Skip to Content

Using CO2 to Extract Geothermal Energy

Carbon dioxide captured from power plants could make geothermal energy more practical.
November 16, 2009

Carbon dioxide generated by power plants may find a second life as a working fluid to help recover geothermal heat from kilometers underground. Such a system would not only capture the carbon dioxide and keep it out of the atmosphere, it would also be a cost-effective way to use the greenhouse gas to generate new power.

Hot air: The Soultz-sous-Fôrets geothermal plant in Alsace, France, pumps water into fractured rock to extract heat and thus generate electricity. Researchers backed by $16 million in federal stimulus funds seek to prove that such geothermal plants could generate 50 percent more heat by cycling carbon dioxide underground instead.

Backers of this as-yet-unproven concept secured a big endorsement and much-needed cash with the U.S. Department of Energy’s recent award of $338 million in federal stimulus funds for geothermal energy research. Some $16 million of the funds will be shared by nine carbon dioxide-related projects led by Lawrence Berkeley National Laboratory and other national labs, Sunnyvale, CA-based combinatorial chemistry firm Symyx Technologies, and several U.S. universities.

The idea: Carbon dioxide that’s cycled through hot regions kilometers underground can efficiently bring heat to the surface, where it can be used to generate electricity. The likelihood is that the process would leave lots of carbon dioxide underground, and thus out of the atmosphere, according to Symyx project leader and materials scientist Miroslav Petro. “You’re sequestering CO₂ and at the same time generating power from it.”

The concept was first proposed as a way to improve systems that pump water deep underground to fracture hot rocks, then bring the heated water up via a second well to generate power, and then cycle the water back down. The technology has been thwarted to date because it’s so difficult to fracture rock to get at the geothermal heat and sustain its flow. The European Union’s Soultz-sous-Fôrets project in Alsace, France, the most advanced such project worldwide, has taken 20 years to reach just 1.5 megawatts of power generation (enough to supply roughly 1,500 homes). And the process has antagonized nearby communities because of the small earthquakes sparked by the aggressive fracturing required.

In 2000, Los Alamos National Laboratory physicist Donald Brown proposed replacing water with supercritical carbon dioxide, a pressurized form that is part gas, part liquid. Supercritical CO2 is less viscous than water and thus should flow more freely through rock. Brown noted that a siphoning effect should help cycle the carbon dioxide, thanks to the density difference between the supercritical CO2 pumped down and the hotter gas coming up, slashing power losses from pumping fluid. Plus, Brown argued, instead of using precious fresh water resources, a carbon dioxide-based project could sequester the equivalent of 70 years worth of CO2 emissions from a 500 megawatt coal power plant.

Six years later, Lawrence Berkeley hydrogeologist Karsten Pruess performed the first detailed modeling of the technology. Pruess projected that a project such as Soultz-sous-Fôrets could produce approximately 50 percent more heat with carbon dioxide than with water. Most of the DOE-funded projects seek to test Pruess’s optimism.

The most important question, according to Petro, is how supercritical carbon dioxide will interact with rock and minerals. Supercritical CO2 also has a particularly complex relationship with water. On its own, supercritical CO2 is not expected to dissolve minerals from rocks - a major problem encountered in the water-based approach. But, says Petro, adding a fraction of water to supercritical CO2 could form a super-dissolving “acidic soda water.”

At least one developer, meanwhile, is seeking financing for a field demonstration of carbon dioxide-based geothermal. In September, Salt Lake City-based geothermal developer GreenFire Energy announced a joint venture with small oil developer, Enhanced Oil Resources, to build a two-megawatt CO2-based demonstration plant near the Arizona-New Mexico border. The companies propose to commence drilling wells in 2010 to access hot rock underlying a natural underground carbon dioxide reservoir. They project that the location could yield enough heat to generate up to 800 megawatts of power and, in the process, could absorb much of the carbon dioxide generated by the six large coal-fired power plants in the region.

Instead of adding CO2 to geothermal energy plans, the University of Minnesota’s geofluids research group, one of the DOE’s awardees, proposes to add geothermal energy extraction to existing plans for carbon capture and storage. Martin Saar, the University of Minnesota geophysicist who leads the geofluids group, says this scheme will yield additional value out of operations that already pump supercritical CO2 into deep saline aquifers for storage, or into oil and gas formations to accelerate production. That carbon dioxide will pick up heat from the surrounding rocks, says Saar, so why not circulate some of it to generate power? This eliminates the need to fracture rocks. And it takes advantage of existing equipment and drilled wells, thus reducing the cost of the geothermal plant.

Saar is researching how supercritical CO2 interacts with rock, minerals, and water. Understanding the latter is critical to the Minnesota scheme, since carbon dioxide injected into a saline aquifer will mix with water. However, Saar says that may be less of a problem than it appears, because large volumes of CO2 injected into a saline aquifer should separate to form a distinct layer: “Supercritical CO2 is actually less dense than the brine, so in an aquifer it will rise and pool underneath the cap rock.”

If the lab work confirms that and other predictions, Saar says, they could be testing CO2 geothermal in the field in as few as three years.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.