Greenhouse gas dissolves in water rather than becoming locked into minerals.
In a paper in this week's Nature1, UK researchers have firmed up our notions of what happens to carbon dioxide buried deep underground over millions of years. They find that in sedimentary rock, it's unlikely to become locked into solid minerals — instead, it will either dissolve in water or stay trapped in a CO2 bubble. Nature News finds out what this means for the prospects of long-term carbon storage.
Didn't we already know that CO 2 can be buried safely?
Yes. Oil companies routinely inject CO2 down wells to help recover crude oil. In test sites where this injection is monitored, no leakage has been seen over decades. And experience with old oilfields shows that reservoirs with the right geology can lock up their CO2 for millions of years2. But some uncertainty remains over what happens to CO2 underground in the long term.
What did the researchers do?
To work out what's most likely to happen — in sedimentary rock, at least — Stuart Gilfillan, of the University of Edinburgh, UK, and his colleagues tracked CO2 in nine natural underground reservoirs across the United States, China and Hungary. The youngest of the reservoirs, the Bravo dome in New Mexico, was filled with CO2 around 10,000 years ago, whereas the gas in the oldest, McElmo in Colorado, was formed 40–70 million years ago.
How did they track the fate of the gas?
The researchers compared data on various isotopes sampled from the bubble of gas trapped in the reservoir and found a surprising link between the levels of CO2 and noble gases.
The ratio of CO2 to helium-3 is already known to reveal whether any CO2 has been lost from the reservoir after both gases arrived together from the Earth's mantle. The researchers found that the more CO2 was lost, the higher the concentration of noble gases such as helium-4 and neon-20. The researchers argue that this link shows how much CO2 has dissolved in water, kicking out the noble gases from solution into the gas bubble.
When CO2 dissolves in water it forms carbonic acid. And that in turn may transform into carbonate minerals. The team knew that the heavier isotope of carbon, carbon-13, prefers to precipitate in minerals over the lighter carbon-12. And using the 13C/12C ratio, combined with the isotope data showing how much CO2 had been lost from the original gas bubble, they showed that most of the lost CO2 had dissolved in water and not formed minerals.
Is it a surprise that CO 2 didn't get trapped as minerals?
Not really: this study is mainly confirmation of theoretical suspicions.
"Geochemists worldwide have been excited about trapping CO2 in minerals, but I think we've known for a long time that it is small stuff for most sedimentary rocks," says Sue Hovorka, a geochemist at the University of Texas, Austin, who has worked on pilot experiments injecting CO2 into saline aquifers. For rocks such as basalts, it could be a different story, she adds.
Doesn't the water leak out from the reservoir — taking the dissolved CO 2 with it?
When CO2 dissolves in water, it creates a denser liquid that sinks. This is far less likely to escape than if the CO2 remains gaseous, when it must be contained by an impermeable 'cap' of rock.
When gas is deliberately captured and pumped underground, it is injected deep as a supercritical fluid — a state halfway between a liquid and a gas. "The major risk with CO2 storage occurs when CO2 is in the buoyant phase (gas or supercritical liquid)," says David Keith, who works on carbon capture and storage at the University of Calgary in Canada. "As soon as the CO2 is dissolved, the risk is near zero because there is no longer a strong driving force to bring CO2 to the surface. In my view, it makes little difference whether the CO2 simply remains dissolved or is mineralized." He and other researchers are working on actively accelerating the rate at which CO2 dissolves in water3.
But CO2 may still escape, points out Chris Ballentine of the University of Manchester, UK, who also took part in the work published in Nature1. The injection of a vast amount of CO2 may help squeeze out water from the reservoirs, or create acidic fluids that eat holes in the rock.
How do we ensure that doesn't happen?
Simply put, this work reminds us that we need to survey the hydrogeology of any proposed storage site carefully, Ballentine says. Admittedly, CO2 injection sites trialled so far — such as Norway's Sleipner field project in the North Sea, which has been working since 1996 — have shown little risk of CO2 leakage through water flow.
But Curt Oldenburg, head of geologic carbon sequestration at Lawrence Berkeley National Laboratory in California, points out that these trials are still extremely small-scale: "A single power plant can produce 8 megatonnes of CO2 a year, and these projects are injecting 1 megatonne of CO2 a year," he says. Nobody knows if large-scale CO2 injection might disrupt water flows as feared.
And Eric Oelkers, who works on carbon sequestration in Toulouse at France's national research agency, CNRS, warns that the knowledge obtained by geologists is not being used in commercial storage plans. "There are a huge number of projects, and some people are just pumping CO2 down a hole and hoping it stays there. Somebody's going to goof up," he sighs.
Gilfillan, S. M. et al. Nature 458, 614–618 (2009).
Lu, J., Wilkinson, M., Haszeldine, R. S. M. & Fallick, A. E. Geology 37, 35–38 (2009).
Leonenko, Y. & Keith, D. W. Environ. Sci. Technol. 42, 2742–2747 (2008).