Science 1663

Containing Carbon Dioxide

During the Cold War, Los Alamos geologists proved their ingenuity by successfully containing nuclear tests underground. The latest global threat, climate change, is challenging them again. Can they successfully contain carbon dioxide, the most troublesome greenhouse gas, deep beneath the earth?

by Anthony Mancino

The average American family of four puts about three tons of garbage per year by the curb, but because we burn fossil fuels for electricity, heat, and transportation, that same family is annually responsible for dumping about 80 tons of carbon dioxide (CO₂) into the atmosphere—CO₂ that contributes to global climate change.

Projections show that coal-fueled electricity generation will increase 74 percent by 2030 and that coal will soon surpass oil to become the largest source of energy-related CO₂ emissions. Geologic sequestration may offer a way to mitigate the climatic impacts of this trend.

Some believe we should just stop using fossil fuels, but that won't happen in the near future unless we're willing to cripple our economy and keep the developing world in poverty. Eighty-six percent of the world's energy comes from fossil fuels, and projections show energy demand and fossil fuel use rising dramatically. China, India, and the United States are planning to add 850 new coal-fired power plants to the 2,100 worldwide that currently chug out one-third of the world's human-generated CO₂ emissions.

We need a way to cut those emissions and slow climate change now without precipitously abandoning the abundant and affordable coal resources that fuel the majority of power plants. The solution may lie in how we handle garbage. We bury it. Burying CO₂ might be the best short-term option for slowing climate change.

It's called geologic sequestration. Instead of releasing CO₂ into the atmosphere, we would capture it at its industrial sources, turn it into a fluid, and inject it into deep geological formations. (See "The Basics of Geologic Storage") There is a neat symmetry to that. Carbon we've released from deep underground in the form of oil, gas, and coal can be returned to subterranean storage.

The global need to slow climate change while still using coal has led Los Alamos scientists to the small town of Snyder, Texas.

A Ready Testing Ground

The drive to Snyder cuts through the expansive flat and treeless plains of west Texas, punctuated here and there by cotton fields and oil wells. Beyond the many wells, on a ridge in the distance, sit several wind turbines, a symbol of our energy situation, in which fossil fuels are still king but alternatives are on the horizon.

A minute fraction of the thousands of CO2 injection and oil recovery wells (bright spots connected to dirt access roads) that dot the landscape near Snyder, Texas. The edge of Snyder, where people have lived atop CO₂ injection operations for 35 years, is visible at lower right.

Originally a buffalo-hunting settlement, Snyder became an oil boomtown in 1948. The boom was over by late 1951, but in the 1970s, Snyder still managed to extract more oil than any other area in Texas. Producers there had developed a method of injecting CO₂ into the oil-bearing rock to force out hard-to-get reserves. This method, called "enhanced oil recovery," or "EOR," is now used worldwide.

When Los Alamos scientists wanted to understand the effects of large-scale geologic sequestration, they found willing partners in the operators of the SACROC oil field near Snyder, where EOR has been performed with CO₂ since 1972.

SACROC is a large, successful operation, but its numbers reveal the challenge of geologic sequestration. In SACROC's entire 35 years of injecting CO₂, only 70 million metric tons of CO₂ have accumulated in its reservoirs. The world's coal-fired power plants alone emit about 11 billion metric tons of CO₂ per year.

"We'll need to sequester amounts of CO₂ that are orders of magnitude larger than anything attempted in EOR," explains Phil Stauffer, a Los Alamos hydrogeologist.

"With EOR," adds chemical engineer Hari Viswanathan, "we have around three decades of history, but we need to consider possibly hundreds of years."

Increased Scale Brings Increased Risk

Increasing the time and volume of sequestration multiplies the potential risks, so Viswanathan and Stauffer are working as part of a team of scientists led by Rajesh Pawar to develop a comprehensive risk assessment framework and computer model to evaluate potential problems associated with geologic sequestration. The model is called CO₂-PENS, with "PENS" standing for "predicting engineered natural systems."

The Basics of Geologic Storage The geologic formations that would store CO₂ have many porous and nonporous layers. The sponge-like porous ones would hold the CO₂. Most of these porous layers already have fluid in them, usually saline water and in some cases, oil. These fluids will have to move to make space for the CO₂, which is how residual oil is mobilized in EOR operations.

For sequestration, the CO₂ is first compressed until the combined heat and pressure make it "supercritical," a state in which it displays both gas and fluid properties. When injected, this supercritical fluid diffuses like a gas into the porous rock but takes up less space than a gas because of its fluid-like density. The reservoir layers must be deep enough (under high-enough pressure) to maintain the supercritical state.

Supercritical CO₂ is buoyant and will rise above the other fluids. If it rises high enough (above a depth of 2,600 feet), it will return to a gaseous state, expand, and slip through any available escape route, indicated in the figure by question marks. To keep this from happening, a sequestration site must have an impermeable cap rock, past which the CO₂ cannot rise. For eons, such trapping mechanisms have worked efficiently to contain natural gas, oil, water, and volcanic CO₂, but even impermeable cap rocks are typically cracked, and the CO₂ could seep up faults or fractures in that ceiling.

Los Alamos scientists learned to predict the movement of fluids in the subsurface during nuclear testing and while planning for nuclear waste repositories. To contain the effects of underground nuclear tests or store radioactive waste beneath the earth for 10,000 years, they had to understand and predict how materials move and react in complex geological systems. While CO₂ sequestration is far safer than these activities, it still poses potential safety and economic risks.

You can't be sure CO₂ will stay where you put it because the subsurface is home to a staggering number of chemical reactions, physical processes, and potential escape pathways. But with the proper scientific understanding, you can predict how the CO₂ will behave, quantify the uncertainties in those predictions, and identify the experiments and field observations needed to reduce that uncertainty.

"There's uncertainty in how fast CO₂ can be injected into a given hole," says Stauffer, "how fast it will move through porous rock, how fast it will mineralize, and how fast it might leak up through different pathways. We're trying to reduce those uncertainties so decision makers can act on a known level of risk."

Scientists can already make fairly good predictions for individual subsurface processes because data and computer models already exist for them. What's been missing is a way to pull them together into an accurate picture of the whole complex system. That's the gap CO₂-PENS fills. It is an overarching program that taps a shared cyber library in which separate process models are dynamically linked. It goes beyond the individual results of each model to reveal how all the many processes affect each other.

For example, if a porous-rock model shows a plume of CO₂ and brine moving toward old cement-plugged oil wells, CO₂-PENS could pass those results to another model that would calculate the likelihood that the plume will remain contained. If a leak seemed likely to occur near a freshwater aquifer, CO₂-PENS would update a groundwater-impact model to determine the possible effects on water chemistry. If an atmospheric leak were possible, CO₂-PENS would access data from an atmospheric-circulation model to show how CO₂ would locally concentrate or dissipate.

Rajesh Pawar (center), project leader for geologic sequestration, works with Phil Stauffer to develop the computer code for CO₂-PENS, a science-based prediction tool designed to ensure safe and effective geologic sequestration operations.

CO₂-PENS also quantifies the uncertainty in its predictions and indicates if it needs more information—more experiments or more sensors at the site—to decrease that uncertainty. The program will initially be used to select the best sequestration sites, but once a sequestration operation is underway, it will learn from site-generated data and sharpen its predictive accuracy.

It can also help analyze economic risks. For one site, CO₂-PENS unexpectedly predicted that drilling wells 1 kilometer deep would be less economical than drilling them 3 kilometers deep.

"At 1 kilometer, you needed 80 wells to handle the CO₂ volume from a power plant," Stauffer explains. "At 3 kilometers, you needed only 10 wells because the increased temperature lowered the viscosity of the CO₂, and the higher injection pressure allowed it to slide more easily into the reservoir."

Balancing Competing Factors

A core sample extracted from a plugged wellbore in a geologic formation where CO₂ has been injected for over 30 years. The detail shows a polished section of the core. The orange carbonated zone shows how CO₂ affected the cement next to the formation's shale cap rock.

"Sequestration will be a balancing act between risks and benefits," says Viswanathan. "An existing oil field might seem an economical choice for a site because existing wells would lower drilling costs, but there's some question about the containment properties of old wells. It may be cheaper in the long term to drill new wells into a saline reservoir where old wells are not piercing the cap rock." CO₂-PENS will help manage that balancing act.

Oil companies typically plug abandoned wells with cement, and since a combination of CO₂ and brine is acidic, there is a risk that the CO₂-brine solution would eventually eat its way through the plug. Initially, not enough was known to quantify the risk.

To study the question, Los Alamos researchers recovered cement samples from a SACROC well where cement had been exposed to CO₂ and brine for decades. Laboratory analysis, led by Bill Carey, showed that the cement was still an effective barrier after 30 years. CO₂ did migrate along the well hole's casing and along the nonporous rock above the oil reservoir (the cap rock), but minerals had filled the gaps over time. The results give cause for optimism but also indicate further attention is needed, given the long time scales involved.

The CO₂ can still cause trouble without reaching the surface. It needs only to escape beyond the cap rock to find pathways through porous layers above. If that happens, it could affect freshwater resources by introducing brine or mobilizing inorganic contaminants, such as metals. It could also seep into nearby hydrocarbon resources, such as natural gas deposits, owned by other companies, resulting in legal and financial penalties. So there are risks, but with tools like CO₂-PENS, the risks are manageable and pose far less danger than letting atmospheric CO₂ accumulate unchecked.

Los Alamos researchers are working on other unknowns in the risk assessment framework as well. In one project, scientists are developing complex parallel computer codes to understand exactly how CO₂ and brine mix and what mineral reactions result. Those processes will greatly affect the flow of a CO₂ plume over time.

Geochemist Bill Carey analyzes a cement sample from a CO₂-enhanced oil recovery operation.

Researchers are also studying an area in Canada where hundreds of thousands of wells have been exposed to acid gas (CO₂ and hydrogen sulfide). Data from so many wells hold a wealth of information relevant to CO₂ sequestration and should generate statistics invaluable to CO₂-PENS.

Yet another important piece is quantification of CO₂ flux at the surface, where CO₂ constantly cycles in and out of the terrestrial ecosystem through plants, soils, and oceans. You have to characterize the natural variations and inventories of CO₂ before you can tell if new CO₂ from a sequestration operation is seeping into the mix. Julianna Fessenden-Rahn and other Los Alamos researchers are obtaining baseline measurements and using analytical methods to determine if CO₂ is of biological or industrial origin. The data they collect will increase the safety and effectiveness of sequestration operations and increase the predictive certainty of CO₂-PENS.

Hopeful Signs

All the initial signs indicate that geologic sequestration can be safe and effective, but the challenge of implementing it on a scale grand enough to affect climate change is somewhat daunting. The greenhouse gas engine driving global warming is like an onrushing freight train. It would take decades to stop even if we hit the brakes right now. And we're not ready to hit the brakes. Existing coal plants are not equipped to separate CO₂ from their exhaust streams (a nontrivial technical problem that Los Alamos scientists are also working to solve). New coal plants won't include such a capability because there is no regulatory pressure to do so and because it would increase the price of electricity.

Julianna Fessenden-Rahn works on an eddy covariance tower used to measure CO₂ flux at the land-atmosphere interface.

So can we implement geologic sequestration in time to make a difference? Stauffer answers with a blunt rhetorical question: "If your car is heading toward a tree and you know you're going to hit it, do you still put on the brakes?"

"We've got to try," adds Viswanathan. "Even if things start getting bad, they won't be as bad if we do this."

George Guthrie, program director for the Laboratory's Fossil Energy and Environment Program Office, is a bit more optimistic. "Geologic sequestration is being implemented today, and it will take off. The momentum in the last 2 or 3 years is just amazing, and there's a lot of optimism that it will be able to make a dent. Internationally, you see large-scale field efforts proposed and deployed. So far these operations involve natural gas production instead of coal plants because the ability to separate CO₂ already exists as part of the gas-purification process. There is still a lot of economic uncertainty for the electric power industry's coal-fired plants, but that's why we're doing all these pilot studies and experiments."

In the end, the right economics and regulatory policies will drive industry's willingness to implement geologic sequestration. When that time comes, Los Alamos researchers plan to be equipped with the scientific and engineering knowledge needed to do it right.

• Fossil Energy & Environment
• Energy Security
• Carbon Sequestration