Perovskite solar cells have made remarkable progress in the past few years, and the rising efficiency of the low-cost devices has begun to rival conventional silicon cells. Now perovskite photovoltaics have been used to split water to produce hydrogen, offering a promising route to the clean-burning fuel.
Hydrogen is very slowly gaining ground as a transportation fuel, but it has a dirty secret: most of the world’s supply of the gas is made via reactions with methane and steam at high temperatures, releasing the greenhouse gas carbon dioxide in the process. Developing a less-polluting source of hydrogen could help to bolster its green credentials and speed its adoption. Using electricity from renewable sources to split water by electrolysis, however, is far too expensive today.
That’s partly because it takes a voltage of at least 1.23 V to split water, with commercial systems running at about 1.8 V to 2.0 V. Conventional silicon or cadmium-telluride solar cells cannot deliver that voltage, because their bandgap is not wide enough. So three or four cells must be connected in series to reach the electrolysis threshold. At large scales, such systems are therefore uneconomical.
Enter the clean energy savior du jour: perovskites, which have a wide band gap, enabling each cell to produce a relatively beefy voltage of up to 1.5 V. What’s more, they rely on a cheap and easy-to-manufacture light-absorbing layer such as methyl ammonium lead iodide. (The cell’s name refers to this material’s crystal structure). And since 2009, improvements in the chemistry and design of the cells’ various layers have pushed their efficiency from just 3.8 percent to a whopping 17.9 percent, with some labs reporting unconfirmed results up to 19 percent.
Michael Grätzel at the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland, one of the pioneers of perovskite cells, has now shown, along with his colleagues, that using just two cells in parallel is enough to start the hydrogen fizzing out of water. Their research was published this week in Science.
His team connected the cells to cheap, efficient electrocatalysts of their own design that use the solar power to split water. One electrode delivers electrons to water, splitting the molecules into hydrogen gas and hydroxide anions. The other electrode uses those hydroxide anions to produce oxygen gas and release more electrons back to the circuit.
Previous electrocatalysts like platinum and iridium are cursed by high costs, and cheaper materials that contain cobalt or molybdenum are finicky. Electrolysis systems generally need the water to be spiced with acid or base to help the current flow, but here's the rub: Electrocatalysts that are good at generating hydrogen are generally only stable if the water is acidic, while those adept at generating oxygen prefer basic conditions.
Grätzel’s system builds on previous work with an electrode made of iron and nickel. His team coated porous nickel “foam” with iron and nickel hydroxide, and tested the electrodes in a basic solution of sodium hydroxide. They found that the electrodes generated oxygen more effectively than a platinum-nickel electrode, and were almost as good at generating hydrogen— quite a surprise, given that this reaction usually runs far more slowly in basic solution. “The real advance is being able to use this less-than-ideal catalyst, because the voltage of the perovskite cell is so high,” says Thomas Hamann at Michigan State University in East Lansing, who wrote a commentary on the work in Science.
In the researchers’ prototype system, a pair of perovskite cells converted light into electricity with an efficiency of 15.7 percent, and generated a healthy 2.0 V. That was more than enough to make hydrogen and oxygen stream from the electrodes. Measuring the rate of gas production, Grätzel’s team calculated that the overall solar-to-hydrogen conversion efficiency was 12.3 percent. That matches the performance of a gallium-indium-phosphide/gallium-arsenide tandem cell linked to platinum catalytic electrodes—but at a much lower cost, says Hamann. Indeed, of all similar systems that use cheap, abundant materials, nothing has topped 10 percent efficiency, he adds.
As perovskite cells continue to improve, Grätzel’s team reckons that the overall solar-to-hydrogen conversion efficiency could rise to 15 percent. That’s in line with the trajectory of the US Department of Energy's goals for solar hydrogen production: 15 percent efficiency by 2015, 20 percent efficiency by 2020, and significant cuts to the cost of the system’s components.
If those targets could be met, hydrogen might also be used to bank excess energy generated by solar power or wind turbines and be deployed when these facilities idle or when demand surges. In theory, storing energy in the form of free hydrogen could be more convenient than pumped hydropower, and a lot cheaper than batteries.
There is one major drawback: Grätzel’s perovskite cells degraded after only a few hours. “Stability of the perovskites is a known issue,” says Jingshan Luo, part of Grätzel’s team at EPFL. But he points out that other researchers have recently shown that perovskite solar cells can be made much more resistant to moisture or heat, and in some cases, made to operate for more than 1000 hours without any loss in performance.
If the stability issues can be conquered, Hamann suggests that there are other ways to improve the performance of the system. Rather than using two perovskite cells, one of them could be placed on top of a silicon cell. Since silicon has a smaller band gap, it would absorb light from the red end of the spectrum that currently streams through the semi-transparent perovskite cell. This could halve the area of solar panels needed, increase the overall efficiency of the solar energy conversion, and boost hydrogen production. “That’s exactly what I’m doing now!” says Luo excitedly.
This strategy comes with a cost, however: It would lower the overall voltage of the tandem cell, requiring a more efficient hydrogen-evolving electrode. Luo is working on this, too, and says he and his colleagues are making good progress: “Most perovskite groups are focused on the cells, and they’re not experts in solar fuel generation,” he says. “We’re doing research on both sides.”