Hydrogen is currently used to upgrade crude oil and synthesize ammonia, a critical building block of the fertilizers applied in modern agriculture. It also could be valuable as a feedstock for generating green electricity and as an ingredient in environmentally friendly fuel cells to power cars and trucks. But hydrogen is commonly produced from natural gas heated by steam, which results in greenhouse gas emissions and other environmental problems. Thus, scientists have been working to replace this process with one that taps a renewable energy source—and just such a breakthrough was announced in a paper recently published in Nature Energy.
The new approach relies on a photoelectrochemical (PEC) device, a type of solar cell that can potentially split water molecules more efficiently than other methods. Scientists have long struggled to design a PEC device that is both efficient and durable enough to be cost-effective. A key advance came 18 years ago, when John Turner, an electrochemist at the U.S. National Renewable Energy Laboratory, designed a device that comprised layers of gallium indium phosphide and gallium arsenide semiconductors. These materials convert sunlight to electricity more efficiently than other options. Turner's design held the record for the highest solar-to-hydrogen conversion efficiency until 2015. But the acidic solution to which the cell was exposed while in use quickly broke it down, making the hydrogen it produced too expensive.
For the new design, researchers led by chemist Jing Gu of San Diego State University added coatings to the semiconductor layers to prevent acid corrosion. These protective coatings significantly extended the life of Turner's high-efficiency design and produced a PEC device that retains 80 percent of its capabilities in durability tests. A “hydrogen economy” in which consumers can make their own hydrogen to power their cars and heat or cool their homes may not yet be imminent, but at least this engineering feat makes such a future sound a little less like utopian hype.