Will carbon capture and storage (CCS) ever pay off?
For many years, we've been told that CCS systems and processes will allow us to reduce carbon emissions and stop global warming while continuing to use fossil fuels. CCS has been a key assumption of the "450 Scenario" in the International Energy Agency's annual energy outlook reports, in which the world can meet its energy needs while keeping atmospheric carbon concentration below 450 parts per million (ppm). If you read the news, you might even think CCS systems are well on their way to becoming a commercial reality.
But the fact is, they aren't. And current trends suggest they never will be.
The main reason is the cost. Finding good data on the cost of CCS is difficult, because it simply doesn't exist. No commercial-sized power plants equipped with the technology have been built yet. All of the cost data we have are estimates based on engineering designs, which are notorious for being much lower than reality.
Two things are clear: Since 2004, the cost of building a new power plant equipped with CCS has been escalating rapidly along with the costs of all construction commodities (like oil and steel). And those costs are now rising above the cost associated with power generation from renewables.
CCS is really a catch-all term for a variety of technologies and processes. The first part, carbon capture, usually imagines that devices will be integrated into the exhaust end of coal- or natural gas-fired power generation plants, removing some of the carbon dioxide (CO2) emissions. The second part, storage (aka sequestration), imagines that the captured CO2 will be compressed into a liquid, then either buried permanently underground or sold for use in industrial processes. For example, CO2 is used to make soft drinks fizzy, and to loosen up oil from old reservoirs as part of "enhanced oil recovery" (EOR) operations.
Various kinds of CCS have been imagined for capturing CO2 out of ambient air, as a way to deal with widely dispersed emissions from things like vehicles, but those ideas really belong in the category of "geo-engineering" and would be far more expensive and difficult than capturing concentrated CO2 straight out of a power plant. CCS has also been eyed for emissions from cement factories, blast furnaces for steel production, fertilizer factories, and other industrial facilities, but the main focus is on power plants. If we can't make CCS work for power plants, we probably can't make it work anywhere, so I am focusing on that here.
Cost estimates for CCS vary widely, by whether the capturing technology is to be added to an existing plant as a retrofit, or built into a new plant; by the type of power plant (such as a "supercritical" or "ultra supercritical" coal plant, or an "integrated gas combined cycle" plant); by the type of fuel (usually coal or natural gas); by when the carbon is captured (post-combustion, pre-combustion, or "oxy-fuel," in which coal is burned in pure oxygen rather than air to produce purer CO2 emissions); by the type of CO2 transport (pipeline or another method); and by the type of storage (porous underground saltwater formations, EOR projects, depleted oil and gas reservoirs, unmineable coal seams, and so on).
Given all the variations, one 2011 study from the Global CCS Institute, a group whose membership "covers more than 80 percent of the world's CO2 emissions from energy and industrial sources," offered a cost range of $38 to $107 per tonne of captured CO2, which isn't terribly helpful.
Many CCS cost studies cite data from 2009 and earlier, so they don't reflect current costs (after the commodity boom). The most recent cost data I was able to find was in a June 2012 paper from the U.S. Congressional Budget Office (CBO), "Federal Efforts to Reduce the Cost of Capturing and Storing Carbon Dioxide," which analyzed five engineering studies on building a new coal-fired power plant equipped with CCS.
In addition to ruling out retrofits, which are generally deemed to be cost-prohibitive, the CBO report focused exclusively on the post-combustion approach, "because that technology is the only one that is compatible with the most commonly used designs for electricity-generating plants."
The summary results are shown in the following table.
The right way to compare different power generation technologies is on a levelized cost of energy (LCOE) basis, which gives the average cost of producing electricity over the lifetime of a plant, including the costs of construction, financing and operation. Adjusted for inflation, the CBO chart shows that the LCOE of a new coal plant with CCS is about $90 to $150 per megawatt-hour (MWh) in 2013 dollars, or $0.09 to $0.15 per kilowatt-hour (kWh).
The main reason CCS is expensive is that it takes a lot of equipment to capture, purify (if the CO2 is to be sold), liquefy, transport and bury CO2. According to the CBO analysis, the average capital cost of a CCS-equipped coal plant would be 76 percent higher than a conventional plant and the LCOE for a CCS-equipped plant would average 76 percent more than for a conventional plant.
All that equipment consumes a lot of energy.
The U.S. Department of Energy estimates that the energy requirements of post-combustion carbon capture reduce the plant's efficiency by 20 to 30 percent. A 2007 study from MIT found that a CCS retrofit of an existing subcritical pulverized coal plant would reduce the plant's electrical output by more than 40 percent. Reduced energy output means higher prices for the energy that's not consumed by the plant itself.
How does the $0.09 to $0.15/kWh cost of a CCS coal plant compare with competing alternatives, such as conventional natural gas-fired power plants not equipped with CCS or renewables?
According to the Annual Energy Outlook 2012 from the U.S. Energy Information Administration (EIA), the cost of power from a conventional natural gas power plant without CCS is $0.0686/kWh, making it the cheapest clean(-ish) way to generate power. This is one reason why natural gas has been pushing coal off the grid, as I detailed last year in "Regulation and the decline of coal power."
Now consider the cost of utility-scale solar power, which has been dropping like a stone in recent years. That data can be hard to come by, since the price of power embedded in a power purchase agreement (PPA) isn't usually disclosed, but we do have numbers from a few new PPAs.
In the California Public Utility Commission's Renewable Portfolio Standards report for the first and second quarters of 2012, the weighted average price of approved contracts for 140 MW of distributed solar PV was less than $90/MWh ($0.09/kWh).
On Jan. 8, 2013, the Los Angeles Department of Water and Power reported, "Currently, the cost of solar energy through a power purchase agreement from a large solar power plant over 200 MW [megawatts] is about $0.095/kWh."
On Feb. 2, 2013, Greentech Media reported that the PPA price for the 50 MW Macho Springs solar project in New Mexico was $0.0579/kWh; after including the state production tax credit, the price was $0.0849/kWh.
Recent plants under contract in Michigan are coming in at about $0.091/kWh, according to Energy Fact Check.
Greentech Media solar analyst Scott Burger gave me a few final data points. The 23 MW SunEdison project in Hemet, Calif., came in at around $0.08/kWh, he estimated. Generally, his organization is seeing PPAs in the $0.07 to $0.09/kWh range.
At that price, a new coal plant is already a non-starter. Michigan's Public Service Commission estimates that a new coal plant would cost ratepayers around $0.133/kWh, and Bloomberg says the average price of power from a new coal plant is $0.128/kWh.
Meanwhile, the cost of CCS keeps going up, and the cost of solar keeps going down.
Massive subsidies needed
The fossil fuel industry and its partners in government realize that CCS isn't going to work without massive public sector investment.
In answer to my questions about Shell's Quest CCS project in the Alberta tar sands, a PR spokesperson with Edelman Digital Public Affairs in Washington, DC, told me that "current carbon prices do not support the economics of the project, which is why government support is required for the project to proceed." But Shell hopes that "as the costs of the technology come down and the price of carbon increases, CCS will become more economic."
"Without government support, CCS will not become economic," echoes ICO2N (the Integrated CO2 Network), a group of Canadian companies representing the coal and tar sands industries. They hope that with hundreds of millions of dollars of investment by the Canadian government into CCS research, CO2 sales for EOR will make the economics work in the meantime.
But the CBO study questions whether CCS could ever scale up to the point where it could stand on its own. It would take more than 200 gigawatts (GW) of new generating capacity equipped with CCS to reach that point, the report suggests, and "under current laws and policies, utilities are unlikely to build that much new generating capacity. . . or invest in adding CCS technology to much of their existing capacity for many decades."
All these studies agree that carbon emissions allowances must be priced much, much higher for CCS to become economically viable.
The Global CCS Institute estimates that CO2 would have to be priced at $23 to $92 per tonne. In a 2011 paper, the European Technology Platform for Zero Emission Fossil Fuel Power Plants found that coal-fired CCS power plants would be "close to becoming commercially viable" only at €35 ($46) per tonne of CO2.
The non-industry view is much less optimistic. A new research paper by Richard Middleton of Los Alamos National Laboratory and Adam Brandt of Stanford University estimates that "significant capture and storage occurs only above $110/tonne CO2 in our simulations."
Unfortunately, global carbon markets aren't pricing CO2 emissions anywhere near these levels.
The current price for carbon in the EU Emissions Trading System is just €5 per tonne. The new California CO2 allowance is $13.62 per ton, and the proposed U.S. carbon tax would be about $20 per ton.
With the future of carbon policy so uncertain in the United States and abroad, it's not at all clear that carbon prices will rise to the point where investing in CCS makes sense.
At the World Future Energy Summit in Abu Dhabi in January, I asked Maria van der Hoeven, Executive Director of the IEA, what would happen to their 450 Scenario if CCS didn't work out. "We need at least 10 large demonstration projects to scale up from pilots to commercial installations," she acknowledged. "But nevertheless you can see it has slipped down the political agenda. It's true. It's costly, there are too many other issues at stake, people are afraid of it sometimes because they don't want to have it stored underground, because it's 'terrible,' it's something they are afraid of. So in some way or the other at this moment CCS doesn't fly. It doesn't."
Cancellations and delays
A handful of demonstration-scale CCS projects are in progress, with most of them hoping to become operational over the next several years.
Aside from the Shell project, a December 2012 update by Politifact lists six more that are well under way, including the roughly $200 million Illinois Industrial Carbon Capture and Storage project in Decatur, Ill.; the $2.88 billion Kemper Integrated Gasification Combined Cycle plant in Kemper County, Miss.; the $2.5 billion Texas Clean Energy Project in Penwell, Texas; the $2.8 billion Hydrogen Energy California Project in Bakersfield, Calif.; and the $1.65 billion FutureGen 2.0 project in Meredosia, Ill., a revival of the original (and much-ballyhooed) FutureGen plant that was scotched in 2011 over escalating costs.
But the recent cancellation of several large projects doesn't bode well.
The $278 million Swan Hills Synfuels project and the $1.4 billion TransAlta project, both in Canada, have been scrapped due to cost and cheap natural gas. American Electric Power's $668 million Mountaineer Station in New Haven, West Va., which used gasified coal, was recently cancelled due to the "current uncertainty of U.S. climate policy and the continued weak economy." FutureGen 2.0 is reportedly months behind schedule. SourceWatch lists many other delayed or cancelled projects around the world.
These may sound like big-ticket items, and they are. One wonders if that money wouldn't be better spent on building already-cheaper renewables, and it would. But the coal companies and tar sands operators are fighting for their lives, and spending a few billion here or there on a saving grace like CCS probably seems like a relatively small price to pay.
I am doubtful that CCS will ever pay. The cost curves for renewable power suggest that solar and wind will undercut the cost of CCS on new or retrofitted gas and coal plants before 2020, when CCS proponents hope that it will become economically viable.
One new analysis by UBS found that the LCOE from unsubsidized new rooftop solar photovoltaic in the United States is $0.24/kWh, making it cheaper than grid power in 11 European countries. And U.S. utility-scale solar power priced at less than $0.09/kWh (and falling) is very tough to beat.
Unless the world decides to price carbon aggressively before those renewable cost curves fall much lower, CCS looks like a dead man walking.
* Correction: Several cost/kWh figures were incorrectly stated in the original version of this article. They have been corrected. I regret the errors.