The One That Got Away

This is not a fishing story, nor one of unrequited love... It is a story of a scientific journal article that made its way through peer review and was published when it should not have. The article appeared in Environmental Science and Technology (full text available here, and supporting information here), and claims to shed new light concerning the efficiency penalty that coal-fired power plants incur when they install Carbon Capture & Storage (CCS) technology. The authors report higher values for this penalty, and state that this has competitiveness implications against other low carbon technologies, and even natural gas combined cycle plants without CCS. Their analysis however is flawed on several counts.

The first puzzling assumption is the choice of sources for the plant and capture system efficiency assumptions. Some of these date back to the early 1990s, and correspond to sub-critical coal plants and less advanced CO2 scrubbing systems than are available today. It is true that these systems are less efficient, but it is commonly understood that the units that are candidates for CCS today are the more modern, supercritical and ultrasupercritical coal plants. These, of course, would run using the CO2 scrubbers that are available today, and which feature lower energy expenditures (for a comparison with recent estimates, see NETL's report). The choice of outdated values is misplaced.

The treatment devoted to CO2 purification under one of the configurations considered is also of limited applicability. The notion here is that in order to remove impurities from the final CO2 stream, additional energy will need to be expended and in some cases dedicated equipment used. Modern systems, however, report purities approaching 100%. Although academically valid, this train of thought does not extend far into real life. It is hard to imagine why today's commercial operators would prefer a system that achieves lower performance and necessitates additional capital and operational expenditures over one that outperforms it and is backed by commercial warranties.

Another assumption that is hard to justify is that plants fitted with CCS will continue to generate the same amount of grid electricity as before ("All scenarios assume that an additional boiler will be needed to meet the steam demand not met by the turbine or recovered heat"). In order for that to happen, additional investment will be needed, essentially for a supplemental power plant. For a retrofit situation on a connected grid with many other supply options, there is no need to maintain the output constant (for a new build, the issue is purely theoretical). The authors claim that broad application of the technology will decrease the installed of capacity on the grid unless the power is supplemented, which may seem logical, but in fact is simplistic and unrealistic. Choices about what generation to build are made on an integrated basis, considering many factors such as capital costs and fuel prices, the supply-demand balance, competing generation options, and government policies and regulations. For example, a coal plant without CCS may face closure due to emission regulations if it does not reduce its carbon emissions. The owner of such a plant would consider choices that would likely include plain retirement, retrofit with CCS, and replacement with a plant using a different fuel or technology. For such an operator, a slight loss in output after compared to not operating at all may be a straightforward choice.

More generally, the article makes misleading cost comparisons. Even though the costs under fairly bizarre assumptions are outlined, the benefits are not. Installing pollution control equipment does cost money, but that says nothing about the costs that would be incurred if the pollution were not abated, or what the alternatives to CCS would be. Weighing those factors would involve considering laws and regulations that limit carbon emissions, the presence and level of a carbon price, as well as competing technologies (including their price, intermittency, and availability). For example, in a market where polluters have to pay for their carbon emissions and intermittent renewable energy sources make up a considerable portion of generation, retrofitting a fairly modern coal plant with CCS might be preferable to scrapping the plant, especially if it can ramp its output in a way that facilitates further deployment of renewable sources. Ignoring the whole suite of considerations when evaluating a generation technology is neither realistic nor useful.

All in all, the article contains some logic, but predominantly manages to confuse and mislead by looking at part of the picture, using antiquated assumptions and configurations that do not correspond to the real world. It is also an outlier in terms of the costs it reports. If you are looking for a more reliable comparison of costs of generation with and without CCS, look here instead.