Computational studies using quantum calculations could provide a framework for identifying the best catalysts

US researchers have developed a new strategy for predicting bimetallic catalysts. They have already put it to good use in identifying a nickel-platinum catalyst that they claim is the best ever for the ammonia decomposition reaction - potentially a vital reaction if ammonia is to become an important means of storing hydrogen.

Making predictions about catalysts that contain more than one metal is complicated because the properties of a mixed metal are not necessarily intermediate to those of the parent metals. Nevertheless, this has remained the common assumption in the literature, says Dionisios Vlachos, one of the authors of the new study based at the University of Delaware. His team has developed a method that gets around the problem using density functional theory (DFT) calculations - calculations based on quantum mechanics.

’What the density functional theory calculations do is they account for the architecture of the catalyst,’ explains Vlachos. ’You need to account for the unique architecture of the atoms in space - where they actually reside - in order to be able to predict the properties of the correct material.’ The simplified approaches used previously, he says, don’t account for the full chemistry of the reaction.


Source: © Nature Chemistry

The architecture of bimetallic catalysts affects how they function

Vlachos’s team used DFT calculations to screen a library of microkinetic models -detailed reaction mechanisms - for ammonia (NH3) decomposition, with different bimetallic materials as catalysts. Noting that for single metal catalysts, the binding energy for nitrogen at the catalyst surface was an important indicator of catalytic activity, they calculated this value for mixed metals. One, a nickel-platinum catalyst, had a nitrogen binding energy close to that of the current best single metal catalyst, ruthenium. And the researchers’ predictions were born out experimentally, with activity starting at 50?C, compared to 350?C for ruthenium. 

The real value of the work, however, is in creating a framework for identifying bimetallic catalysts more generally says Vlachos. Platinum may not be the best substitute for ruthenium, simply because of its cost, but Vlachos believes the same principles could be applied to search for cheaper materials, and for bimetallic catalysts for other reactions, including the selective oxidation of carbon monoxide for hydrogen purification and hydrogenation of hydrocarbons. 

Claus Hviid Christensen, a catalysis expert working for catalysis firm Haldor Topsoe in Denmark, is interested to see more and more groups going down the computational route. ’It’s a very efficient way to narrow down the enormous range of possible catalyst candidates,’ he says. ’And this reaction is itself interesting because ammonia is a carbon-free energy carrier.’ 

Hayley Birch