Carbon Order

Atmospheric CO2 Reduction

Over the past decade, human activities have produced approximately 34 GtCO2/yr annually with about 16 GtCO2/yr, or about 2 ppm/yr, accumulating in the atmosphere (more recent estimates of annual emissions sources are ~39 GtCO2/yr: 36 GtCO2 from fossil fuel combustion and cement production and ~3 GtCO2 from land use changes [Global Carbon Project, 2014]). Note that less than half of current and historical anthropogenic CO2 emissions remain in the atmosphere; the remainder (18 GtCO2/yr) has been taken up by the ocean and the terrestrial biosphere. This existing uptake and removal of CO2 from air, natural “carbon dioxide removal” (CDR), already moderates the impacts of human emissions on atmospheric CO2 levels and global climate. Indeed this uptake is seasonally so great that atmospheric CO2 concentrations intra-annually decline. Nevertheless, substantially increasing existing CDR by natural or unnatural means such that the average annual growth rate of atmospheric CO2 is reduced or reversed poses a significant challenge. One reason is that if enough CO2 were removed from the atmosphere to cause a decline in overall atmospheric concentrations, CO2 would “outgas” from the ocean into the air and the terrestrial land sink would be less effective.1 Over a period of several decades, this would replace up to half of the CO2 that had been removed by CDR (IPCC, 2013a). Reducing CO2 concentration by 1 ppm/yr would require removing and sequestering CO2 at a rate of about 18 GtCO2/yr; reducing CO2 concentrations by 100 ppm would require removing and sequestering a total of about 1,800 GtCO2, or roughly the same amount of CO2 as was added to the atmosphere from 1750 to 2000.

Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration, National Research Council of the National Academies 2015

Simplified schematic of the global carbon cycle

A simplified schematic of the global carbon cycle.  Numbers present reservoir mass, also called 'carbon stocks' in PgC (1PgC = 1015 gC = 3.67 GtCO2) an annual carbon exchange fluxes (in PgC yr-1).  Black numbers and arrows indicate reservoir mass and exchange fluxes estimated for the time prior to the Industrial Era, about 1750.  Fossil fuel reserves are from GEA (2006) and are consistent with numbers used by IPCC Working Group III for future scenarios.  Red arrows and numbers indicate annual anthropogenic fluxes averaged over the 2000-2009 time period.  These fluxes are a perturbation of the carbon cycle during Industrial Era  post-1750.  Red numbers in the reservoirs denote cumulative changes of anthropogenic carbon over the Industrial Period (1750-2011).  By convention, a positive cumulative change means that a reservoir has gained carbon since 1750.  Uncertainties are reported as 90% confidence intervals. For more details see IPCC (2013a)

SOURCE: Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration, National Research Council of the National Academies 2015

Figure 1a: Annual anthropogenic CO2 emissions.

Figure 1(b)  Warming vs cumulative CO2 emissions

Figure 1:  (a) Annual anthropogenic CO2 emissions; (b) Warming versus cumulative CO2 emissions

SOURCE: Wg3 AR5 Summary for Policy Makers, IPCC March 2014

Figure 2: a) decay of instantaneous (pulse) injection and extraction of atmopheric CO2; (b) atmospheric CO2 if fossil fuel emissions terminated at end of 2011, 2030, 2050

Figure 2: a) Decay of instantaneous (pulse) injection and extraction of atmopheric CO2; (b) atmospheric CO2 if fossil fuel emissions terminated at end of 2011, 2030, 2050

SOURCE: The Case for Young People and Nature:  A Path to a Natural, Prosperous Future, James Hansen et al

Figure 3

Figure 3: Geological storage options for CO2

SOURCE: Addressing the grand challenge of atmospheric CO2: geologic sequestration vs biological recycling, Ben J Stuart, Journal of Biological Engineering 2011 S:14 doi:10.1186/1754-1611-5-14

Figure 4

Figure 4: A schematic of CO2 injection

SOURCE: Safe Storage of CO2 in Deep Saline Aquifers, Robert G Bruent, Andrew J Guswa, Michael A Celia and Catherine A Peters.  American Chemical Society. Environmental Science and Technology 1st June 2002

Figure 5

Figure 5

SOURCE: IPCC Special Report on COCapture and Storage, IPCC 2005

Figure 6:

Figure 6

SOURCE: Feasibility of Injecting Large Volumes of COinto Aquifers. Seyyed M Ghaderi, David W Keith, Yuri Leonenko.  Energy Procedia 1, (2009) 3113-3120 Science Direct

Figure 7

Figure 7

SOURCE: A.2.1 World Scale (Lithographical map of the world, modified from Commission for the Geological Map of the World,

Figure 8

Figure 8

SOURCE: Storage Prospectivity, 

Figure 9


Figure 9


Figure 10


Figure 10SOURCE: Researchers examine carbon capture and storage to combat global warming, Annie Jia, Stanford News 13.06.07

Figure 11

Figure 11

SOURCE: Sliepner

Figure 12

Figure 12 Carbon Capture & Storage - Potential Sites in North Sea Basin

SOURCE: The glaciation of the North Sea Basin and its implications for carbon capture and storage sites.  Presentation by Tom Bradwell at UKCCSRC Glacistore meeting 27.02.15


Figure 13


Figure 13


SOURCE: IPCC Special Report on CO2 Capture and Storage, IPCC 2005


Figure 14: Alternative schemes for ocean storage


Figure 14: Alternative schemes for ocean storage


FigureSOURCE: Ocean Storage of CO2: Pipelines, Risers and Seabed containment (draft). Andrew Palmer, David Keith and Richard Doctor


Figure 15: Containment schemes


Figure 15: Containment schemes


SOURCE: Ocean Storage of CO2: Pipelines, Risers and Seabed containment (draft). Andrew Palmer, David Keith and Richard Doctor


Figure 16: Cross-section of cylindrical containment membrane


Figure 16: Cross section of cylindrical containment membrane


SOURCE: Ocean Storage of CO2: Pipelines, Risers and Seabed containment (draft). Andrew Palmer, David Keith and Richard Doctor


Figure 17


Figure 17: Four phases of a CO2 geological sequestration project


SOURCE: Regulating the Geological Sequestration of CO2, Elizabeth Wilson et al, Environmental Science and Technology, April 15 2008


Figure 18: Oman - 70,000km3 of 30% olivine


Figure 18: Oman  - 70,000km3 of 30% olivine


Figure 19

Figure 19

SOURCE: A review of mineral carbonisation technologies to sequester CO2. A Sanna, M Uibu, G Caramana, R Kuusik, and MM Maroto-Valer. Chem. Soc. Rev 2014 43, 8049

Figure 20

Figure 20: Sites investigated for potential deep-sea carbon sequestration

SOURCE: A global assessment of deep-sea basalt sites for carbon sequestration.  Dave Golberg and Angela Slagle, Lamont-Dogherty Earth Observatory. Energy Procedia 2008

Table 1

Table 1

SOURCE: A review of mineral carbonisation technologies to sequester CO2. A Sanna, M Uibu, G Caramana, R Kuusik, and MM Maroto-Valer. Chem. Soc. Rev 2014 43, 8049


table 2


Table 2
SOURCE: A global assessment of deep-sea basalt sites for carbon sequestration.  Dave Golberg and Angela Slagle, Lamont-Dogherty Earth Observatory. Energy Procedia 2008


Geological Sequestration of Carbon Dioxide


Total capacity estimates show that geological sequestration has the potential to sequester large amounts of CO2. In Global Energy Assessment: Toward a Sustainable Future, Benson et al. (2012) estimate that global sequestration capacities for depleted oil and gas reservoirs are ~1,000 GtCO2 for coal beds up to 200 GtCO2 and sequestration in saline aquifers is highly variable between 4,000 and 23,000 GtCO2 (Benson et al., 2012). A recent study by Dooley (2013) provides updated geologic sequestration capacities, with a global “theoretical” capacity of 35,300 GtCO2, an “effective” capacity of 13,500 GtCO2, and a “practical” capacity of 3,900 GtCO2. The IPCC (2005, 2011a) estimates a minimum sequestration capacity in geologic formations of 550 GtC (~2000 GtCO2), with the potential to be significantly larger (i.e., thousands of gigatonnes of carbon), due to the uncertainty associated with saline aquifers. In 2012, the U.S. Geological Survey (USGS) identified technically accessible sequestration resources totaling 3,000 GtCO2 in 36 geological formations in the United States (USGS Geologic Carbon Dioxide Storage Resources Assessment Team, 2013). Figure 20 shows the estimated CO2 sequestration potential of saline aquifers, depleted oil and gas, and coal-bed reservoirs in North America.


Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration, National Research Council of the National Academies 2015


US CO2 sequestration capacity estimates for various geological reservoirs


US CO2 sequestration capacity estimates for various geological reservoirs.  Saline acquifers have the highest potential for CO2 sequestration, followed by depleted coal beds and oil and gas fields.

Current emission trajectories will result in atmospheric CO2 levels of potentially 650ppm by 2100, with worst-case IPCC projections (RCP 8.5, see figure 1) in the order of 1300 ppm (1). Paleoclimate records indicate that CO2 levels of 400 ppm are associated with polar ice melt sufficient to cause a sea level rise of 20-25 m (2), a process which all the evidence suggests is already underway (3), and may reach several metres this century (4). Large-scale sequestration will therefore be necessary to reduce atmospheric CO2 to a maximum of 350 ppm (5). Re-emission from natural carbon sinks increases a 300ppm requirement to around 500 ppm (6) (see Figure 2), representing circa 4 trillion tonnes CO2.  Over 100 years, this equates to 40 Gt CO2/year – roughly equivalent to current global anthropogenic CO2e output, which of course, also needs to be decarbonised over approximately a third of this timeframe.

Underground accumulation of carbon dioxide is a widespread geological phenomenon, with natural trapping of CO2 in underground reservoirs. Information and experience gained from a large number of existing enhanced oil recovery (EOR) and acid gas projects, as well as from the Sleipner, In Salah and Weyburn projects, indicate that it is feasible to store CO2 in geological formations as a CO2 mitigation option. Industrial analogues, including underground natural gas storage projects around the world, provide additional confidence that CO2 can be safely injected and stored at well-characterised and properly managed sites, with 99% or more of the injected CO2 being retained for 1000 years. (7, 8).

CO2 sequestration can be achieved through either burial or mineralization. Although the former includes various proposals for injection into deep oceans, depleted oil reserves or unmineable coal seams, saline aquifers offer the greatest potential in terms of storage capacity, locational advantages and state-of-the-art technology. Theoretical capacity is vast: hundreds to thousands of years’ of anthropogenic carbon emissions for saline aquifers, and effectively limitless in the case of mineralization.


Geologic sequestration is accomplished by injecting CO2 at depths greater than 1 km into porous sedimentary formations using well-established drilling and injection technologies derived from the oil and gas industry (see figures 3-7).

At ambient ground-surface temperature (25°C) and pressure (0.1 MPa), COis a gas with a density of 1.8 kg/m3 (the density of air under these conditions is 1.2 kg/m3). Assuming a geothermal gradient of 30 °C/km and a pressure gradient of 10.5 MPa/km, CO2 can be stored as a supercritical fluid at injection depths greater than ~800 m, as shown in Figure  3.  The density of supercritical CO2, ~260 kg/m3 at 800 m depth, permits far greater quantities of CO2 to be stored per unit volume than as a gas at shallower depths (9).

At 800-m depth, water with 15% total dissolved solids by mass has a density of ~1100 kg/m3 (4, 13, 16, 33). This density difference generates buoyancy forces that drive injected CO2 upward (10).

Storage in aquifers may occur via several mechanisms: injecting supercritical CO2 into formations with overlying layers of low permeability strata (caprock or stratigraphic trapping); retention as an immobile phase trapped in the pore spaces of the storage formation (residual phase saturation); dissolution in the formation fluids (solubility trapping); and through reaction with most minerals to precipitate carbonates (mineral trapping). Low groundwater flow velocities, often in the order of 1–10 cm/yr, limit lateral movement of dissolved CO2 in these systems (hydrodynamic trapping) (11).

CO2 dissolved in formation waters is not subject to upward buoyant migration. CO2 solubility decreases with increasing temperature and salinity, and increases with increasing pressure. At ground-surface conditions, the solubility of CO2 in pure water is ~1.7 kg CO2/m3. At the higher temperature and pressure conditions at 800-m depth, the solubility of CO2 in formation waters with 15% total dissolved solids by mass is ~35 kg/m3 (12).



There are many sedimentary basins around the world variously suited for CO2 storage (figures 8 & 9). Such basins frequently have sedimentary formations extending to a depth of greater than 2 km and are composed of horizontally stratified porous rocks with mean porosities often greater than 10% (13). Key criteria are: capacity, confinement effectiveness and geological stability. Suitability in part depends on their location on the continental plate – mid-continent or near the edge of stable plates are ideal because of their stability and structure: such basins are found within most continents and around the Atlantic, Arctic and Indian Oceans. Storage potential is similarly good in basins found behind mountains formed by plate collision, such as the Rocky Mountain, Appalachian and Andean basins in the Americas; the Alps, Carpathians and west of the Urals in Europe; and those to the south of the Zagros and Himalayas in Asia. Basins in tectonically active areas need to be selected carefully.

The efficiency of CO2 storage in geological media, defined as the amount of CO2 per unit volume, increases with CO2 density, as does storage safety, due to reduced upward migration. ‘Cold’ sedimentary basins, characterized by low temperature gradients, are more favourable for CO2 storage because CO2 attains higher density at shallower depths (700-1000m) than ‘warm’ basins where dense fluid conditions are reached at greater depths (1000-1500m). Adequate porosity and thickness (for storage capacity) and permeability (for injection) are critical; porosity usually decreases with depth because of compaction and cementation which reduces capacity and efficiency. The storage formation should be capped by extensive confining units, such as shale, salt and anhydrite beds to minimize leakage into overlying rock formations and ultimately into the atmosphere.


Depleted oil and gas reservoirs are prime candidates for CO2 storage, for several reasons: first, confinement integrity has already been demonstrated, in some cases over millions of years; secondly, the geological structure and physical properties have been extensively studied and characterized; thirdly, computer models have been developed to predict movement, displacement behaviour and trapping of hydrocarbons; and finally, some of the existing infrastructure and wells may be used for handling CO2 storage operations.

For operational wells, EOR can enhance oil extraction by 7-23% (average, 13.2%) compared with 5-40% for conventional primary production, and an additional 10-20% for secondary recovery using water flooding. For enhanced CO2 storage in EOR operations, oil reservoirs may need to meet additional criteria than those applicable solely to EOR.

Although up to 95% of gas can be extracted, CO2 could potentially be used to enhance gas recovery by repressurizing the reservoir.


Saline formations are deep sedimentary rocks saturated with formation waters or brines containing high quantities of dissolved salts. Although widespread and containing vast quantities of water, they are unsuitable for agriculture or human consumption. The best available example is the Sleipner project in the North Sea which currently sequesters around 1 Mt CO2/year (see below).



The worldwide CO2 storage capacity of deep saline aquifers has been estimated to range from 100 to 10,000 GtCO2, based on analyses by Koide et al, Hendriks and Blok, and other researchers. These analyses differ in their assumptions about volumes of sedimentary basins, aquifer characteristics, CO2 storage density, and technological and economic constraints.  Capacity estimates fall into two categories: the first assumes CO2 remains as a separate fluid phase, while the second assumes all CO2 dissolves in the brine (14).

Initial simulation studies have shown that a relatively small fraction of an aquifer will be filled with separate-phase CO2 because of hydrodynamic and buoyancy effects (22). If it is assumed that all of the CO2 is dissolved in the brine -- which will occur eventually as a result of interphase mass transfer even if CO2 is injected as a separate fluid phase -- the solubility constrains the capacity. Bruant et al therefore define CO2 storage capacity solely on the basis of physical parameters, that is, volumes and solubilities, such that capacity is the maximum amount of CO2 that can be stored in an aquifer by solubilization. On this basis, the authors calculate global capacity as follows:

The areal extent of worldwide sedimentary basins, not including those located under offshore seabeds, is ~70million km2. Assuming an average useful formation thickness of 200 m and an average aquifer porosity of 10%, the available volume is ~1.4 million km3. Assuming a CO2 solubility of 40 kg/m3, the available global storage capacity is ~56,000 GtCO2. Each of the parameters used in this calculation might reasonably be considered to have an uncertainty factor of 2. For example, an average aquifer porosity of 20% is also a reasonable estimate, and other studies have assessed the long term total storage capacity to be 30 gm/m3 for all storage mechanisms. Taking this uncertainty into consideration, the global storage capacity of land-based deep saline aquifers is estimated to be between 10,000 and 200,000 GtCO2. This would accommodate hundreds to thousands of years of CO2 emissions (15).


Hydrocarbon-producing basins represent numerically less than half the sedimentary provinces in the world. Other than in EOR, injected CO2 occupies the pore volume previously taken by the oil and gas, less that filled with residual water, which can be in the order of 30-50%. Estimates of global capacity vary according to assessment criteria, and the more accurate ‘bottom up’ evaluations are only available for north-western Europe, the US, Canada and Australia, based on URR (ultimately recoverable reserve) volumes. CO2 storage capacities are: Europe: 40Gt (gas), 7Gt (oil); Canada 4Gt (gas), 1Gt (oil); US 98Gt (oil and gas); Australia 15Gt (gas), 0.7Gt (oil). Significant yet-to-be-assessed storage potential also exists in the Middle East, Russia, Asia, Africa and Latin America.

Global capacity for CO2-EOR opportunities is estimated at 61-123 Gt CO2, though optimizing storage beyond current practice (which often minimizes CO2 injection) may increase these figures. Global estimates for storage capacity (as distinct from bottom-up approaches) vary from 126-400 Gt CO2 for oil and 800 Gt CO2 for gas. The combined estimate for discovered oil and gas fields is likely in the range 675-900 Gt CO2. Including undiscovered fields, this increases to 900-1200 Gt CO2.


Although it is recognized that deep aquifers offer a very large potential storage capacity for CO2 sequestration, the implementation of large volume injection to fill this capacity seems to be very complex. In low permeability formations or compartmentalized reservoirs the CO2 injection will result in an increase in pore pressure. This increase in pore pressure can limit the ability to inject CO2 into the subsurface, because overpressure-associated geomechanical damage needs to be avoided. In this case, the storage capacity mainly depends on pore and brine compressibilities that provide extended pore space availability, and on the maximum pressure buildup that the formation can sustain (16) (see side panel for detail).


Evidence from oil and gas fields indicates that hydrocarbons and other gases and fluids including CO2 can remain trapped for millions of years. Carbon dioxide has a tendency to remain in the subsurface (relative to hydrocarbons) via its many physico-chemical immobilization mechanisms. World-class petroleum provinces have storage times for oil and gas of 5-100 million years, others for 350 million years, while some have been stored for up to 1400 million years. Nonetheless, some natural traps do leak, reinforcing the need for careful site selection, characterization and injection practices.

The economic implications of such leakage are, however, minimal, even at rates far higher than would be considered likely from a typical storage site. In a 2003 paper (18) exploring the trade-offs between discounting, leakage, the cost of sequestration and the energy penalty (the cost of capture, transport and injection underground), it was found that if the annual leak rate is 1%, and the discount rate is 4%, then CO2 mitigation using leaky storage is worth 80% of mitigation with perfect storage, while a leakage rate of 0.1% is nearly the same as perfect storage. A rate of 0.5%, however, renders storage unattractive.


The cost of geological storage of CO2 is highly site-specific, depending on factors such as storage formation depth, number of injection wells, location (onshore or offshore), but lies in the range US$0.6-8.3/tCO2.


Whilst there is sufficient storage capacity in geologic formations for possibly hundreds of years of anthropogenic output, mineralization has even greater potential.  The process mimics natural weathering, in which gaseous and/or aqueous CO2 reacts with metal cations such as magnesium, calcium and iron, to form carbonate minerals:

Metal oxide + CO2 → Metal carbonate + Heat (1)

For example:

Mg2SiO4 + 2CO2 + 2H2O → 2MgCO3 + H4SiO4 (2)

Mg3Si2O5(OH)4 + 3CO2 + 2H2O → 3MgCO3 + 2H4SiO4 (3)

Fe2SiO4 + 2CO+ 2H2O → 2FeCO+ H4SiO4 (4)

CaSiO+ CO2 + 2H2O → CaCO3 + H4SiO4 (5)

Due to the abundance of silicates around the world, the storage capacity of mineral carbonation is very large (>10,000 GtC) (29).  Significantly also, there are none of the issues of permanence, long-term monitoring and verification associated with geologic sequestration. The inherent stability of mineral carbonation is evidenced by the distribution of carbon in the lithosphere, where more than 99% of the world’s carbon reservoir is locked up in limestone (Ca CO3), dolomite (MgCO3) and other types of carbonates.

Mineralisation can be broadly categorized as in situ or ex situ, the latter being an above-ground industrial process. Only the former will be considered here.

Carbonation is a natural process where CO2 reacts with different minerals forming solid precipitates leading to the weathering of the rocks. The reactions are spontaneous and exothermic but kinetically slow. They are exemplified by (6) and (7) where calcium and magnesium oxides are considered to react with CO2.

CaO + CO2 → CaCO3 + 179 kJ mol-1 (6)

MgO + CO→ MgCO+ 118 kJ mol-1 (7)

The most reactive compounds for CO2 mineralization are oxides of divalent metals, Ca and Mg, and their availability in nature is mainly in the form of silicates, such as olivine ((Mg,Fe)2SiO4) orthopyroxene (Mg2Si2O6–Fe2Si2O6), clinopyroxene (CaMgSi2O6–CaFeSi2O6) and serpentine ((Mg, Fe)3Si2O5(OH)4), the latter originated by the hydratation of olivine. When CO2 dissolves in water, it reacts with these silicates forming corresponding carbonates, where CO2 is fixed in a mineral form.

Mantle peridotite and basalts deposits, enriched in Mg, Fe and Ca silicates, are the main targets for in situ CO2 mineralisation projects, as outlined in further detail in the side panel opposite.


Transport and storage costs for mineral carbonation have been estimated at $17/t CO– about double the upper range of costs for geological storage in sedimentary basins (32). However, several factors are relevant here. First, given the locational flexibility of DAC, the need for transportation and storage is minimized. Secondly, geological storage costs do not take into account potential long-term monitoring. Finally, mineralization dramatically reduces potential leakage.


At present, there is no market for atmospheric reduction through DAC, as distinct from that applicable to carbon offsets or ’virtual’ zero-carbon fuels achieved through hypothecated sequestration of CO2 equivalent to that emitted during combustion. However, momentum is building in policy circles for some form of  global carbon pricing which, if it is to have the desired effect in terms of driving decarbonisation, will need ultimately to be at a realistic level, even if this is reached through progressive increase. Global policy, as currently anticipated following COP 21, is hamstrung by fiscal disincentivization in the form both of fossil fuel subsidies (directly, and indirectly as barely-concealed externalities), and the absence of a tax on carbon. As emphasized throughout the IEA’s special report, Energy and Climate Change 2015, ahead of the Paris summit, it is essential that these anomalies are addressed if there is to be any possibility of adhering to the 2oC maximum target. On this basis, the only realistic assumption is that global carbon pricing is at some point inevitable.

Irrespective of when such pricing is implemented, the requirement for atmospheric carbon reduction is, and will be, substantial. Even now, CO2 levels are at least 50 ppm above the level considered safe by leading climate scientists, and it is unlikely, on current projections, that these will peak below 550 ppm, and more likely will reach 650 ppm or higher. As explained in the introduction to this section, any reduction in atmospheric CO2 levels will have to account for re-emission from natural carbon sinks (see figure 2), increasing by around 67% the amount to be sequestered. Thus even a 50ppm requirement becomes 83.5ppm which, at 7.8 Gt CO2 per ppm, translates to 651.3 Gt CO2, or around 8.1 Gt CO2 per annum over an 80 year timeframe. Assuming 50% of this reduction is met through DAC, it would require an average of 4050 one-million tonne DAC plants for the duration of the programme.  DAC requirements for a range of peak CO2 emission levels are set out in Table 3 below.

Table 3: DAC Plant requirements for a range of peak atmospheric levels of CO2 400ppm-600ppm

Adopting a projected peak CO2 emissions level of 500ppm (as the median of the range 350-650ppm), reducing atmospheric CO2 to 350ppm would involve an average capital expenditure on DAC installations of around $215bn/year.

Although reducing atmospheric levels of CO2 is a global problem requiring a global response, this could be achieved substantially through the private sector, given an appropriate policy context and investment incentives. The key element in a favourable policy environment would be a global price on carbon — specifically at a level which facilitates large-scale DAC in conjunction with other GGR strategies. A carbon price of $140/tonne CO2, phased in over the course of a decade, would provide a clear market signal and timeline for development of a global GGR industry, including DAC,  across multiple sectors, in addition to atmospheric reduction.

Whilst, as mentioned elsewhere on this site, support is building steadily for a global price on carbon, there are many competing proposals as to what form this might take, ranging from largely ineffectual cap and trade markets such as the ETS, to Fee and Dividend schemes (as for instance legislated in British Columbia), to carbon rationing (Tradable Energy Quotas or TEQs). The challenge will be to devise a system which provides both a universal fiscal incentive, and the flexibility to accommodate a multiplicity of differing approaches by nations, sectors and markets.

For the purposes of DAC deployment, however, a global price on carbon need not be incompatible with any of these options. On the contrary, it would promote conformity in carbon-pricing between regions and countries, whilst allowing flexibility in distribution or allocation of the carbon tax income where it is imposed. F & D or carbon rationing could each co-exist with carbon pricing, allowing fees to be applied variously to redistribution, or investment in decarbonisation, or a combination of the two.

More challenging is the issue of atmospheric reduction under such a regime. No one country or region would act unilaterally to allocate tax revenue to incentivise DAC investment where the benefit accrues to the world as a whole, so there needs to be a mechanism whereby this can be facilitated at a global level, through an appropriate global institution. Given that the IMF is charged with responsibility for promoting financial stability, and that climate change threatens the entire global economy, it does not seem unreasonable for it to assume a lead role in creating a financial framework in which large-scale DAC deployment for atmospheric reduction could take place.

It is suggested that, for plant capital costs, the necessary funds are made available on a long-term basis (say 40 years) to financial institutions and DAC developers on terms which provide an attractive return for both lender and borrowers — say a flat rate of 2.5% per annum (equivalent to circa 5% APR). All of the capital costs of the DAC atmospheric reduction programme (an average of 305 plants/year over 40 years, at a cost of $213.5 bn annually) would be funded from this source, and fully repaid over the 40 year term of the loans.

Funding provision for the $140/tonne fees for CO2 sequestration is a much larger issue. Conceivably, this could be provided through the world’s central banks working in conjunction with the IMF, possibly in the form of QE. Part at least could be met through revenue from carbon taxation. Because the income stream for the DAC plants would be AAA grade quality, the return to the DAC financier/developers could be limited to (say) $20m per annum. Of the $140m per annum income from each one-million tonne DAC plant, operating costs would be $65m; $17.5m would be allocated to capital repayment; $17.5m to interest; $20m for the licence fee to Carbon Engineering; and the balance of $20m retained by the funder.

The total cost of carbon sequestration payments, calculated by multiplying the fees earned by each plant annually ($140m) by the average number of DAC plants required from 2020 to 2060 (12,220), results in a cost of around $1.7 trillion a year — or less than a third of the current annual cost of global fossil fuel subsidies. Over 40 years, it is also less than the $71 trillion the IEA calculates would accrue in net savings from decarbonising the energy sector by 2050 (33).

A cumulative limit of 1,440 GtCO2 would lead to a 50 percent probability of warming beyond 2oC  (Allen et al., 2009; Meinshausen et al., 2009).  The corresponding stabilization scenario developed by the IPCC ... has total emissions of about 1,600 GtCO2-eq from 2000 to 2100.  For comparison, business-as-usual scenarios (scenarios that do not assume additional policy action to reduce emissions) forecast 2,500 to 4,000 GtCO2-eq from 2000 to 2050, and 4,600 to 7,300 GtCO2-eq from 2000 to 2100. Thus, limiting warming to 2oC will require CO2 emissions reduction, post-emissions consumption by CDR, or some combination of these in the amounts of roughly 1,000 to 3,000 GtCO2 before 2050, and 3,000 to 6,000 GtCO2 before 2100.

Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration, National Research Council of the National Academies 2015


The typical benchmark for the rate of CO2 injection is 1 Mt/year when studying storage performance. This rate is very low compared to the scale necessary for the storage technology to play a significant role in managing global emissions. Quantitative investigation of a large volume CO2 injection and its potential barriers (20 Mt/year during 50 years of continuous injection resulting in a total sequestration of 1 Gt CO2), and sensitivity analysis of the results (plume area and CO2 storage capacity) within the range of aquifer parameters (thickness (50-100 m); permeability (25-100 mD); rock compressibility (from 9 x10-10 to 2 x10-9 (1/Pa)) as well as different injection arrangements, shows that the capacity of a reservoir in the case of large injection volumes should be evaluated not by available pore volume, but by the ability to inject some amount without exceeding fracture pressure of formation (17).

This work demonstrates that underground injection of substantial amounts of CO2 may be a very difficult task. The injection capacity might be limited by injection of large volumes of gas within a relatively small area and within a relatively short period of time. Injection capacity may be much lower than estimated by available pore space on its own. The reservoir pressure during injection may exceed the fracture pressure very fast and injection should be stopped before the target amount is injected. Large volumes would require a multiple injection well design, but it was shown that increasing the number of injection wells has diminishing returns. The sensitivity study provides an illustration of the degree to which each reservoir parameter influences CO2 injectivity and capacity. Permeability and net thickness of the formation have a direct impact on injectivity, but rock compressibility manifests its effect when a larger number of injectors is used (figure 9).


The risk of leakage is mainly from free-phase CO2. Dissolution effectively eliminates this, but it is a slow process taking thousands of years.

Computer modeling suggests that it may be possible to engineer CO2 storage in deep saline aquifers by accelerating the dissolution of CO2 in brines in order to reduce the long-term risk of leakage (a, b, c). Techniques include:

  1. optimizing the geometry of brine injection wells to maximize the rate at which buoyancy-driven flow of CO2 and brines drive dissolution, and 

  2. use of wells and pumps to transport CO2 or brines within the reservoir in order to increase contact between CO2 and undersaturated brines, thereby accelerating the rate of dissolution. Pumping brines from regions where it is undersaturated to those occupied by CO2 accelerates dissolution levels to 100% within 300 years at an energy cost that is less than 10% of the cost of CO2 compression to reservoir pressures (19). Injecting brine on top of the injected CO2 improves CO2 dissolution from 8% to over 50% over a 200 year period (20).

Such reservoir engineering techniques could play an important role in geologic storage for three distinct reasons. First, active reservoir engineering can reduce the risk of leakage. Second, such engineering may increase available storage capacity by increasing the range of aquifers in which CO2 can be safely stored. Third, by shortening the time scale over which free-phase CO2 remains in the reservoir, such methods might facilitate risk analysis and reduce regulatory and other uncertainties related to storing mobile CO2 underground for long durations (21).

The economics are compelling. Forcing the dissolution of ~80% of the injected CO2 over 300 years is estimated at ~$0.08/tonne CO2, approximately one thousandth the cost of air capture (22).


Another option for CO2 sequestration is ocean storage, where it would remain isolated from the atmosphere for centuries. CO2 could be transported by pipeline or ship for release in the ocean or on the sea floor. Despite 25 years of promising theoretical, laboratory and modeling studies, and small-scale field experiments, it has yet to be deployed or thoroughly tested.

Some 30% of anthropogenic COis already absorbed by the oceans, totaling around 500 GtCO2 out of 1300 GtCO2 over the last 200 years. Models predict that oceans will take up most of the CO2 released into the atmosphere over several centuries. Anthropogenic CO2 primarily resides in the upper ocean with the result that pH has been reduced by around 0.1, causing adverse effects on marine organisms. Deep oceans are largely unaffected and therefore represent a favoured option for burial.

Oceans cover 70% of the Earth’s surface to an average depth of about 3,800 metres; hence, there is no practical physical limit to storage potential. However, over millennia, CO2 injected into the oceans at great depth will approach the same equilibrium as if it were released to the atmosphere. Sustained atmospheric CO2 concentrations in the range 350-1000 ppmv imply 2,300 ± 260 to 10,700 ± 1000 Gt of anthropogenic CO2 will eventually reside in the ocean.

Analyses of ocean observations and models agree that injected CO2 will be isolated from the atmosphere for several hundreds of years and that the fraction retained tends to increase with depth of injection. Over centuries, ocean mixing results in loss of isolation of injected CO2, and exchange with the atmosphere. Injection of a few Gt CO2 would produce measurable change in ocean chemistry in the region of injection, whereas injection of hundreds of Gt CO2 would eventually produce measurable change over the entire ocean volume. Confinement in some form of container, such as inflatable bags (see below) would of course avoid this.

Liquid COinjected at depths above 2750 m is lighter than the surrounding seawater, and tends to rise towards the surface and release to the atmosphere. However, liquid CO2 is more compressible than seawater, and at pressures corresponding to depths greater than 2750 m liquid COis denser than seawater and therefore sinks towards the bottom. A CO2 lake would become covered with a layer of hydrate, retarding CO2 exchange with the atmosphere, although some would diffuse to higher levels, ultimately venting to the air.

The possibility of exploiting this to create lakes of liquid CO2 in depressions on the ocean floor has been the subject of a growing body of research (24, 25), and in some technical detail (26). Such schemes involve either under-sea pipelines or shuttle tanker delivery via a tailpipe, feeding a large inflatable membrane bag. A 1km long containment vessel would hold 0.16 Gt CO2, equivalent to about 1.8 days of current global emissions of 32.3 Gt CO2/year (figures 14-16).

Preliminary cost estimates are encouraging. Assuming a membrane cost of $100/m2, the membrane cost per tonne of CO2 stored is $3.1 – a small fraction of the capture cost. Installation and connections costs at scale are estimated at $0.035/tonne, yielding a total storage cost of $3.135/tonne CO2, in principle clearly commercially viable (27).


Peridotite is a component of ophiolites which are complex geological sequences representing the emplacement on land of sections of oceanic crust. The world’s largest ophiolitic outcrop, representing around 70,000 km3 (350 x 40 x 5 km) is the Samail Ophiolite in Oman, of which some 30% by volume is mantle peridotite, three quarters of which is olivine. Water infiltration through a network of fractures and reaction with the preidotite trigger the formation of serpentine, brucite, magnesite and dolomite leading to strongly alkaline (pH 12) springs in which contact with the atmosphere precipitates Ca-carbonates, mostly in the form of terraced travertine. The total volume of carbonate in the Samail Ophiolite is 5.5 x 10>7m3 with an average age of 26,000 years indicating that about
4 x 10>7 kg CO2/year (40,000 tonnes) are consumed by the precipitation of carbonates (figure 18).  

Artificial enhancement of this natural process could be achieved through injection of CO2 at a higher concentration and temperature:  at 90oC and 100 bar pCO2 about 0.63 kg CO2 can be permanently stored as carbonates for each kg of peridotite. A typical in situ project would include drilling of the peridotite, hydrofracturing, injection of heating fluids at 185oC (the optimum temperature for olivine carbonation), followed by injection of pure CO2 at 25oC. The exothermal reaction (producing 760 kJ kg_1) and the geothermal gradient (up to 20 1C km1) both contribute to the reduction in the energy needed for heating the fluids. The resulting enhancement of the carbonation rate following this process is considered to be one million times faster than the natural occurring process.


Basalts are found mostly on the oceanic crust, and to a lesser extent on the continental crust. The largest layered onshore basalt formations are located in India (provinces of Deccan Traps), USA (Columbia River basalts), Russia (Siberian Traps) and UAE/Oman (figure 19 and table 1).

A global assessment of deep-sea basaltic sites (30) reviewed the most secure locations that offered all of the key trapping mechanisms (figure 19 and table 1) and concluded that this provides ‘unique and significant advantages over other potential storage options, including: (a) vast reservoir capacities with high porosity and permeability, sufficient to accommodate centuries-long US production of fossil-fuel CO2 (b) chemical reactivity of CO2 with basalt and in situ fluids to produce stable, non-toxic carbonates; and (c) significant risk reduction for post-injection leakage by geological, gravitational, and mineral trapping mechanisms’.  The study found that the largest volumes and most secure sites occur in regions adjacent to intermediate to fast-spreading seismic ridges as well as deep sea seismic ridges: even individual oceanic basalt ridges have sufficient capacity for tens to hundreds of years of US anthropogenic carbon. Worldwide, the total volume is 4.1 Tt-C assuming minimum thickness estimates, which may understate the storage potential by as much as ten times. Aseismic ridge sites may provide three times greater capacity than ridge sites; however, ‘pore volume estimates and area assessments are significantly less well-constrained’.

Basalts have a good degree of secondary permeability due to networks of fractures during or after their deposition. The resulting pore space may be filled with circulating water enriched with Ca or Mg ions which can react with the CO2, precipitating carbonates:

(Ca2+, Mg2+) + CO2 + H2O → (Ca, Mg)CO3 + 2H+ (8)

The reaction rate is controlled by the concentration of H+ and its neutralization by Mg, Al and Ca silicates which therefore represent the controlling factors in the development of in situ mineralization based on this process. Injecting CO2 within Basalts on the ocean floor benefits from a further series of trapping mechanisms: below 2,700 m where the temperature is below 2oC, CO2 is denser than seawater, causing it to sink  (gravitational trapping); encagement in hydrate ice lattices limits solubility; and low-permeability of the thick ocean floor further reduces the possibility of leakage.

A number of injection-test and feasibility projects are currently underway to address the potential of Basalts, both on- and offshore.  For instance, the Kevin Dome Project in Montana being undertaken by the Blue Sky Carbon Sequestration Partnership, aims to inject 1 million tonnes of COinto the regionally extensive Duperow carbonate formation, utilizing a wide range of characterization techniques, evaluation procedures and modeling analyses to monitor the geology, geochemistry, water quality, air quality and CO2 behaviour, with the objective of demonstrating the potential for long term basalt storage (31).



The Statoil Sleipner project has been in operation in the Norwegian sector of the North Sea since autumn 1996, incentivized through Norway 315 NOK/tonne (about $55/tonne) carbon tax (Figures 10-12) By early 2005, more than 7Mt CO2 had been injected at a rate of approximately 2700t/day, out of an expected 20Mt CO2 over the project lifetime. Sleipner gas is 9 per cent carbon dioxide. The project extracts carbon dioxide from the gas on the production platform, and injects it into the Utsira formation, at a depth of 800 m below the seabed. The sandstone aquifer formation is 250 m thick, and is thought to have a capacity of 600Gt CO2. The cap is an 80 m layer of shale, several hundred km long, and 150 km wide. Statoil has stated that ‘The entire carbon dioxide emissions from all the power stations in Europe could be deposited in this structure for 600 years’ (this presumably refers to contemporary emissions rates, and does not allow for exhaustion of fossil fuels).

The project has three phases: (i) baseline data-gathering and evaluation (completed in November 1998) (ii) status evaluation following three years of injection (reservoir geology; reservoir simulation; geochemistry; assessment of monitoring requirements; and geophysical modeling), and data interpretation and verification. Monitoring by seismic time-lapse has confirmed that the caprock is an effective seal that prevents CO2 migration out of the storage formation. Reservoir studies and simulations covering thousands of years have shown that CO2 will eventually dissolve in the pore water, which will become heavier and sink, minimizing the potential for long-term leakage.


Located in the central Saharan region of Algeria, In Salah is the world’s first large-scale CO2 storage project in a gas reservoir. A joint-venture between Sonatrach, BP and Statoil, the Krechba field at In Salah produces natural gas containing up to 10% CO2 which is re-injected into a sandstone reservoir at 1800m depth, storing up to 1.2 MtCO2/year from April 2004, toward an expected total of 17 Mt CO2 over the project lifetime. A preliminary risk assessment of CO2 storage integrity has been carried out and baseline data acquired.


A CO2-EOR project established in Weyburn, southern Saskatchewan is expected to inject 23 MtCO2, and extend the life of the oilfield by 25 years, using CO2 piped from a coal gasification facility 325 km south in Beulah, North Dakota. The CO2-EOR project is expected to take CO2 for about 15 years, with delivered volumes dropping from 5000 to 3000 t/day over the life of the project.

Since CO2 injection began in 2000, the project has performed largely as predicted, producing around 10,000 barrels/day of incremental oil, with all produced CO2 (circa 1000t/day) captured and reinjected. Extensive surface monitoring has revealed no leakage to date either to the surface or near-surface environment.


In the US, approximately 73 EOR projects inject 30 Mt CO2/year, most from natural CO2 accumulations. CO2-EOR projects are also underway in a number of countries, including Trinidad, Turkey and Brazil. In addition to these commercial projects , a number of pilot storage projects are underway or planned: The Frio Brine Project in Texas, involving injection in a highly permeable formation with a regionally extensive shale seal; Ketzin, west of Berlin, Germany; the Ottway Basin of southeast Australia; and the Teapot Dome in Wyoming. There are others in the US, Europe, including Italy and Poland, and Japan (case studies all taken from 23) (figure 13).


For deep GS to be safe and secure, with minimal risk of leakage, surface disruption, or contamination of other geological resources, appropriate care and monitoring need to be regulated.

The US and many other countries already inject large volumes of fluid underground, and one application in particular – fracking – is a source of growing public and community opposition. Given the nature of CO2, its role in climate change and the potential for large-scale sequestration, regulating deep GS will require particular attention and some degree of international coordination. This is paramount because the volume of fluid will be higher than in many other injection projects and because, at the time of injection, CO2 is a buoyant fluid. Moreover, public skepticism concerning industry assurances regarding the risks associated with fracking will inevitably require the highest possible standards in the sequestration of CO2.

The life cycle of a GS project (figure 17) involves four separate stages (28, from which the remainder of this section is also sourced): site characterization and permitting before any injection; site operation; post-closure operations by the site operator; and long term stewardship (figure 15). If large-scale GS is to proceed, the competing needs and interests of the public, project developers, financial and insurance institutions, government regulatory agencies, non-governmental organizations, and national and international agencies managing CO2 trading must be appropriately balanced. The goal is to create an efficient regulatory regime that ensures safe and responsible GS deployment, protects local health and environments, meets the needs of national and international  climate frameworks, and is cost-effective.

Although existing law can be used to cope with experimental projects, full-scale commercial deployment will require a much more comprehensive approach to selecting sites and allocating responsibility, more appropriate rules for accounting and monitoring, and minimum standards to ensure an adequate level of safety wherever GS is deployed worldwide. Indeed, past experience suggests that simply scaling up existing regulations for commercial-scale GS projects can have serious pitfalls.  For example, the U.S. experience in Florida, where wastewater that had been injected underground migrated into underground sources of drinking water, illustrates both the problems that can arise when very large quantities (~3 Gt/yr of wastewater) are injected into unsuitable geological formations and the subsequent difficulties that can result from making ad hoc modifications to an existing regulatory regime. In addition, the current EU and US regulatory regimes do not deal adequately with pre-injection site characterization, ongoing monitoring during site operation, large-scale fluid displacement, continued post-closure site monitoring, long-term liability, and other issues. Many existing schemes do not clarify subsurface property ownership of pore space or subsurface trespass (by post-injection COmovement) laws, which vary significantly across jurisdictions. Although specific rules will vary across nations, these and other issues must be addressed before large-scale deployment will be possible.

Experts from the insurance industry indicate that existing health, safety, and environmental liability frameworks can cover the potential risks of GS, with the notable exception of risks associated with long term stewardship. Because most firms do not last for centuries, there is wide agreement that long-term responsibility for the stewardship of closed sites must be assumed by national governments or institutions designed to last for many hundreds of years

Governments worldwide should provide incentives for initial large-scale GS projects to help build the knowledge base for a mature, internationally harmonized GS regulatory framework. Health, safety, and environmental risks of these early projects can be managed through modifications of existing regulations in the EU, Australia, Canada, and the US.  An institutional mechanism, such as the proposed Federal Carbon Sequestration Commission in the US, should gather data from these early projects and combine them with factors such as GS industrial organization and climate regime requirements to create an efficient and adaptive regulatory framework suited to large-scale deployment. Mechanisms to structure long-term liability and fund long-term post-closure care must be developed, most likely at the national level, to equitably balance the risks and benefits
of this important climate change mitigation technology.

'Adopting a projected peak CO2 emissions level of 500ppm (as the median of the range 350-650ppm), reducing atmospheric CO2 to 350ppm would involve an average capital expenditure on DAC installations of around $215bn/year.  This is less than 5% of the $5.3trn in estimated subsidies and cost externalities for the fossil fuel sector in 2015 alone.'

'Assuming that solar energy is used to fuel the DAC process and that ~100,000,000 acres of Bureau of Land Management (BLM) land are available in the southwestern United States, this could lead to a removal of ~13 GtCO2/yr and a cumulative removal of ~1,100 GtCO2 up to 2100'

Climate Intervention: Carbon Dioxide Removal and Reliable Sequestration, National Research Council of the National Academies 2015