Rhodium nanoparticles have demonstrated the ability to capture the energy in UV light and use it to selectively catalyze the conversion of carbon dioxide (CO2) to methane. Such light-driven catalysis could be used to help reduce the level of CO2 in the atmosphere and for industrial applications. In the past two decades, the field of plasmonics has explored novel ways to use light to add energy to metal shrunk down to the nanoscale. In a study at Duke University, the plasmonic behavior of rhodium nanoparticles showed an improvement over rhodium’s already excellent catalytic properties by simultaneously reducing the activation energy and selectively producing methane, a desired product for the CO2 hydrogenation reaction. Duke University researchers have engineered rhodium nanoparticles (blue) that can harness the energy in ultraviolet light and use it to catalyze the conversion of carbon dioxide to methane, a key building block for many types of fuels. Courtesy of Chad Scales. Researchers synthesized rhodium nanocubes at an optimal size for absorbing near-UV light. They then placed small amounts of the nanoparticles into a reaction chamber and passed mixtures of CO2 and hydrogen through the chamber. When the nanoparticles were heated to 300° C, they reacted by generating an equal mix of methane and carbon monoxide (CO). While CO and methane were equally produced without illumination, methane was almost exclusively produced when rhodium nanoparticles were mildly illuminated. The reduced activation energy and super-linear dependence on light intensity caused the unheated photocatalytic methane production rate to exceed the thermocatalytic rate at 350 °C. “We discovered that when we shine light on rhodium nanostructures, we can force the chemical reaction to go in one direction more than another,” said professor Henry Everitt. “So we get to choose how the reaction goes with light in a way that we can’t do with heat.” The ability to selectively control the chemical reaction could be a factor in determining the cost and feasibility of industrial-scale reactions, said researcher Xiao Zhang. “If the reaction has only 50 percent selectivity, then the cost will be double what it would be if the selectivity is nearly 100 percent,” Zhang said. “And if the selectivity is very high, you can also save time and energy by not having to purify the product." Rhodium nanocubes were observed under a transmission electron microscope. Courtesy of Xiao Zhang. Small amounts of rhodium are used to catalyze a number of industrial processes, including processes for making drugs, detergents and nitrogen fertilizer. Rhodium accelerates reactions with a boost of energy that is typically in the form of heat. Heat is easily produced and absorbed, but can cause problems such as shortened catalyst lifetimes and the synthesis of undesired byproducts. “Effectively, plasmonic metal nanoparticles act like little antennas that absorb visible or ultraviolet light very efficiently and can do a number of things like generate strong electric fields,” said Everitt. “For the last few years there has been a recognition that this property might be applied to catalysis.” The team plans to test whether its light-powered technique to catalyze rhodium could drive other reactions that are currently catalyzed with heated rhodium metal. By tweaking the size of the rhodium nanoparticles, the team also hopes to develop a version of the catalyst that is powered by sunlight, creating a solar-powered reaction that could be integrated into renewable energy systems. “Our discovery of the unique way light can efficiently, selectively influence catalysis came as a result of an ongoing collaboration between experimentalists and theorists," said professor Jie Liu. “Professor Weitao Yang’s group in the Duke chemistry department provided critical theoretical insights that helped us understand what was happening. This sort of analysis can be applied to many important chemical reactions, and we have only just begun to explore this exciting new approach to catalysis.” The research was published in Nature Communications (doi:10.1038/ncomms14542).