Goodbye coal, oil? Team tests cheaper way to use sunlight to produce hydrogen

Researchers have developed a tool for splitting water into hydrogen and oxygen that holds the potential to significantly cut the costs of using sunlight to drive the reaction.

The team, led by Jingshan Luo, a researcher at the École Polytechnique Fédérale de Lausanne in Switzerland, used chemical elements that are abundant and relatively cheap to make solar cells and give them electrodes that reach what some researchers have called "exceptional" efficiency.

At this stage, the devices represent a proof of concept, cautions Thomas Hamann, a chemist at Michigan State University who didn't take part in the study but is also conducting research on photovoltaic cells and their applications.

"It's not game over, but it's an important step with a lot of promise," he says.

Using sunlight to produce hydrogen is a way to capture the sun's energy and store it for future use, much as a plant captures sunlight and through photosynthesis uses the sun's energy to produce hydrocarbons – the basis of fossil fuels.    

In principle, solar-derived hydrogen could then replace fossil fuels for uses ranging from powering vehicles to producing electricity via fuel cells.

The basic approach for splitting water molecules might be familiar to anyone who has visited an elementary school science fair. Connect a wire to each terminal of a battery, then put the other ends into a jar of water with the bare wire exposed. Oxygen bubbles up from the end of the positive wire, and hydrogen bubbles up from the negative wire.

The earliest effort to use sunlight, instead of a battery, to split water molecules and produce electricity dates back to 1972: A pair of Japanese scientists found that when an electrode made from titanium oxide was submerged in water and exposed to light, it split water into hydrogen and oxygen.

Since then, researchers have been exploring ways to use photovoltaic cells to do the same thing. But these have been made largely with more-expensive minerals, including rare ones such as indium and platinum. Their relatively high cost represents a challenge to large-scale commercial use. In addition, many of these cells have to be grouped in gangs of three or four to provide enough voltage to split water, researchers say.

Dr. Luo and colleagues used solar cells made from carbon, hydrogen, nitrogen, lead, and iodine with a crystal structure that chemists call a perovskite. The cells' electrodes were made from a blend of nickel hydroxide and iron.

To generate the water-splitting voltage, the team needed only two cells, which they showed to be 12.3 percent efficient at converting sunlight to hydrogen.

"This is the first time we have been able to get hydrogen through electrolysis with only two cells!" Luo said in a prepared statement.

The efficiency achieved is higher than other systems built from earth-abundant chemicals, Dr. Hamann notes. And it's a scant 1/10th of a percent below the 12.4 percent efficiency of a system using cells and electrodes made from more-expensive, less abundant elements such as platinum.

The results appear in Friday's issue of the journal Science.

The team acknowledges one hitch: The recipe for its solar cell grows less efficient over the course of about 10 hours, although the cell appears to reconstitute itself when it isn't being used – a phenomenon the team uncovered when it cycled the system through simulated day and night cycles.

This instability "clearly needs to be addressed before commercialization would be considered," says Alex Martinson, an assistant chemist at the Argonne National Laboratory in Lemont, Ill., who focuses on developing new approaches to solar-cell and solar-fuel production.

Still, "putting everything together into an efficient water-splitting system is never an easy task," he says, tipping his hat to the team's achievement. Dr. Martinson made his comments via e-mail.

If that problem is solved, one additional improvement could come by twinning one of the perovskite cells with a silicon-based cell. In principle, the water-splitting efficiency could rise to 20 percent or more, because the two in tandem would cover more wavelengths of light than either alone, Hamann says. Although the voltage would be lower, researchers could make up for that by choosing the right materials for the electrodes, which include a catalyst to enhance electrode performance.

Researchers have been tantalized by the potential of perovskite-based photovoltaic cells because they convert sunlight to electricity with high efficiency and potentially inexpensive materials, Martinson says.

That this technology can be arrayed "to produce a voltage that conveniently splits water is a nice added bonus," he says.