Carbon dioxide removal (CDR) refers to a number of technologies, the objective of which is the large-scale removal of carbon dioxide from the atmosphere. Among such technologies are bio-energy with carbon capture and storage, biochar, ocean fertilization, enhanced weathering, and direct air capture when combined with storage. CDR is a different approach than removing CO2 from the stack emissions of large fossil fuel point sources, such as power stations. The latter reduces emission to the atmosphere but cannot reduce the amount of carbon dioxide already in the atmosphere. As CDR removes carbon dioxide from the atmosphere, it creates negative emissions, offsetting emissions from small and dispersed point sources such as domestic heating systems, airplanes and vehicle exhausts. It is regarded by some as a form of climate engineering, while other commentators describe it as a form of carbon capture and storage or extreme mitigation. Whether CDR would satisfy common definitions of "climate engineering" or "geoengineering" usually depends upon the scale on which it would be undertaken.
The likely need for CDR has been publicly expressed by a range of individuals and organizations involved with climate change issues, including IPCC chief Rajendra Pachauri, the UNFCCC executive secretary Christiana Figueres, and the World Watch Institute. Institutions with major programs focusing on CDR include the Lenfest Center for Sustainable Energy at the Earth Institute, Columbia University, and the Climate Decision Making Center, an international collaboration operated out of Carnegie-Mellon University's Department of Engineering and Public Policy.
The mitigation effectiveness of air capture is limited by societal investment, land use, availability of geologic reservoirs, and leakage. The reservoirs are estimated to be sufficient to for storing at least 545 GtC. Storing 771 GtC would cause an 186 ppm atmospheric reduction. In order to return the atmospheric CO2 content to 350 ppm we need atmospheric reduction of 50 ppm plus an additional 2 ppm per year of current emissions.
Carbon dioxide removal is different from reducing emissions, as the former produces an outlet of carbon dioxide from Earth's atmosphere, whereas the latter decreases the inlet of carbon dioxide to the atmosphere. Both have the same net effect, but for achieving carbon dioxide concentration levels below present levels, carbon dioxide removal is critical. Also for meeting higher concentration levels, carbon dioxide removal is increasingly considered to be crucial as it provides the only possibility to fill the gap between needed reductions to meet mitigation targets and global emission trends.
In the OECD Environmental Outlook to 2050 released at the 2011 United Nations Climate Change Conference, the authors commented on the need for negative emissions, stating "Achieving lower concentration targets (450 ppm) depends significantly on the use of BECCS".
A carbon dioxide sink such as a concentrated group of plants or any other primary producer that binds carbon dioxide into biomass, such as within forests and kelp beds, is not carbon negative, as sinks are not permanent. A carbon dioxide sink of this type moves carbon, in the form of carbon dioxide, from the atmosphere or hydrosphere to the biosphere. This process could be undone, for example by wildfires or logging.
Carbon dioxide sinks that store carbon dioxide in the Earth's crust by injecting it into the subsurface, or in the form of insoluble carbonate salts (mineral sequestration), are considered carbon negative. This is because they are removing carbon from the atmosphere and sequestering it indefinitely and presumably for a considerable duration (thousands to millions of years). However, Carbon Capture technology remains, at best, theoretical and is yet to reach more than 33% efficiency. Furthermore, this process could be rapidly undone, for example by earthquakes or mining.
Bio-energy with carbon capture and storage, or BECCS, uses biomass to extract carbon dioxide from the atmosphere, and carbon capture and storage technologies to concentrate and permanently store it in deep geological formations.
BECCS is currently (as of October 2012) the only CDR technology deployed at full industrial scale, with 550 000 tonnes CO2/year in total capacity operating, divided between three different facilities (as of January 2012).
The Imperial College London, the UK Met Office Hadley Centre for Climate Prediction and Research, the Tyndall Centre for Climate Change Research, the Walker Institute for Climate System Research, and the Grantham Institute for Climate Change issued a joint report on carbon dioxide removal technologies as part of the AVOID: Avoiding dangerous climate change research program, stating that "Overall, of the technologies studied in this report, BECCS has the greatest maturity and there are no major practical barriers to its introduction into today's energy system. The presence of a primary product will support early deployment."
Biochar is created by the pyrolysis of biomass, and is under investigation as a method of carbon sequestration. Biochar is a charcoal that is used for agricultural purposes which also aids in carbon sequestration, the capture or hold of carbon. It is created using a process called pyrolysis, which is basically the act of high temperature heating biomass in an environment with low oxygen levels. What remains is a material known as char, similar to charcoal but is made through a sustainable process, thus the use of biomass. Biomass is organic matter produced by living organisms or recently living organisms, most commonly plants or plant based material. The offset of GHG emission, if biochar were to be implemented, would be a maximum of 12%. This equates to about 106 metric tons of CO2 equivalents. On a medium conservative level, it would be 23% less than that, at 82 metric tons. A study done by the UK Biochar Research Center has stated that, on a conservative level, biochar can store 1 gigaton of carbon per year. With greater effort in marketing and acceptance of biochar, the benefit could be the storage of 5-9 gigatons per year of carbon in biochar soils.
Enhanced weathering refers to chemical approach to remove carbon dioxide involving land or ocean based techniques. Examples of land based enhanced weathering techniques are in-situ carbonation of silicates. Ultramafic rock, for example, has the potential to store thousands of years worth of CO2 emissions according to one estimate. Ocean based techniques involve alkalinity enhancement, such as, grinding, dispersing and dissolving olivine, limestone, silicates, or calcium hydroxide to address ocean acidification and CO2 sequestration. Enhanced weathering is considered as one of the least expensive of geoengineering options. One example of a research project on the feasibility of enhanced weathering is the CarbFix project in Iceland.
Carbon dioxide can be removed from ambient air through chemical processes, sequestered, and stored. Traditional modes of carbon capture such as precombustion and postcombustion CO2 capture from large point sources can help slow the rate of increase of the atmospheric CO2 concentration, but only the direct removal of CO2 from the air, or "direct air capture" (DAC), can actually reduce the global atmospheric CO2 concentration if combined with long-term storage of CO2.
A few engineering proposals have been made for removing CO2 from the atmosphere, but work in this area is still in its infancy.  Among the main technologies proposed, three of them stand out: Causticization with alkali and alkali-earth hydroxides; Carbonation  and Organic-inorganic hybrid sorbents consisting of amines supported in porous adsorbents. For a review on the most recent research on these areas, see 
One proposed method is by so-called artificial trees. This concept, proposed by climate scientist Wallace S. Broecker and science writer Robert Kunzig, imagines huge numbers of artificial trees around the world to remove ambient CO2. The technology is now being pioneered by Klaus Lackner, a researcher at the Earth Institute, Columbia University, whose artificial tree technology can suck up to 1,000 times more CO2 from the air than real trees can, at a rate of about one ton of carbon per day if the artificial tree is approximately the size of an actual tree. The CO2 would be captured in a filter and then removed from the filter and stored.
The chemistry used is a variant of that described below, as it is based on sodium hydroxide. However, in a more recent design proposed by Klaus Lackner, the process can be carried out at only 40 °C by using a polymer-based ion exchange resin, which takes advantage of changes in humidity to prompt the release of captured CO2, instead of using a kiln. This reduces the energy required to operate the process.
In 2008, the Discovery Channel covered the work of David Keith, of University of Calgary, who built a tower, 4 feet wide and 20 feet tall (1.2×6.1 meters), with a fan at the bottom that sucks air in, which comes out again at the top. In the process, about half the CO2 is removed from the air.
This device uses the chemical process described in detail below. The system demonstrated on the Discovery Channel was a 1/90,000th scale test system of the capture section; the reagents are regenerated in a separate facility. The main costs of a full plant will be the cost to build it, and the energy input to regenerate the chemicals and produce a pure stream of CO2.
To put this into perspective, people in the U.S. emit about 20 tonnes of CO2 per person annually. In other words, each person in the U.S. would require a tower like the one featured by the Discovery Channel to remove this amount of CO2 from the air, requiring an annual 2 megawatt-hours of electricity to operate it. By comparison, a refrigerator consumes about 1.2 megawatt-hours annually (2001 figures). But, by combining many small systems such as this into one large system, the construction costs and energy use can be reduced.
It has been proposed that the Solar updraft tower to generate electricity from thermal air currents also be used at the same time for amine gravity scrubbing of CO2. Some heat would be required to regenerate the amine.
A similar CO2 scrubber has also been built by Carbon Engineering. Besides simply focusing on capturing the CO2, the company also puts emphasis on reuse of the CO2, for example in the production of fuels, which would thus be carbon-neutral.
Direct air capture has been proposed as a way of generating carbon-neutral organic chemicals, by harvesting the atmospheric compounds and then using them in the production and synthesis of polymers and fuels.
The Swiss-based Climeworks built the first industrial scale direct air capturing plant in Hinwil, in the canton of Zurich, Switzerland. It started up in May 2017, and could scrub up to 900 metric tons of CO2 per year using heat from a local waste incineration plant. The CO2 was then pumped into local greenhouses, where it was used to help grow vegetables like tomatoes and cucumbers.
Climeworks and Reykjavik Energy also started up a small test direct air capturing plant in Hellisheidi, Iceland in October 2017. Using waste heat from a local geothermal power plant, the test plant captured up to 50 tons of CO2 per year which was bound with water and injected over 700 metres underground into basaltic bedrock. The CO2 reacted with the basalt to form solid carbonate minerals. 
Ocean fertilization or ocean nourishment is a type of climate engineering based on the purposeful introduction of nutrients to the upper ocean to increase marine food production and to remove carbon dioxide from the atmosphere. A number of techniques, including fertilization by iron, urea and phosphorus have been proposed.
Calcium oxide (quicklime) will absorb CO2 from atmospheric air mixed with steam at 400 °C (forming calcium carbonate) and release it at 1,000 °C. This process, proposed by A. Steinfeld, can be performed using renewable energy from thermal concentrated solar power. Quicklime is made by heating limestone to release the CO2 within it. Quicklime is mixed with sand for brick building as mortar, where it hardens by absorption of CO2.
Zeman and Lackner outlined a specific method of air capture using sodium hydroxide. Carbon Engineering, a Calgary, Alberta firm founded in 2009 and partially funded by Bill Gates, is developing a process to capture carbon dioxide using a solution of potassium hydroxide mixed with some water at their pilot plant . They hope to create and sell synthetic fuels at a cost of $100 a ton.
Among the technologies studied for direct air capture (DAC), the use of aqueous hydroxide sorbents is one of the most promising approaches. In this process, CO2 from the air is chemically dissolved into NaOH(aq) solution as Na2CO3; the Na2CO3 is then reacted with solid Ca(OH)2, which regenerates the solvent and produces CaCO3 crystals; lastly, heat is applied to the CaCO3 crystals to produce pure CO2 gas.
Air is pumped through the CO2 absorber as the first step of this process. CO2 absorber for DAC are designed either as a counter-current spray tower or as a counter-current thin-falling-film contractor to maximize the contact area between the air and the solvent and thus maximize the absorption driving force. The solvent is regenerated in the causticization unit by reacting the Na2CO3 with Ca(OH)2, which also transfers the captured CO2 to the form of CaCO3 solid crystals. A mechanical filter is then used to separate the CaCO3 crystals form the water. Since the crystals come out wet from the filter, they are dried in a steam dryer. Then the dry crystals are heated in a furnace to produce CaO and pure CO2 gas. The CaO is then hydrated to regenerate the Ca(OH)2 used for the causticization reaction. The pure CO2 stream is then compressed and ready to be transported for geologic sequestration, EOR, or other commercial applications.
1 M NaOH (aq) is a typical solvent concentration because this concentrations is limited by the causticization reaction that regenerates the solvent it is not too far from the practical maximum of 2 M NaOH. The furnace/kiln can be powered renewably or by burning fuel on-site with pure oxygen produces in an on-site air separation unit.
NaOH is economically competitive with other absorbents (e.g. Amines) used for DAC processes. DAC processes are energy intensive. Calcination (at the furnace) is the most energy intensive step of this process.
A crucial issue for CDR methods is their cost, which differs substantially among the different technologies: some of these are not sufficiently developed to perform cost assessments. The American Physical Society estimates the costs for direct air capture to be $600/tonne with optimistic assumptions. The IEA Greenhouse Gas R&D Programme and Ecofys provides an estimate that 3.5 billion tonnes could be removed annually from the atmosphere with BECCS (Bio-Energy with Carbon Capture and Storage) at carbon prices as low as EUR50 per tonne, while a report from Biorecro and the Global Carbon Capture and Storage Institute estimates costs "below EUR100" per tonne for large scale BECCS deployment.
CDR is slow to act, and requires a long-term political and engineering program to effect. CDR is even slower to take effect on acidified oceans. In a Business as usual concentration pathway, the deep ocean will remain acidified for centuries, and as a consequence many marine species are in danger of extinction.