One way to combat global climate change is to directly capture carbon dioxide, the main greenhouse gas, as it is being emitted, and store it safely. But methods of carbon dioxide sequestration, notably, pumping the gas into underground geologic structures such as exhausted oil reservoirs, are not practical in many areas, and raise fears that the stored carbon dioxide will escape.

Now researchers at Harvard University and Columbia University have proposed a new method for trapping nearly limitless amounts of carbon dioxide – a technique they say will be secure, as well as a practical option for areas located far from underground reservoirs.

The researchers, in an article posted online this week in the Proceedings of the National Academy of Sciences, propose that carbon dioxide be pumped into the porous sediment a few hundred meters into the sea floor in deep parts of the ocean (greater than 3,000 meters deep), in what one of the researchers, Dan Schrag, professor of geochemistry at Harvard, calls “a fairly simple, permanent solution.”

The key was finding a “sweet spot,” where the pressure and temperature of the surrounding environment make carbon dioxide more dense than surrounding fluids, thereby trapping it in place. This situation occurs at the bottom of the ocean because of a combination of high pressure and low temperatures – a fact others have also noted in proposals to store carbon dioxide in deep parts of the ocean.

But such injections would kill ocean life, and, unless sequestered in deep trenches, the carbon dioxide could be carried by currents to shallow areas, where it could reenter the atmosphere.

The researchers’ insight was that injections into the sea floor could take advantage of the pressure and temperature of the ocean, while avoiding the negative side effects of earlier proposals. The carbon dioxide, in liquid form, would be brought to the sequestration site by ship or pipeline, and piped into the sea floor with equipment like that used by the oil industry for drilling deep-sea wells. Once beneath the sea floor, the carbon dioxide would interact with the surrounding fluids and produce hydrate ice crystals, which would plug the rock pores, serving as a secondary cap on the carbon dioxide. Over hundreds of years, the carbon dioxide would dissolve in the surrounding water, and then would only have the potential of leaking out by diffusion, a slow process that would take millions of years, the researchers say. Within the next five years they hope to run a large-scale field test of this new approach.

As worries over the impact of carbon dioxide emissions on global climate change soar, researchers are increasingly searching for ways to rid the atmosphere of the greenhouse gas. But, so far, industrial-scale projects have been limited. Notable among them, oil giant BP and GE recently announced a project to build power plants in Scotland and California that derive hydrogen from fossil fuels and sequester the carbon dioxide by-product. And Statoil in Stavanger, Norway, separates excess carbon dioxide in natural gas extracted in its North Sea mining operations and injects it into underground reservoirs. While these reservoirs are under the ocean, they are under too little water, and too deep below the sea floor to use the mechanisms described by Schrag and his colleagues.

The most prominent storage method nowadays (Statoil’s project is an example) involves depositing carbon dioxide in underground geologic formations such as depleted oil fields. Here the dynamics between carbon dioxide and surrounding fluids are different than those in the sea floor, where the ocean keeps the fluids cool. Rather, these formations are heated by the earth’s crust, and the high temperature make the carbon dioxide less dense than the water in the surrounding rock, making it prone to rising to the surface, Harvard’s Schrag says.

Sea-floor injections also offer an immense amount of storage capacity. If all the known geologic reservoirs for conventional storage were useable, they could store all the carbon dioxide currently produced each year, and continue doing so for 80 years at current emission rates. In contrast, sea-floor storage around the United States alone could store thousands of years worth of U.S. carbon dioxide production, the researchers estimate.

Robert Socolow, co-director, of Princeton University’s Carbon Mitigation Initiative, notes that the sea-floor injection method has the advantage of being intrinsically secure. But he says that well-mapped reservoirs, away from seismically active areas, can be effectively capped to prevent the greenhouse gas from escaping, and therefore these methods will continue to have a place.

Indeed, the costs for the new sea-floor method will vary, Schrag says, but will probably be slightly more than for land-based storage. It could, however, be more economical for areas near the ocean, especially those far from a known geological reservoir. “If you’re sitting right next to a big basin, it’s probably slightly more expensive. If you’re in New Jersey, and you have to pump the carbon dioxide 300 miles to get to such a basin, then I would say no.” He notes that the cost for any method of large-scale sequestration is still unclear.

“The need for robust, potentially inexpensive carbon sequestration schemes is enormous,” says Nathan Lewis, a professor of chemistry at Caltech. While it still requires more experimental validation, he says Schrag’s work “is potentially very important. It ought to be considered very seriously.”