A ceramics company, Ceramatec Inc., and a US government laboratory, and the Idaho National Engineering and Environmental Laboratory (INEEL) said a computer model has showed they could use a nuclear reactor to separate hydrogen from hot water.

Officials from the two organisations said this method would yield more hydrogen than can be produced by electrolysis, which runs electricity through water to separate hydrogen and oxygen. The conversion rate of water into hydrogen ranged between 45 and 50% high temperatures, compared with about 30% in electrolysis, the researchers said.

Researchers said the process of obtaining hydrogen by splitting water using electric energy has been known for about 150 years, but costs in terms of dollars and electric energy made it an unpopular choice.

"High temperature electrolysis has the potential to change that by reducing the amount of electrical energy required and using a proportion of thermal energy in its place," said Joseph Hartvigsen with Ceramatec.

"We have been able to show that we can produce hydrogen at commercially attractive rates in a very small unit and at conditions that are typical of a high-temperature, helium-cooled reactor," said INEEL lead researcher Steve Herring.

The sample, about the size of a paperback book, was successfully tested in a kiln that can simulate the high temperatures that would be created in so-called Generation 4 nuclear reactors – about 980 degrees Celsius.

The new method involves running electricity through water in the kiln. As the water molecule breaks up, a ceramic sieve separates the oxygen from the hydrogen.

Ceramatec and INEEL will partner with Hoeganaes Corp. in New Jersey and the University of Washington for the project to increase the sample size 100-fold over the next three years.

But the researchers admit it would be decades if not a generation before hydrogen power and its infrastructure is as commonplace as the one in place today for petroleum-based energy, such as refineries and gas stations.

Instead, Herring said, the most immediate use of hydrogen using the new process would be to upgrade poor quality petroleum for use as motor fuels and then synthesizing existing fuels that cars can use, like gasoline, diesel and jet fuel, he said.

"Then the third stage in the use of hydrogen would be the use of pure hydrogen as a transportation fuel," he said.

But the plan requires building a new kind of nuclear reactor, at a time when the US is not even building conventional reactors. And the cost estimates are uncertain.

The long-term goal is to create a reactor that could produce about 300 megawatts of electricity.

The Department of Energy (DOE) is hoping for a demonstration of commercial-scale hydrogen production using the process by 2017. The government is considering building the necessary next-generation power plant at INEEL, said DOE-INEEL initiative lead researcher Michael Anderson.