24 August 2012
Pour yourself a flute of champagne and bubbles of carbon dioxide will rise, bursting when they reach the surface. A question troubling some researchers is whether carbon dioxide behaves similarly underground. They’d like to know whether the gas, stored beneath the surface to cut down on greenhouse emissions, would make its way through cracks in the rock, and bubble out into the atmosphere.
To the surprise of engineers investigating this question, their experiments have found that an abundant underground fluid – a salty, water-based material called brine – will actually impede bubble rise.
Energy companies pump thousands of gallons of carbon dioxide underground each year as part of an effort, known as carbon sequestration, to keep the greenhouse gas from entering the atmosphere. Scientists don’t have any indication that the carbon dioxide injected into the ground can escape – but some are concerned about potential leakage as the number of these carbon sequestration projects grows.
A pair of environmental engineers at the University of Virginia set out to understand how a leak might occur. Instead, they were surprised to find that carbon dioxide bubbles, when mixed with brine, increased the brine’s viscosity, or resistance to flow. This makes it harder for other bubbles to leak through the mixture.
“The biggest surprise was that these bubbles would change the long-term viscosity of the fluids,” said Andres Clarens, a co-author of the study. “In some cases the effective viscosity doubled or tripled from the initial value.” A similar transition – in terms of viscosity, not composition – would be to change a glass of water into a glass of mercury or whole milk.
The two scientists found that carbon dioxide bubbles are fairly stable in brine, so a number of them may accumulate over time, increasing the brine’s viscosity. Basically, bubbly brine will have a higher viscosity than bubble-free brine.
Since the bubbly brine is more viscous, new bubbles just entering the fluid will rise slower, Clarens said. How much slower depends, in part on the size of the bubbles and the viscosity of the brine. But, the researchers suggest the bubbles could be slowed by more than 75 percent of their original upward velocity.
The study will publish today in Water Resources Research, a journal of the American Geophysical Union.
With growing attention on commissions to inject carbon dioxide underground, Clarens and his colleague Shibo Wang wanted to chronicle a bubble’s possible journey to the surface. Because brine is abundant in the rock types at or near injection sites, they decided to investigate the interactions between brine and carbon dioxide gas.
Because it is tricky to perform experiments deep under the ground, the researchers conducted their experiments in the lab. They used synthetic brine that reflects the chemistry of water in the pores of rock. The researchers put the solutions into a rheometer – a device that measures viscosity – with a pressure vessel capable of operating at 15 megapascals to simulate conditions in the deep subsurface. The instrument then spun the carbon dioxide-brine mixture at high speeds to create the kind of forces a bubble would face while flowing through brine-filled pores in the ground.
Until now, brine has not been considered a potential obstacle for carbon dioxide rise. Instead, scientists thought that layers of dense rock – such as clay or shale – would trap carbon dioxide. Ideally, these clay or shale boundary layers would be flawless, Clarens said. But small cracks or fractures are likely present, which would act as an escape highway for large volumes of liquid or gaseous carbon dioxide.
Jan Nordbotten at the University of Bergen in Norway – although uncertain about the role of bubbles specifically – agrees that interactions between carbon dioxide and brine are important for understanding potential leakage.
“Predicting the migration of carbon dioxide requires an understanding of a broad suite of physical relationships,” said Nordbotten, who was not involved with the research but studies potential leaks from carbon storage reservoirs using mathematical models. “The impact of carbon dioxide on the viscosity of brine is of course one of these relationships, and as such it deserves attention.”
Clarens said that this new study is just the beginning of understanding the many different drivers behind a potential leak.
“Eventually this type of understanding is going to be really important for setting policy,” Clarens said. “But we’re still in the early days.”
-Jessica Orwig, AGU science writing intern