Recently, the American Physical Society (APS) released a report on the direct capture of carbon dioxide from air. The report concludes that air capture could be a powerful tool for mopping up carbon dioxide emissions that otherwise would escape to the air, for providing carbon dioxide for synthetic liquid fuels in the transportation sector, and for gradually reducing the concentration of carbon dioxide in the atmosphere. The report rightly notes that air capture cannot reduce atmospheric concentrations rapidly, and thus provides no excuse for delaying action against climate change. However, the report recognizes that air capture is one among few options that can reduce atmospheric carbon dioxide concentrations. Even with an all-out effort to stop emissions, we will likely overshoot any safe stabilization target, and returning to the target is best accomplished with air capture. Air capture does not replace capture at power plants, renewable energy, and energy efficiency; it is simply a complementary technology. Yet to the committee the glass is half empty, because it deems air capture too expensive.
Pessimism colors the report and frames its conclusions. For example, it states that a well-designed device capable of capturing the emissions of an average American would fit into a box with a square meter front and less than a meter depth. Most people own bigger refrigerators and drive much larger cars, suggesting that approaching air capture one-person-at-a-time is not overly daunting. Nonetheless, the committee wrings its hands over the size of a unit capturing the emissions of a big coal plant. Numbers like this can be made to look scary even for well-established technologies. For example, the line of cars and trucks produced in a year would be so long that the last car could not drive to the front of the line before the engine breaks down. Such an analysis helps to visualize the world’s car production, but it has no bearing on its feasibility.
The single and seemingly lethal criticism of the report is that the cost of air capture is and will always be too high. From this perspective, air capture may be a desirable tool, but is unavailable. The conclusion of the report rests on this single cost assessment rather than on first-principle objections. How reliable are such cost estimates, and does it make sense to base R&D policy on them?
Cost estimates of novel technologies have often been wrong. New technologies present moving targets. Costs can drop by orders of magnitude as technology develops. Examples are plentiful, with computer hardware leading the field. The cost of solar panels has dropped almost hundredfold since the 1950s. Efficiency improvements in gas turbines have moved them from a scientific curiosity in the 1930s to a mainstay in power generation and aviation today. Once sulfur emission trading was enacted, sulfur reductions at power plants proved within four years to be ten times cheaper than experts predicted.
Observed cost reductions as technologies mature pose a conundrum. Clearly, if one could design the lower cost version at the outset and thus correctly assess its cost, one would not be stuck with higher cost predecessors. Were assessments easy and accurate, subsequent progress would have to be minimal. But this goes against empirical evidence. Applying the logic of this study to solar energy, gas turbines or fuel cells would have stopped all these technologies dead in their tracks.
The APS study puts an upper limit on the cost of air capture. Experience shows that improvements in technology often reduce costs ten or twentyfold relative to awkward first tries. Just like wind and solar energy, air capture is within striking distance of economic feasibility. The APS study does not account for the dynamic of cost reduction via learning. The assessment methodology chosen stifles innovation and strangles new ideas before they have a chance to mature.
Today, capture of carbon dioxide is being developed for markets unrelated to climate management. Companies (one of us is involved with one company) are aiming to address this market demand at a price much lower than the APS estimate. These companies are reducing costs by developing innovative technologies rather than going through the academic exercise of designing a baseline plant without deviating from conventional practice. Nobody doubts that their technologies will have to get cheaper, but experience indicates that this is not too much to ask for and the report does not raise any scientific or technical issues that would preclude the necessary improvements.
Klaus S. Lackner, Maurice Ewing and J. Lamar Worzel Professor of Geophysics, Columbia University; Director, Lenfest Center for Sustainable Energy, The Earth Institute; Founder, Advisor, and Director, Kilimanjaro Energy, Inc.
Sarah Brennan, Columbia University; Associate Director, Lenfest Center for Sustainable Energy, The Earth Institute