling porosity after it is created. This secondary carbonate precipitation is also controlled by very local back-diffusion of calcium ions creating very narrow reaction/precipitation fronts. Thus a ‘dynamic’ zone of increased permeability appears to move through the sample, and is longer- lasting than one specific shrinkage crack.

Laboratory experimental studies and observations of boreholes as part of CO₂ storage studies have shown that imperfections in the sealing of steel/cement and cement/rock interfaces can have an important control on gas migration (e.g. Carey et al., 2007: Rochelle et al., 2009). For example, a poor steel/cement bond or an unsealed engineered damage zone (EDZ) at the cement/rock interface can allow CO₂ migration. Poor seals might also result from poor well management, or temperature cycling during periods of alternating production, injection, or shut-in. In the case of the study of borehole cement recovered from the SACROC Unit (Carey et al., 2007), zones of cement carbonation extended several metres up the well (i.e. into the overlying caprock) along the cement-rock interface (Figure 14). It is possible therefore, that if there was significant flow of acidic, CO₂-rich water along this interface, then there is potential to enhance dissolution and magnify interface permeability. On the other hand, significant dissolution will only occur if there is sufficient flow to remove dissolved components. In conditions of low or no flow, then reactions may occur initially, but once they approach local equilibrium will do very little thereafter. Indeed, if carbonation reactions resulted in a reduction in porosity/permeability and reduced potential for fluid migration, then they might improve the sealing of the cement. Cement carbonation therefore has the potential to be either beneficial or deleterious depending upon the specific environment and degree of reaction. Given the smaller amounts of CO₂ in the repository environment compared to CO₂ storage, and the larger (and potentially excess) amounts of cement, there would appear to be a relatively low risk that cement carbonation would cause major problems in the repository environment. In terms of interfaces, CO₂ storage studies need to consider just those between steel/cement and cement/rock (e.g. Carey et al., 2007; Rochelle et al., 2009). Also, these interfaces may not be clean as they may be variably covered with drilling mud. Whilst residues from drilling fluids will not be an issue in a repository setting, it will however be important to consider cement/cement interfaces. These could occur, for example, between cement grout/backfill and concrete boxes containing steel drums of waste, between layers of grout/backfill if emplacement was staged, or between cavern structural cement and grout/backfill. There could also be chemical or mineralogical differences across the cement/cement interfaces — such as when different formulations are used, between different ages of cement (hydrated phases in cement becoming more ordered over time), or if some of the cement were exposed to higher than ambient temperatures due to the presence of heat-emitting waste.

There are other reactions that will also occur along interfaces. Carbonation reactions generate water as hydrated calcium silicate hydrate phases convert to carbonate minerals. If this water were to migrate towards grains of non-hydrated cement clinker, then new cement minerals may form (e.g. calcium silicate hydrates), and these may help lower local permeability. There are indications from a limited number of laboratory experiments undertaken as part of CO₂ storage studies (see Rochelle et al., 2009) that sealing along a cement-steel interface might actually improve with ingress of limited amounts of water or CO₂, as this may result in new cement minerals growing and an improvement in sealing.

The behaviour of cement interfaces during carbonation is thus complex. However, one clear conclusion coming out of several CO₂ storage-related cement carbonation studies was that a high quality initial bond/seal was the best way to minimise gas migration. A good seal may actually improve with limited carbonation, but a ‘leaky’ seal is likely to get worse. Similarly, well-sealed interfaces will be needed in a repository setting.

File:OR13004fig14.jpg

Figure 14     Image of samples of borehole cement recovered after 30 years in a CO₂-rich environment (modified after Carey et al., 2007). Note that carbonation was focussed along the cement/steel and especially the cement/shale interfaces.