A Better Solar-Powered Water Splitter Devised

To withstand the highly corrosive environment inherent in the process of splitting water, electron-producing silicon-based solar electrodes, which deteriorate almost immediately when exposed to oxygen, are now being coated with an ultrathin layer of titanium dioxide, rendering them durable and efficient for transferring electricity.

Conceptually, splitting water could not be simpler. Scientists have long known that applying a voltage across two electrodes submerged in water splits the water molecules into their component elements, oxygen and hydrogen.

From an environmental standpoint, the process is a dream: an electrochemical reaction whose only requirements are water and electricity and whose only byproducts are pure oxygen and hydrogen, a clean-burning fuel applicable in a promising new class of renewable energy applications. In fact, hydrogen is the cleanest burning chemical fuel known.

"In theory, water splitting is a clean and efficient energy storage mechanism. Unfortunately, solving one problem creates another," said Paul McIntyre, associate professor of materials science and engineering at Stanford University, who led the study with chemist Christopher Chidsey. "The most abundant solar electrodes we have today are made of silicon, a material that corrodes and fails almost immediately when exposed to oxygen, one of the byproducts of the reaction."

This problem had vexed researchers since at least the 1970s. Many had given up, but McIntyre and Chidsey devised the clever solution of coating their silicon electrodes with a protective, ultrathin layer of titanium dioxide.

"Titanium dioxide is perfect for this application," McIntyre said. "It is both transparent to light and it can be efficient for transferring electricity, all while protecting the silicon from corrosion."

Sunlight travels through the protective titanium dioxide into the photosensitive silicon, which produces a flow of electrons that travels through the electrochemical cell into the water, splitting the hydrogen from the oxygen. The hydrogen gas can be stored and then, when the sun is not shining, the process can be reversed, reuniting hydrogen and oxygen back into water to produce electricity.

Other researchers had attempted to protect the electron-producing silicon electrodes. Some tried other materials, which failed for reasons of performance or durability. Some had even tried titanium dioxide, but those efforts also fell short. Their layers were either materially flawed, allowing oxygen to seep through and corrode the semiconductor, or too thick to be electrically conductive.

Yi Wei Chen and Jonathan Prange, the lead doctoral students on the McIntyre-Chidsey team, discovered that the key to the titanium dioxide's protectiveness is achieving a very thin, yet high-quality layer of material. They found that a layer just 2 nm thick was sufficient as long as it was free of the pinholes and cracks that doomed earlier titanium dioxide experiments.

With their electrodes successfully shielded from corrosion, the researchers revealed yet one more engineering ace in the hole, adding a third layer of ultrathin iridium, a catalyst, atop the titanium dioxide. Iridium boosts the rate of the splitting reaction and improves performance of the system.

In side-by-side durability experiments, the researchers put their creation to the test. Control samples without the protective layer corroded and failed in less than a half-hour, while those with the titanium dioxide lasted the full duration of the test, eight hours without apparent corrosion or loss of efficiency.

The authors pointed out that their approach is general enough to work on other semiconductor substrates and to integrate other catalysts, allowing for fine-tuning of electrodes to maximize performance. Likewise, atomic layer deposition, the technique that allowed such fine and flawless layering, is in wide application in the semiconductor industry today. It should, therefore, lend itself to application on a large scale. Lastly, the results were achieved without exploring the use of other efficiency-enhancing techniques, such as surface texturing, which could further improve performance.

"We are excited about the possibilities of this technology," said McIntyre, "as much for the electrode itself, as for the process used to create it."

Their success might just push a promising technology one step closer to practical application and the world one step closer to a clean-energy future.

Seed funding for this work was provided by these Stanford groups: Precourt Institute for Energy, the Center for Integrated Systems and the Global Climate and Energy Project.

The study was published in the journal Nature Materials.

For more information, visit: www.stanford.edu