Carbon dioxide (CO2) capture from large stationary point sources and subsequent long-term geological storage inside deep rock formations provides a method for avoiding the emission of CO2 into the atmosphere. According to Benson and Cole, technical and economic assessments suggest that carbon capture and sequestration can account for up to 20% of CO2 emission reductions.  It is also believed that 99% or more of the injected CO2 will be retained underground for 1000 years assuming suitable site selection and monitoring practices. 
Bachu states that large sedimentary basins show good potential for geological storage due to their high porosity and permeability, and the presence of impermeable cap rock to trap it underground.  In North America alone, oil and gas reservoirs are estimated to have a capacity of about 80 gigatons of carbon (Gt C), saline aquifers between 900 and 3300 Gt C, and coal beds about 150 Gt C. This combined capacity corresponds to several hundreds of years of emissions.
The technology for injection of CO2 in deep geological formations already exists due to similar processes (such as well drilling, gas injection, seismic monitoring, and reservoir simulations for enhanced oil recovery (EOR)) already employed by the oil and gas industry. In the short run, utilizing the large capacity of oil and gas fields for geological storage cannot be realized unless enhanced oil recovery is implemented or operators declare such fields depleted. Therefore there is a timing issue matching carbon dioxide sources with storage sites.  Bachu also notes that the economic benefits of enhanced oil and gas recovery, as well as enhanced coal bed methane recovery makes them most likely to be implemented first, as compared to storage in saline aquifers which provide no economic incentive. While the capacity of oil and gas reservoirs is at least one order of magnitude smaller than that of saline aquifers, such EOR operations provide large scale pilot projects to lay the foundation for storage and monitoring technologies for use in non-EOR applications in the future.
At typical storage conditions at depths of 800 to 1000 m, CO2 exists as a supercritical fluid. This gives it liquid-like characteristics as well as a greater density for more efficient utilization of the storage volume as compared to gaseous CO2. However, due to the interactions between injected CO2 and in-situ fluids such as water and oil, its behavior is characterized by multi-phase flow. A consequence of this is that only about 30% of the pore volume is saturated with CO2 due to the flow dynamics and capillary pressure. 
CO2 is less dense than the water trapped in the original rock formation, causing it to become buoyant. Thus, an impermeable cap rock is necessary to trap the CO2 to prevent it from migrating upwards through porous rock out of the formation. This process is known as structural or straitigraphic trapping. Caprocks are usually low permeability shales, salt domes, or carbonate rocks. Structural trapping is especially important during the initial stages immediately following CO2 injection.
Hydrodynamic or residual-phase trapping occurs when CO2 becomes trapped in the pore spaces of the reservoir formation.  This mechanism is especially important in saline formations that do not possess caprocks.
Solubility trapping occurs when CO2 dissolves in the formation water to form carbonic acid, carbonate ions, and bicarbonate ions, lowering the pH from near neutral to 4. Carbon dioxide solubility increases with increasing pressure but decreases with increasing temperature and salinity. This has the beneficial effect of minimizing buoyant forces since the dissolved CO2 no longer exists as a separate phase, aiding in immobilization of the CO2 plume.
The dissolved CO2 can further react to dissolve primary minerals, precipitating thermodynamically stable secondary minerals such as carbonates. Such mineral trapping mechanisms are able to immobilize CO2 for geologically significant timeframes and have the greatest potential for long term sequestration.  However, the kinetics of mineral trapping are very slow and this trapping mechanism only becomes significant on the order of a thousand years or more. A paper by Krevor et al. cites ultramafic rocks as having great potential for mineral storage.  Mafic rocks contain a high concentration of reactive magnesium and iron but ultramafic (also ultrabasic) rocks have increased reactivity due to their lower silica contents of less than 45%. Ultramafic rocks undergo metamorphic hydration to produce serpentinites which are also suitable for mineral storage of CO2. Below are carbonation reactions for serpentine and olivine:
|Serpentine:||1/3 Mg3Si2O5(OH)4 + CO2 → MgCO3 + 2/3 SiO2 + 2/3 H2O|
|Olivine:||1/2 Mg2SiO4 + CO2 → MgCO3 + 1/2 SiO2|
It is important to be aware of some of the environmental consequences of geological storage. Dissolution of carbon dioxide can lead to the dissolution of minerals such as iron oxyhydroxides. This dissolves toxic metal compounds into the formation water and poses significant environmental risks if the water migrates to aquifers or groundwater meant for human consumption. The supercritical carbon dioxide is also a good organic solvent and may mobilize toxic organic compounds present in residual crude oil.  Therefore, in addition to suitable site selection, a robust regulatory and monitoring framework must be in place in order to safeguard the integrity of water bodies near the injection site.
The Sleipner project operated by Statoil is the first commercial application of CO2 storage in deep saline aquifers in the world, and it is also the largest carbon capture and storage project in Europe today. A paper by Torp and Gale summaries the findings of the Saline Aquifer CO2 Storage (SACS) Project, demonstrating that the successful CO2 injection has been achieved by monitoring the injected CO2 using seismic data to verify models and tools originally developed for the oil and gas industry.  The offshore Sleipner gas field produces natural gas with a CO2 concentration of 9% which is too high for pipeline transport due to corrosion issues. The CO2 is removed via amine scrubbing and injected into a brine-containing sand layer called the Utsira formation for geological storage 1000 m under the sea floor. CO2 injection commenced in 1996 with an injection rate of 1MT CO2 per year, and estimates indicate that it will store about 20 MT CO2 over the lifetime of the gas field's operations within 12 km of the injection site.
The Utsira formation is between 200 and 300 m thick, porosity estimates of the formation core range from 27% to 31% and it spans total area of 26,100 km2. The overlying caprock can be divided into the lower, the middle and the upper seal, and it has been empirically determined from caprock samples that they serve as an effective seal, with capillary leakage of CO2 unlikely to occur. The presence of thin shale layers in the formation allows for the lateral migration of CO2 beneath the intra-reservoir shales, increasing the effective reservoir capacity due to increased dissolution of CO2. In addition, the presence of a 50 m thick sand wedge near the base of the lower seal further increases the reservoir capacity as CO2 can migrate into the wedge in addition to being confined beneath the top of the formation. This also aids in reducing the migration distance of CO2, keeping the CO2 plume closer to the injection site.
Torp et al. highlight from geochemical experiments and modeling studies that limited reaction has occurred between CO2 and the rocks and sand in the Utsira formation. They conclude that mineral trapping will not contribute significantly towards trapping of the injected CO2. They instead point to solubility trapping as being significant, with reservoir simulations showing this mechanism accounting for about 18% of the lifetime trapped CO2, with a solubility of CO2 in brine of 53 kg/m3 at formation temperatures and pressures. However, this result is to be expected as mineral trapping only becomes significant on the timescale of thousands of years.
Torp also evaluates the long term fate of CO2 in the formation. While their model assumes an impermeable caprock which prevents buoyant migration of CO2 out of the formation, it does allow for molecular diffusion of CO2 through the overlying strata. The downward diffusion into the brine column activates convection currents within the column due to density changes, aiding in circulation of the CO2 and increasing the amount of CO2 dissolved in the brine column. Calculations on the upward migration through overlying shale layers reveal that CO2 will not reach the seafloor for hundreds of thousands of years, indicating the formation's long term storage ability.
The SACS project has successfully shown that conventional time lapse seismic data is suitable for the monitoring of CO2 injected into saline aquifers. Furthermore, this project has bolstered confidence in their monitoring ability as CO2 accumulations with a thickness as low as about one meter can be easily detected and monitored using seismic techniques. Torp concludes that after taking characteristics of the formation into account, namely its high permeability, large pore volume, and shallow dome-shaped caprocks, monitoring reservoir pressure is not a priority as only minor pressure buildups will occur over the lifetime of the storage reservoir. It is also pointed out that the cost of monitoring is very high at Sleipner by virtue of it being an offshore project, requiring significant rig rates and drilling time. This is in contrast to onshore projects which are generally less complex and easier to monitor.
The Sleipner Project is pioneering the first generation of carbon capture and sequestration using deep geological storage and plans to store about 20 MT CO2 over the lifetime of each respective project. To put this figure into context, a coal fired power plant emits about 8 MT CO2 annually, indicating that such carbon storage projects need to be scaled up by several orders of magnitude to truly be robust enough to tackle current emissions and past emissions. As pointed out earlier, adoption of geological storage in the short run will require economic incentives either in the form of carbon taxes (in the case of Sleipner) or increased revenue through EOR activities. It is encouraging to note that the technologies required for geological storage and monitoring already exist and are maturing and improving as a direct consequence of projects such as Sleipner. Public perception is crucial in determining the successful implementation of geological storage especially now in this technology's infancy stage. As such, a robust regulatory and monitoring framework must be in place to safeguard the integrity of water bodies and the environment near injection sites.
© Alexander McCurdy. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 S. M. Benson and D. R. Cole, "Sequestration in Deep Sedimentary Formations," Elements 4, 325 (2008).
 S. Benson et al., "Chapter 5: Underground Geological Storage" in IPCC Special Report on Carbon Dioxide Capture and Storage (Cambridge, 2005) [Available for download from the Intergovernmental Panel on Climate Change].
 S. Bachu, "Screening and Ranking of Sedimentary Basins for Sequestration of CO2 in Geological Media in Response to Climate Change," Environmental Geology 44, 277 (2003).
 B. P. McGrail et al., "Potential for Carbon Dioxide Sequestration in Flood Basalts," J. Geophys. Res. Solid Earth 111, B12201 (2006).
 S. C. Krevor et al., "Mapping the Mineral Resource Base for Mineral Carbon-Dioxide Sequestration in the Conterminous United States," U.S. Geological Survey, Digital Data Series 414, January 2009.
 T. A. Torp and J. Gale, "Demonstrating Storage of CO2 in Geological Reservoirs: The Sleipner and SACS Projects," in Greenhouse Gas Control Technologies - 6th International Conference, ed. by J. Gale and Y. Kaya (Pergamon, 2003), pp. 311-316.