Terrestrial Saline Aquifer Sequestration
Terrestrial saline aquifers are saline aquifers located on land; they are often made up of sedimentary rock and may be covered by a layer of relatively impermeable rock, called the caprock. Because the CO2 is less dense than the brine in terrestrial injection, some CO2 will rise up and become trapped by the cap-rock above. In addition, some CO2 will interact with minerals within the aquifer and become trapped.
Injection of CO2 should be confined to depths of 500 m, the shallowest depth at which CO2 is super-critical, to 3000m, the depth below which effective storage decreases and price per unit of CO2 injected increases (Eccles et al, 2009). Efficiency of CO2 storage increases with density; density generally increases with depth, but temperatures rise as depth increases and pore space decreases as the rock becomes more compacted, resulting in decreased storage efficiency past certain depths (Intergovernmental Panel on Climate Change, 2008).
There are few national regulations specifically dealing with CO2 storage, but regulations dealing with oil and gas, groundwater, and the underground injection of fluids can, in many cases, be readily adapted and/or adopted. Policy makers must still determine the classification of CO2 (as a waste, pollutant, or otherwise), different levels of jurisdiction (international, national, subnational), and the ownership of pore space (Intergovernmental Panel on Climate Change, 2008; Bachu, 2007).
- Underground trapping of CO2 occurs naturally. There are 200Mt of CO2 that have been trapped in the Pisgah Anticline in Mississippi for perhaps 65 million years. With the proper care taken in site selection and management, there is a greater than 90% chance that over 99% of the injected CO2 will stay trapped for the first 100 years, and a chance greater than 66% that it will stay trapped for the first 1000 years. As of 2005, there has been no evidence of terrestrial impact from then existing CO2 storage projects (Intergovernmental Panel on Climate Change, 2008).
- Terrestrial saline aquifers have few proposed applications besides CO2 storage, because drinking water cannot be extracted from such aquifers without great expense (IEA 2008).
- Hydrocarbon pockets are often formed in association with saline aquifers, so known hydrocarbon pockets in sedimentary basins are often found near saline aquifers. This association can be used to find saline aquifers, and also serves, as in the cases of Sleipner West and In Salah, to provide a nearby sink for some CO2 sources, helping to minimize transportation costs (Meng et al, 2006).
- Data from CO2 injection at pilot projects (Frio, Ketzin, Nagaoka) and existing commercial operations (Sleipner, Snøhvit, In Salah) shows that CO2 geological storage in saline aquifers is currently technologically feasible (Michael et al, 2009).
- CO2 may spread after injection; some models suggest that a fracture 8 km from the injection site may leak CO2 after 250 years (IEA, 2008).
- Future drilling may cause fractures and leakage. Preventing leakage may involve replacing cap-rocks with engineered materials, like cement and well-casing, which may have unfavorable properties. Of the nine documented cases of significant leakages from the 470 natural gas storage facilities in the US, with a capacity-weighted median age of 25 years, five were related to the integrity of the wells. Although CO2 is more corrosive than methane to metal parts, natural gas systems are more susceptible to caprock leakages because of rapid changes in pressure, so this experience can be taken to show minimum CO2 storage performance (Intergovernmental Panel on Climate Change, 2008). Abandoned oil wells are sealed, but if the well penetrates a storage zone, the stored CO2 may react with the cement plug and weaken it, eventually causing a leak (IEA, 2008). Though CO2 resistant cement exists it does add 25% to total cementing costs (IFC International, 2008).
- Vertical fractures may occur, allowing CO2 to leak into other aquifers or to seep into hydrocarbon resources and the soil, potentially harming plants and sub-soil ecosystems. Also, SO2 or O2 contaminants may leech heavy metals from the surrounding rock matrix, posing an environmental risk. Contamination of surrounding areas may also occur if brine is displaced by injected CO2 (Intergovernmental Panel on Climate Change, 2008).
- Not all saline aquifers are suitable for CO2 sequestration. Some may not have sufficiently impermeable cap-rocks above them, or have too many previous drilling sites in the vicinity to ensure secure storage. Crystalline, metamorphic, and volcanic rocks are often too impermeable to hold much CO2 and too fractured to securely hold CO2. Those found in mountain-forming areas would also be more susceptible to leaks (IEA, 2008) While basins formed in mid-continent locations are likely to be stable and structurally favorable and those found behind mountains formed by plate collisions are likely to have high storage potential, areas closest to the actual mountain-forming faults should be used only with caution (Intergovernmental Panel on Climate Change, 2008).
- Any impurities in injected CO2 may negatively impact the compressibility of the CO2 stream, thus decreasing the storage capacity (Intergovernmental Panel on Climate Change, 2008).
- Pilot sites generally have comprehensive monitoring programs, but injection rates per volumes are low compared to potential commercial projects, while current commercial projects inject larger volumes of CO2, but monitoring programs are often limited (In Salah, Alberta acid-gas), or reservoir properties are “unrepresentatively good” (relatively high permeability at Sleipner). Also, better technology must be developed for seismic imaging of CO2 and detecting leaks, and more cost-effective methods must be found and tested (Michael et al, 2009)
- Current CO2 storage capacity estimates are imperfect (Intergovernmental Panel on Climate Change, 2008).
- Currently, more investigation is needed of risks associated with possible leakage, impacts on underground microbes and human health need further investigation, and effective monitoring strategies (Intergovernmental Panel on Climate Change, 2008).
- The risk of CO2 leakage can be reduced by producing brine from the aquifer and re-injecting it at some distance from the CO2 injection wells. This technique can increase the percent of CO2 dissolved in the aquifer in 200 years from about 8% to up to 50%. For an energy cost of <20% of that needed for CO2 compression, brine re-injection can increase dissolution 5-fold, reducing environmental risk, since free-phase CO2 poses the greatest risk. Running the pumps for 200 years, at a 2.4Mt/yr brine production and reinjection rate, would incur energy costs of about $15 million, at $0.05/kWh. Additional costs Increasing brine injection rate increases the pressure in the aquifer and does not always increase dissolution rate (Hassanzadeh, 2009)
- Stimulation techniques, such as hydraulic fracturing and induced micro-seismicity may improve injection capacity, although care should be taken not to compromise the cap-rock in the process (Ghaderi, Keith, & Leonenko, 2009).
Estimates of CO2 storage potentials in deep saline aquifers (including offshore) (IEA, 2008):
|Alberta (Canada)||1000 - 4000|
|USA||900 - 3400|
|Europe||30 - 577|
|Worldwide||2000 - 20,000|
(Bradshaw et al, 2007)
(Sharma et al, 2009)
- According to the analysis of Eccles, et al, optimal storage depth is about 1600 m, where the effective storage capacity is 0.24 to 0.31 tonnes of CO2 per m3 of sandstone reservoir rock. At this depth, a minimum of about 0.7 km3 would be needed to store 7389 tonnes of CO2 per day, which is the amount produced by the average coal-fired power plant. Such a site could be filled at such a rate for 20 years with a capacity factor of 80%. Numerical simulations and pilot projects have shown that the injected CO2 bypasses and does not fill up most of the available space. For example, numerical simulations of the Frio injection project have assumed that between 5% and 30% of pore space will be occupied. If correct, the volume necessary to sequester CO2 will be 3-20 times higher than the minimum predicted by [their] model" (Eccles et al, 2009).
- Ghaderi et al did simulations on the effect of the number of wells on the amount of CO2 injected after 50 years. In their graphs, average injection rates of at least 4MtCO2/yr could be achieved in aquifers with more unfavorable properties, with greater rates achieved with more injection wells. For the simulation specifically for the Nisku formation in Canada, lower-limit properties gave an average yearly injection rate of about 4MtCO2, given about 10 injection wells (Ghaderi et al, 2009)
- Layer thickness and permeability can increase cost by a factor of 50 when these two factors are decreased to their lowest limit. Storage costs are pennies per ton with maximum permeability. Increasing layer thickness has diminishing returns for depths below about 1000 m, due to the nonlinear increase in drilling costs. Reservoir depth only affects cost by a factor of up to two. Because of differences in thickness and permeability, even within a single aquifer, injection costs per site may vary from very inexpensive to extremely expensive (Eccles et al, 2009).
- Costs, including monitoring, are in the range of $0.6-$8.3/tCO2. An "enhanced" monitoring package, including periodic seismic surveys, microseismicity, wellhead pressure and injection-rate monitoring, periodic well logging, surface CO2 flux monitoring and other advanced technologies, would cost $0.069-$0.085/tCO2 (with a 10% discount rate for the first 30 years, 1% afterward and $0.27-$0.30/tCO2 not discounted); this estimate assumes monitoring during a 30-year injection period and for 50 years after project closure (Intergovernmental Panel on Climate Change, 2008).
- With more research, the total cost of storage may decrease (Intergovernmental Panel on Climate Change, 2008).
In 2005, the IPCC gave the following estimates for storage costs in saline aquifers, though prices have likely increased since then (IEA, 2008):
|Location||cost ($ per tonne CO2)|
|USA onshore||0.40 - 4.50|
|Europe onshore||1.90 - 6.20|
|Europe/North Sea||4.70 - 12|
|Australia onshore||0.20 - 5.10|
|Australia offshore||0.50 - 30|
- According to the model used by Eccles, et al, injecting CO2 approximately 1500m deep, a depth at which storage capacity is nearly maximized, at an injection rate of 80-100 tonnes per day costs about $2 - $4 per metric tonne (the model takes into consideration maximum pressure before fractures occur when calculating injection rate), (Eccles et al, 2009).
- One model for injection rates predicts minima for storage costs in a typical basin in the range of $2-$7/tonne CO2 (2005 U.S. dollars) depending on depth and basin characteristics in their base-case scenario. Because the properties of natural reservoirs in the United States vary substantially, storage costs could be lower or higher by orders of magnitude (Eccles et al, 2009).
- The cost of geological carbon sequestration is approximately $3.50/tonne of stored CO2 at the optimal depth (for the model results that most closely match existing pilot projects). The model predicts $2-7/tonne for the range of depth and basin characteristics for the base-case permeability and layer thickness (costs continue to increase at depths greater than 3000 m) with an optimum depth near 1300 m. These costs were calculated assuming the maximum possible injection rate and storage, meaning that these are minimum theoretical costs. In the graphs below, high and low end members correspond to unfavorable and favorable reservoir characteristics, respectively. Drilling costs are taken into account, but not the cost for any additional compression equipment (Eccles et al, 2009).
(Eccles et al, 2009) ("ton" refers to metric tonnes)
- At Sleipner West, preparation works cost about $1.9 million, the compressor train cost about $79 million, the injection well cost about $15 million, and annual operational costs total about $7 million (Torp et al, 2005).
- Meng et al proposed demonstration project costs for four pairs of CO2 sources and sinks for China's nine largest pure CO2 sources among ammonia plants. Sinks were saline aquifers thought to coexist with known hydrocarbon pockets in sedimentary basins. Capital investments ranged from $56 million to $71 million per project, specific costs ranged from $15 to $21/tCO2, and total project costs ranged from $13 to $18 million per year. The majority of the cost (up to 90%) was in compression and transportation (Meng et al, 2006).
- The In Salah project in Algeria, a commercial-scale project injecting about 1MtCO2/yr into a saline aquifer, cost about $100 million and injects CO2 at a cost of about $6/tCO2 (Haddadji, 2006).