Clean power from deserts

Overview of CSP
History of CSP
Activities and events
Press releases
Letters and comments
Mailing list
Site map

Turning CO2 back into hydrocarbons

27 February 2008

New Scientist

Duncan Graham-Rowe

CARBON dioxide is the devil molecule of our time. Belched out from vehicle exhausts and power stations, it is the biggest contributor to global warming. As such it is universally recognised as a Bad Thing. Yet a pioneering band of researchers would like us to see it differently - as a valuable resource. They are developing a collection of technologies to retrieve some of the CO2 that would otherwise pollute the atmosphere, using its carbon atoms to form hydrocarbons. These could then be used as vehicle fuel, or as a feedstock to make plastics and other materials we now derive from oil. So could the expanding clouds of CO2 in our atmosphere really have a silver lining?

The idea is simple. Find a way of removing an oxygen atom from a CO2 molecule and you are left with carbon monoxide (CO). From there it is but a short step to hydrocarbon riches. Mix CO with hydrogen, pass the mixture over a catalyst, and out comes liquid hydrocarbon fuel. This reaction, called the Fischer-Tropsch process, was invented as long ago as the 1920s. It was used by Germany during the second world war, when oil was in short supply, to make petrol from gasified coal, and apartheid-era South Africa did the same when sanctions blocked oil imports.

The hard part is the first step: finding a cost-effective energy-efficient way of creating CO from CO2. The simplest route is to heat CO2 molecules to around 2400 °C, at which point they spontaneously split into CO and oxygen. The problem is finding the energy to do this.

One obvious candidate is sunlight. Los Alamos Renewable Energy (LARE), a company based in Pojoaque, New Mexico, has built a small-scale prototype reactor that demonstrates how it can be harnessed. In the LARE reactor, CO2 is fed into a reaction chamber that is sealed at one end by a quartz window 8 centimetres in diameter. The chamber is fixed at the focal point of a mirrored dish that concentrates sunlight through the chamber's window onto a ceramic rod set inside the chamber to collect the heat. As the gas comes into contact with the rod its temperature rises to around 2400 °C, causing the molecules to break down and release CO and oxygen. Reed Jensen, LARE's managing director, says a larger prototype reactor will be ready for trials in a year's time, though he is not saying how big this reactor will be, nor how much CO it will produce.

One of the drawbacks of this approach is the high operating temperature, says Nathan Siegel of Sandia National Laboratories in Albuquerque, New Mexico, where a rival team is at work. High temperatures lead to heavy thermal losses, which in turn can reduce efficiency. Though the sun's energy is free, equipment to generate and withstand these temperatures is expensive to build, making efficient operation vital if the process is to be cost-effective.

With this in mind, the Sandia team is developing a rival system known as CR5 (short for counter-rotating ring receiver reactor recuperator) which operates at less extreme temperatures. Like the LARE reactor it has a concentrator dish that focuses the sun's rays. In this case, the high temperatures are generated on one side of a stack of 14 rings made of a cobalt ferrite ceramic, a material that when heated releases oxygen from its molecular lattice without destroying the lattice's integrity. The rings, which are about 30 centimetres in diameter, rotate at around one revolution per minute inside a sealed double chamber. Sunlight focused through a window in the hot side of the chamber heats the rings to 1500 °C, causing the ceramic lattice to liberate oxygen atoms. As the rings rotate, the heated section passes to the rear of the chamber, where it cools to around 1100 °C as it is bathed in CO2. At this temperature the deoxygenated ceramic reacts with the CO2 molecules to grab back the oxygen atom missing from its lattice, leaving behind a molecule of CO. As the ring continues to rotate, the reoxygenated section passes back into the hot side of the chamber and the cycle begins again (see Diagram).

Proper heating and cooling of the rings is crucial to the operation of the process. On the heated side, the rings must reach the correct temperature for the ceramic to liberate oxygen, and they must cool by several hundred degrees by the time they reach the cool side in order to react with the CO2. To help achieve this, alternate rings rotate in opposite directions, so as the hot section of each ring moves towards the cool side of the chamber it is cooled by neighbouring rings moving in the opposite direction. Both hot and cooler sides of the chamber are maintained at equal pressure to minimise the flow of gases between them.

The CR5 was originally developed as a way to produce hydrogen, using steam in the cool chamber rather than CO2, but its inventor, Rich Diver, reckoned that splitting CO2 would offer a more efficient way of capturing solar energy. Burning the CO formed in the solar reactor should deliver 10 per cent of the energy that was required to produce it, and in April he and his colleagues will switch on a prototype reactor to put their predictions to the test. They calculate it should be able to produce about 100 litres of CO per hour.

Reversing the fuel cell

The idea of using solar energy to convert CO2 into a carbon-based fuel is being taken a step further by Gabriele Centi at the Department of Industrial Chemistry and Engineering of Materials at the University of Messina, Italy. Rather than producing CO with a view to turning that into something more useful, he is building an electrochemical cell that produces hydrocarbon molecules such as nonane and ethylene - important chemical building blocks for plastics and other materials currently derived from oil.

Centi's cell is a distant cousin of the fuel cells that generate electricity by reacting hydrogen or methanol with oxygen, but with the chemical reaction running in reverse. On one side of the cell is a titanium dioxide catalyst that encourages water molecules to split when hit by photons of sunlight, producing hydrogen ions and oxygen gas. The hydrogen ions migrate through a proton exchange membrane to the other side of the cell, where a catalyst containing platinum nanotubes facilitates the reaction with CO2 to produce hydrocarbons.

The energy that would be liberated by using these hydrocarbons as fuel amounts to just under 1 per cent of the solar energy needed to produce it. This may not seem like much but it's better than the energy conversion rate that plants achieve through photosynthesis, and Centi says there is room for improvement by tweaking the catalysts.

So how do these technologies stack up against biofuels as a way of using solar energy to capture atmospheric carbon and turn it into fuel? Ellen Stechel, manager of Sandia's Fuels and Energy Transitions department, estimates that enough CR5 plants to fuel 100 million domestic vehicles with synthetic gasoline could be accommodated on about 5800 square kilometres of land. "That's actually not very much," she says. A recent survey of seven states in the US Southwest revealed that more than 135,000 square kilometres of suitable land were available there. "This is land that's not being used for anything else," she says.

By contrast, biofuels compete with food crops for fertile land. What's more, the percentage of the solar energy that is available from the fuel is staggeringly small - about 0.1 per cent if you take into account the irrigation, harvesting, transportation and refinery process, Stechel says.

To make the most of the available land, Jensen suggests coupling LARE's carbon capture reactor with an electricity generating station that would use the heat wasted by the reactor itself. He reckons the combined installation could convert as much as 48 per cent of the solar energy into usable energy.

As oil and natural gas become more expensive and scarce, petrochemical companies are increasingly interested in finding new raw materials to replace them. Centi is now working with one French firm to explore the use of recycled CO2 to meet this demand, though he refused to name the company. If competitively priced, hydrocarbons produced from industrial sources of CO2 could one day be used to make plastics and other products, where it would remain fixed for years rather than being pumped out into the atmosphere. The devil molecule may yet redeem itself.

"Hydrocarbons produced from industrial sources of CO2 may replace those from oil"

Duncan Graham-Rowe is a writer based in Brighton, UK

From issue 2645 of New Scientist magazine, 27 February 2008, page 32-34

Last updated: 2009-08-20 (ISO 8601)