Injecting tiny proteins into the hunt for 'clean coal'

As big engineering fixes go, "clean coal" has proved an elusive concept. Carbon capture projects remain experimental, expensive and energy intensive. But working with some of the tiniest things in nature, scientists are engineering proteins found in living things to trap carbon dioxide from coal-fired power plants.

"Biomimetic design" is the idea of using nature as a template to create new technologies. Trees are among nature's most efficient carbon sequestration systems. They trap carbon dioxide and convert it to glucose, placing it in a form in which it stays stable for geologically significant durations.

But at the biochemical level, they are still too slow, according to Michael Drummond, a scientist at the University of North Texas who is trying to identify new "carbon capture" enzymes.

When plants spend about three and half seconds to convert carbon dioxide to glucose during photosynthesis, they are spending an inordinate amount of time. The problem is that an enzyme called RuBisCO, which catalyzes the process, is highly inefficient.

But the basic idea of using biological molecules to capture atmospheric carbon is sound enough to get grants from the U.S. Department of Energy's Advanced Research Projects Agency-Energy.

Scientists are studying faster enzymes. One that is getting much new attention is carbonic anhydrase -- a protein found in blood, among other places, that captures carbon dioxide exhaled by cells. In one second, the enzyme can change a million molecules of the gas into harmless bicarbonate, according to Jonathan Carley, the vice president of business development at CO2 Solution, a Montreal-based company that is one among the few working on biomimetic design.

If it works in the body, can it work in a smokestack?

Scientists at CO2 Solution have been trying to engineer this enzyme to capture carbon dioxide from the harsh flue gas emitted by coal plants.

The traditional way to capture the gas is using a chemical called monoethanolamine. But the technique, developed nearly 60 years ago, is expensive. It takes $60 to capture a ton of carbon dioxide, and this doesn't include separation or storage, said Carley. The process also requires nearly 30 percent of the power generated by the power plant. The inefficiencies of the system have made carbon capture a commercially shunned technology.

Taking the cue from nature may prove to be the solution. With slight genetic modifications, a stable protein that can survive at high temperatures and be dissolved in a water-based solvent could be the answer. The technology from this company is meant to be incorporated into existing smokestacks so that retrofitting would be minimal, said Carley.

"Carbonic anhydrase is nature's solution for capturing and releasing carbon dioxide," he said.

The aim, Carley said, is to reduce the cost of capturing a ton of carbon dioxide by 30 percent, such that the price of capture lines up with carbon credit pricing.

"For instance, if 10 percent of emission comes from the USA, the savings for the USA for a 30 percent decrease in capture costs will be $36.2 million per day," said Ekrem Ozdemir, a researcher at the Izmir Institute of Technology in Turkey who is attempting to create stable carbonic anhydrase complexes.

The enzyme works by hydrating the gas, according to Ozdemir. Trapping carbon dioxide is the slowest step of the process. So, when equilibrium is reached between the gas in the atmosphere and dissolved carbon dioxide such that there is no further driving force for the reaction, having a stable enzyme reach out to grab more molecules will move it forward, Ozdemir said.

Finding clues and perhaps 'keys' in tiny geometries

The enzyme, which is initially isolated from an organic source and then modified, is still in the research and development stage. But the technology should be available by 2013, Carley said.

Others, such as Connecticut-based United Technologies Research Center, are attempting to create a completely synthetic carbonic anhydrase that will survive flue emissions.

Drummond and his colleagues at the University of North Texas are trying to push the frontier in biomimetic design a bit further. They've found patterns in the three-dimensional makeup of catalytic proteins that bind well to carbon dioxide. The work should make it possible to identify other enzymes that can be used in carbon capture technologies, as well.

"Nature does seem to use one or more patterns to use carbon dioxide at the molecular level," said Drummond.

Proteins are, at a basic level, mechanical constructs where, if the grooves match, a reaction takes place. Recognizing the grooves would help identify the enzymes that use carbon dioxide in their reactions.

The technique used is prominent in drug discovery, in which researchers study the geometries of the drug where it interacts with the human body to predict similar drugs that may be effective in the fight against disease. The particular three-dimensional arrangement of the drug that allows the interaction is called a pharmacophore.

"The simplest metaphor is someone trying to open a lock," said John Van Drie, a drug researcher who has written about pharmacophores previously. "You try key after key, and from that, you get a general idea. You can use that general idea to find the key that fits."

Will there be byproducts?

But here, the scientists know what the key looks like. It is the two-pronged shape of the carbon dioxide molecule. They are trying to find the lock, which in this case is enzymes, such as carbonic anhydrase.

Drummond has identified two patterns that seem to be conserved and are likely to signal carbon dioxide binding sites in proteins. Using these patterns as queries in a protein data bank that holds the known structures of nearly 60,000 proteins should yield new results.

"A protein could be really efficient at capturing carbon dioxide, even though its biological function has nothing to do with carbon dioxide," said Drummond. "In such a case, we could repurpose such a protein and put it to work removing CO2 from the air instead of doing whatever it does in living organisms."

And engineering these proteins to act as carbon dioxide sinks that produce a useful byproduct, such as lumber from trees, would be ideal, according to Drummond.

"It'd be great if we can get it to not only hold CO2 but also spit out what can be used," he said.