MIT screens greener ammonia catalysts in computer models

The bottleneck in cleaner ammonia production is not the idea of an alternative. It is finding a material that can make the chemistry run fast enough. At MIT, professor Bilge Yildiz and doctoral students Constantine Athanitis and Filip Grajkowski have developed a computational method to narrow that search, identifying promising catalyst alloys in computer models rather than testing combinations at random.

The stakes are substantial. Ammonia production uses roughly 2 percent of the world's energy and generates about 1.5 percent of global greenhouse gas emissions, mostly to make fertilizer. More than 90 percent of the ammonia produced today still depends on the Haber-Bosch process, a fossil-fuel-dependent method that has remained largely unchanged for more than a century.

The alternative studied by the MIT team uses electrochemistry: nitrogen gas reacts with proton-electron pairs, following principles also found in electrolyzers. It is not ready for industrial use. Production rates and yields remain too low for companies to justify switching, making a better catalyst the central target.

The researchers focused on transition metal nitrides, whose own nitrogen atoms participate in the reaction. That built-in nitrogen creates a sequence of chemical steps in which each stage supplies part of the energy needed by the next. The method also highlights remaining bottlenecks, including nitrogen dissociation and hydrogen transfer, so researchers can focus on the properties that matter most.

So what changes in practice? The team’s model could cut years from the early search for candidate materials by replacing broad trial and error with targeted computational screening. But the result is still a laboratory starting point, not a working ammonia plant: the proposed alloys must first be synthesized, then tested in a reaction cell under real operating conditions. University of Wisconsin engineering professor Dane Morgan, who was not involved, said translating the calculations into effective catalysts will require considerable additional work.