The coordination environments of different platinum atoms in single-atom catalysts have been measured by researchers in Switzerland using platinum NMR spectroscopy. The results, which show how the activity of single-atom catalysts depends on the atoms’ binding to their substrates, could open new opportunities for rational catalyst design.1

Single-atom catalysts usually comprise a metal – often a precious metal such as platinum – atomically dispersed on a support such as nitrogen-functionalised carbon. The catalytic sites are found at the interfaces between metal and support. Microscopy techniques can characterise the dispersion of a metal on a catalyst surface, but Javier Pérez-Ramírez at ETH Zürich explains that this is insufficient. ‘The techniques that exist provide an average,’ he says, ‘but single atoms are not single sites … the coordination environments – the local structures – might be very different.’

In the new work, researchers in the groups of Pérez-Ramírez and Christophe Copéret at ETH Zürich, together with colleagues in France and Denmark, performed solid-state platinum-195 NMR spectroscopy on single-atom catalysts comprising platinum dispersed on nitrogen-functionalised carbon supports. This did not produce high spatial resolution as NMR uses long-wavelength microwave radiation. However, the chemical shift of the platinum-195 atoms depended on the binding to the support, and this information could be recovered from theoretical analysis of the spectra. ‘Imagine you’re at the opera and every singer is singing together: we have to listen to all the individual voices and tell what the distribution of the voices is,’ explains Copéret.

The researchers showed that different synthetic protocols for the design of the same catalyst could yield markedly different coordination environments, even when the catalyst looked the same under a microscope. This could explain the difficulties researchers encounter in reliably producing active single-atom catalysts. The researchers also studied the effects of different supports on the coordination of the catalyst sites. Finally, they examined how a single-atom platinum catalyst for the hydrochlorination of acetylene changed after multiple hours on stream, during which the catalyst was slowly deactivated. ‘The catalyst before and after the reactions still had single atoms, but they were not the same,’ Copéret says.

The researchers now hope that they and others will be able to design modified synthetic protocols that will allow catalysts to have a larger proportion of active sites. ‘In catalysis the structure determines the function,’ says Pérez-Ramírez. ‘It’s meaningless if only a few soldiers are the ones doing the fighting.’

‘They’ve basically covered all the bases for a good quality paper,’ says materials scientist Anatoly Frenkel at Stony Brook University in the US. ‘They’ve not only shown a proof of principle but some non-trivial results – I think it’s great work.’ He believes the protocol now might be able to provide crucial insights into the intra-particle heterogeneity of platinum nanoparticles, which are also very common catalysts.

‘It’s absolutely a landmark paper,’ agrees solid-state NMR spectroscopist Rob Schurko at Florida State University. He says the work’s roots lie in the 1981 report by the NMR pioneer Charles Slichter, then at the University of Illinois at Urbana-Champaign, that the platinum NMR spectra of platinum catalysts was significantly altered by the adsorption of molecules such as carbon monoxide.2 The current work, he says, allows single-atom resolution. ‘This is like a full-circle evolution from that early work by Charlie Slichter to the thing that people really want to do with platinum NMR, and down the road I see what he’s doing here could definitely be extended to other platinum-group elements like ruthenium, rhodium and possibly palladium,’ he concludes.