A chemist has developed catalysts that could power every PC, convert coal to petroleum and clean up greenhouse gases. Geoff Watts reports

I am about to be impressed. Oxford University research chemist Tiancun Xiao's only props are a small polythene bottle of colourless fluid, a valve, a short length of glass tube filled with what looks like pieces of gravel and some plastic piping to connect it all together. After a few minutes of fiddling with a couple of spanners - taken from his briefcase - Xiao has assembled his apparatus and is ready to begin. He upends the bottle, submerges the free end of the piping in a beaker of water, opens the valve and a stream of gas comes bubbling up.

But why should I be so taken with what appears to be a bit of rather elementary chemistry? Because the gas is hydrogen, and the set-up that Xiao is demonstrating has the potential to change the way we power some of the most essential gadgetry of the 21st century. The key to this change lies in the "gravel" in the glass tube. It is in fact granules of a catalyst: a commercially valuable material, the nature and manufacture of which is protected by patent.

Catalysts promote reactions between other chemicals without themselves undergoing change. The search for new catalysts or better ways of using existing ones is a fundamental branch of chemistry. Consider, for example, the furious decomposition that goes on deep in the waste buried within every landfill site. The product of all this chemical and microbiological activity is methane - which percolates upwards, outwards and into the atmosphere. It is a flammable gas and an environmental pollutant; each molecule has 23 times the greenhouse effect of carbon dioxide. Engineers know this and may nowadays trap the methane as it emerges. But storing and using it is often uneconomic; much is simply burnt off. The carbon dioxide produced may be a less potent greenhouse gas; but burning anything without getting something back is wasteful.

³ÉÈËVRÊÓƵ

Xiao is a chemist by training, In China he had completed his PhD and worked for some years in environmental protection. He moved to Oxford in 1999 and joined Malcolm Green's Wolfson Catalysis Research Group.

The first report of the catalytic breakdown of methane dates from 1946; but the method used gave a poor yield of hydrogen and carbon monoxide, the products of the reaction. In the late 1980s, Green devised a new method that raised the yield to something near 97 per cent. The only problem was that it relied on a pricey metal catalyst.

³ÉÈËVRÊÓƵ

Expensive catalysts are fine if you operate on a small scale: producing pharmaceuticals, for example. By contrast, when dealing with methane, the huge tonnage of material to be treated requires large amounts of catalyst - which has to be inexpensive to be economically feasible. "We worked on it for a long time," recalls Green. "Then Tiancun discovered you could use carbides of cheap metals such as cobalt. The catalysts we have now perform as well as the best metal catalysts. In fact in some respects they're better."

Last year Xiao's achievement received the recognition it merits: an innovation award from the Carbon Trust. This body encourages the UK to move towards a low-carbon economy by helping organisations to reduce their carbon emissions and to exploit commercial opportunities created by the need for low-carbon technologies.

"One of the reasons we won the award is that we have a catalyst that can cope with methane from biomass, which is often full of impurities such as sulphur. Most catalysts will become poisoned and inactive," Xiao says.

Landfill is only one source of avoidable atmospheric methane. The oil production industry is another. "Everywhere that oil is dug up there's methane. Normally they burn it," Green says.

To exploit their findings, Green and his colleagues have had the help of Isis, Oxford's technology transfer organisation. The outcome is Oxford Catalysts Ltd, a company registered in 2004 and launched last December with investment from IP2IPO, an intellectual property firm that specialises in commercialising university technology. Xiao is now Oxford Catalysts' chief technical officer.

But the company is no one-trick pony. "We also have a patent on a new method of removing the last traces of sulphur from gasoline," Green says.

³ÉÈËVRÊÓƵ

"It becomes prohibitively more difficult to remove the last hundred parts per million without damaging the quality of the fuel, and we've discovered a method of doing it."

Other innovations include the conversion of natural gas and coal to synthetic petroleum, for example. The process has been around for decades; the South Africans used it during the apartheid years, and Germany relied on it to fuel tanks and aircraft in the final years of the Second World War. The process involves two phases of catalytic conversion: first methane to a mixture of hydrogen and carbon monoxide; then their synthesis into paraffin and other long-chain hydrocarbon molecules that can be split to yield petroleum.

³ÉÈËVRÊÓƵ

The company also has a catalyst with an application in industrial and domestic cleaning. Given an appropriate fuel, the right catalyst will generate a jet of hot steam that can be used to remove chewing gum from pavements, to sterilise operating tables or to clean oil tankers.

The company's slogan - "Catalysing a cleaner future" - chimes nicely with its interests in general; most are concerned with cleaning up existing fuels or generating new ones that are environmentally more friendly.

Back to Xiao's demonstration. The bottle contains methanol mixed with a little water and hydrogen peroxide. When this is allowed to trickle over the catalyst in the glass tube, the methanol is broken down to liberate the hydrogen that I see emerging as a stream of bubbles in the beaker. It is production on a very small scale - but, fed to a fuel cell, it would be sufficient to generate a flow of current.

Fuel cells are devices that turn chemical energy into electricity - and the simplest fuel of all is hydrogen. To power your computer by carrying round a steel cylinder of compressed hydrogen would be absurd. The catalytic answer is to make it as you need it. Methanol is stable and light; a coffee cupful of it will generate enough hydrogen to drive a computer for 48 hours. Fuel cells can now be made light and small enough to fit into a laptop. The hydrogen would start flowing the moment it was needed. The whole powerpack would weigh no more than a conventional battery, and its only byproducts would be water and carbon dioxide. Given the number of laptops around, the size of this market alone would be vast.

But whether Xiao's future lies here in Oxford or in China remains to be seen. The Chinese are interested in industrial catalysis, not least in innovative and less polluting ways of dealing with the coal they have in such vast amounts. Efficient ways of converting it to a liquid fuel - of which they are notably short - would be enormously attractive.

"They might want somebody like him to run it," Green says, nodding towards Xiao.

³ÉÈËVRÊÓƵ

Xiao just chortles.