Carbon dioxide removal (CDR) systems - touted as techno-fixes for global warming - usually put more greenhouse gases into the air than they take out, a recent study has confirmed.
Carbon capture and storage (CCS), which grabs carbon dioxide (CO2) produced by coal or gas fired power stations, and then uses it for enhanced oil recovery (EOR), emits between 1.4 and 4.7 tonnes of the gas for each tonne removed, the research shows.
Direct air capture (DAC), which sucks CO2 from the atmosphere, emits 1.4-3.5 tonnes for each tonne it recovers, mostly from fossil fuels used to power the handful of existing projects.
If DAC was instead powered by renewable electricity – as its supporters claim it should be – it would wolf down other natural resources.
And things get worse at large scale.
To capture 1 gigatonne of CO2 (1 GtCO2, just one-fortieth of current global CO2 emissions) would need nearly twice the amount of wind and solar electricity now produced globally. The equipment would need a land area bigger than the island of Sri Lanka and a vast network of pipelines and underground storage facilities. (See endnote 1.)
Claims made that CCS could be “green” – by generating the energy from biofuels, and/or storing the carbon instead of using it for oil production – do not stand up to scrutiny either, the article shows.
The paper – Assessing Carbon Capture: public policy, science and societal need, by June Sekera, a public policy analyst, and Andreas Lichtenberger, an ecological economics researcher – is free to download on the Biophysical Economics and Sustainability web site.
Sekera and Lichtenberger demolish the case put by governments and fossil fuel companies for investing in CDR systems – and show how research methods are slanted to avoid discussion of the full resource costs.
They also challenge the way that so much CDR research focuses not on its extremely dubious worth as a tool to combat climate change, but on whether it can make money. Economists envisage the CO2, in a gaseous or solid form, being marketed as a commodity – but this, too, could operate at scale only in the world of techno fantasy and/or late capitalist dystopia.
Sekera and Lichtenberger write that “market actors seeing avenues for profit [and] seeking government support” are the main promoters of CDR by mechanical and chemical methods (such as CCS and DAC), as opposed to natural methods such as planting trees.
Fossil fuel producers are keen too. They falsely claim that CCS can help produce “green” electricity from coal or gas. Moreover, the main use to which sequestered carbon is currently put is for enhanced oil recovery (EOR), an oil and gas production technique: the carbon is pumped into underground reservoirs containing oil and gas, helping to push these products to the surface.
Governments have long backed industrial CDR, and that has intensified since the Intergovernmental Panel on Climate Change (IPCC) reports of 2014 and 2018, which pointed to negative emissions technologies, in particular Bioenergy with CCS (BECCS) – producing electricity from biofuels, and sucking back the carbon emitted with CCS – as a way to meet decarbonisation targets.
Climate scientists have slammed the scenarios in these reports that rely on unrealistic and dangerous assumptions about using a vast proportion of the world’s land to grow crops for bioenergy. (See e.g. here, here or here.) But that has not stopped state backing for CDR.
The US government alone sank $5 billion into research of CDR in 2010-18, Sekera and Lichtenberger point out. The UK Committee on Climate Change says CDR is a “necessity”, and the European Commission has incorporated industrial CDR into its “green deal”. The trough of development funding is getting bigger.
Humanity’s “collective biophysical need” to reduce the amount of CO2 in the atmosphere is the standard by which Sekera and Lichtenberger judge CDR technologies.
This helps them to cut through mountains of hype about CDR, including articles focused on how “cost effective” it is, or claiming it is “carbon reductive” compared to a hypothetical “business as usual” case.
They reviewed more than 200 scientific papers, and analysed what they said about: the impact of each CDR technology on the total carbon balance; the resource usage required for it to be used at scale; and the biophysical impacts, especially at scale.
In the case of CCS, Sekera and Lichtenberger point out that it “can not reduce the stock of atmospheric CO2 since it can not store more than it captures”. In order for the CO2 to be captured by this method, it has to be added to the stock in the first place, almost always by a power station or industrial process.
If the carbon is stored, rather than used for oil production, “the process could avoid being net [CO2] additive”, Sekera and Lichtenberger concede. But renewable electricity generation linked to storage would be much more energy-effective, as a research team led by Sgouris Sigouridis at Khalifa university, Abu Dhabi, calculated recently – so why would you invest in fossil fuels plus CCS in the first place?
Furthermore, in the real world, now, CO2 captured by CCS is not stored, but used for EOR – that is, to produce more oil. All five of the largest CCS projects recently listed by an oil industry journal (one of which, Petra Nova, is currently mothballed) send the captured CO2 for EOR. (See endnote 2.)
Other CCS facilities, that take the carbon out of natural gas without the gas being burned, e.g. in the manufacture of hydrogen, also very often send the CO2 produced for use in EOR.
The Global CCS Institute, an industry body, says that, of ten new projects it recently listed, three will send CO2 for EOR, two are “considering” it and only two plan dedicated geological storage. No information is provided for the other three.
The life cycle analysis of CCS+EOR’s global warming impact is dire. Paulina Jaramillo of Carnegie Mellon university, USA, and her colleagues estimated that CCS+EOR emits 3.7-4.7 tonnes of CO2 for each tonne sequestered – and in the ten years since their article was published, no-one has questioned their numbers.
Researchers who claim that CCS+EOR is carbon-negative are not telling the full story, Sekera and Lichtenberger write.
"We found that papers that deem CCS-EOR to be a climate mitigation technique either fail to account for all emissions (i.e. they perform only a partial life cycle analysis) and/or they make an assumption that CCS-EOR-produced oil “displaces” conventionally produced fossil fuel energy."
DAC technologies suck CO2 from the air using sorbents (solid chemicals that absorb the CO2 molecules), or aqueous solutions containing amines (nitrogen-based compounds). All these techniques require a ferocious amount of energy.
High-temperature sorbent-based techniques need masses of heat energy, usually supplied by burning natural gas – which straight away makes them net CO2 emitters.
Pilot projects using low-temperature sorbent and amine systems are being run from electricity alone, or supplemented by spare heat from other processes.
Sekera and Lichtenberger found no published research with a complete life cycle analysis, including the manufacture of sorbents or amines, of renewables-powered DAC.
What they did find is that tremendous quantities of energy would be needed to remove any significant amount of CO2 from the atmosphere.
This is where the issues of scale come in. One study claims that, of the 40+ GtCO2 pumped into the atmosphere by human economic activity, 2.5 GtCO2 could be captured by DAC in 2030, increasing to 8-10 Gt CO2 by 2050.
But, Sekera and Lichtenberger point out, the largest of the handful of existing DAC facilities now working captures 0.000004 GtCO2 per year. (Yes, that’s four millionths of a GtCO2! There are five zeroes there.) The largest projected facility aims to capture 0.001 GtCO2/yr.
For these systems to be scaled up, using renewable energy, implies using prodigious amounts of wind and solar power. (As though there are not enough concerns already about the danger of resources being grabbed from the global south for large-scale renewables.)
For those who like numbers, the estimates of energy required to capture 1 GtCO2 are: 1390-2789 terawatt hours (TWh) (US National Academies, capture only); 3580.9 TWh (Smith et al, capture and storage); 3417 TWh (Climate Advisers, capture and storage). To compare: total world electricity generation from renewables in 2019 was 2805.5 TWh. (See endnote 1.)
On top of the energy, there’s the land. And the “massive mobilisation and diversion of material, human and energy resources”, Sekera and Lichtenberger point out. And, once you scale up technologies of this kind, biophysical impacts include “groundwater contamination, earthquakes caused by vast volumes of CO2 stored underground; [and] ‘fugitive emissions’ that pollute the air.”
Sekera and Lichtenberger found that, for DAC enthusiasts, the “major, but generally ignored policy issue” is: "[W]hether renewable energy should be channelled for carbon removal, rather than used directly to reduce carbon emissions by powering homes, industry, businesses and transport."
Yes. Why not just use the renewably-produced electricity – hard enough itself to produce at scale without causing further resource stresses – and close down fossil-fuelled power stations? Assuming you are not just trying to invent survival strategies for oil and gas corporations, that is.
Politicians and researchers alike are avoiding the question of whether using electricity generated from renewables to power DAC can ever make sense, Sekera and Lichtenberger charge.
"Scientific and technical papers increasingly acknowledge that fossil fuel-powered DAC is thermodynamically counterproductive, yet those same papers fail to tackle the consequential question of whether renewable energy should be funnelled to DAC, rather than used to directly supply energy for buildings and transport."
And the massive land requirements and biophysical impacts are largely “ignored”.
Sekera and Lichtenberger call for all existing subsidies for carbon removal systems for EOR (effectively, subsidies for oil production) to be removed. No system that puts more CO2 into the atmosphere than it takes out should get state support, they argue.
As for DAC powered by renewables, they say that it should be discussed as “a public service to meet a societal need, which is to achieve an absolute reduction in atmospheric CO2”.
They are sceptical that DAC can ever work at scale. The political question is “whether industrial-mechanical carbon removal is a realistic option”. Before any support is given to it, it should be analysed according to (i) its overall impact on the carbon balance, (ii) the effect of using resources (electricity, land, labour, and so on) at large scale, and (iii) the biophysical impacts of using it at scale.
Sekera and Lichtenberger argue that DAC and other industrial techniques need to be compared with biological methods of carbon removal – such as reforestation and afforestation; farming techniques; and grasslands and wetlands restoration – that they propose, preliminarily, will be “more effective and efficient” in energy and resource usage.
A report by Climate Advisers says that natural solutions are “the most readily available”, have already been deployed on a large scale for decades, are more cost-effective, offer numerous co-benefits and “should be a component of all truly visionary international climate action agendas”.
And by the way: BECCS is not a biological method of carbon removal. It is a way of burning biological material as fuel, with the carbon captured mechanically.
For social and labour movements, and all who are concerned about climate change, Sekera and Lichtenberger’s research must surely be taken seriously.
Technologies are not neutral. They work in social contexts. In contrast to biological methods, big, industrial CDR will – for the foreseeable future at least – be controlled by oil companies, or by the state. These technologies are by their nature inimical to collective control or operation.
We should therefore be wary – as Trade Unions for Energy Democracy are – of union bosses who support CCS, supposedly to “protect jobs” but actually to give a new lease of life to fossil fuel industries. And we should have serious, thoughtful discussions about the relationship of technological change and social change.
All our efforts should be directed to changing the big technological systems that consume fossil fuels – urban transport and buildings, industry and agriculture, military and state systems – to reduce wasteful and unnecessary consumption.
We should fight for existing technologies, preferably small scale ones, to change these systems, as trades unionists in Leeds are doing; fight to accelerate the transition away from fossil fuels; and fight to unmask hypocritical “climate emergencies” that are a cover for inaction. GL, 13 November 2020.
Gabriel Levy is a writer with an interest in socialism, ecology and the former Soviet countries, who blogs at People and Nature, where this article was first published. He has been active in the labour movement all his life.
Endnote 1. Sekera and Lichtenberger, using calculations made in Peter Smith et al, “Biophysical and economic limits to negative CO2 emissions”, Nature Climate Change 2016 (6), show that to capture one-tenth of a gigatonne of carbon dioxide (1/10 GtCO2), all wind and solar power generated in the US in 2018 (370 TWh) would be needed. So to capture 1 GtCO2, 3700 TWh (about twice the total world wind and solar output in 2018 of 1853 TWh, as stated in the BP Statistical Review 2019) would be needed. The National Academies of Sciences, Engineering and Medicine report, Negative Emissions Technologies and Reliable Sequestration: a research agenda (2019), gives an estimate of 5-10 GJ (=1.39-2.78 MWh) for capturing one tonne of CO2 by DAC; that figure, on p. 10, apparently refers to a range of values in table 5.11 on p. 222. Climate Advisers, in their report Creating Negative Emissions: the role of natural and technological carbon dioxide removal strategies (2018), give an estimate of 12.3 GJ/tonne (3.417 MWh/tonne) for carbon capture, processing, transportion and injection into storage. On land area, Sekera and Lichtenberger, again citing the National Academies report, say that 10 times the state of Delaware, i.e. 64,460 km2 (or 10 x 6446 km2) would be needed to recover 1 GtCO2. For non-US readers I use the example of Sri Lanka (land area 64,630 km2). That is roughly equal to half the land area of Greece (128,900 km2).
Endnote 2. The NS Energy article to which I linked states that Occidental’s Century plant, the world’s biggest CCS facility, supplied CO2 to an industrial hub. But actually it ends up being used for EOR in the Permian basin oil field, as Occidental states here. The article doesn’t state the destination of CO2 from SaskPower’s Boundary Dam project. If it has changed since 2016, when 90% of the CO2 was going for EOR, the company has kept quiet about it.