LAUSR

Catalytic activity of the hydrogen evolution reaction on nanoclusters depends on diverse adsorption site structures. Machine learning reduces the cost for modelling those sites with the aid of descriptors. We analysed the performance of state-of-the-art structural descriptors Smooth Overlap of Atomic Positions, Many-Body Tensor Representation and Atom-Centered Symmetry Functions while predicting the hydrogen adsorption (free) energy on the surface of nanoclusters. The 2D-material molybdenum disulphide and the alloy copper–gold functioned as test systems. Potential energy scans of hydrogen on the cluster surfaces were conducted to compare the accuracy of the descriptors in kernel ridge regression. By having recourse to data sets of 91 molybdenum disulphide clusters and 24 copper–gold clusters, we found that the mean absolute error could be reduced by machine learning on different clusters simultaneously rather than separately. The adsorption energy was explained by the local descriptor Smooth Overlap of Atomic Positions, combining it with the global descriptor Many-Body Tensor Representation did not improve the overall accuracy. We concluded that fitting of potential energy surfaces could be reduced significantly by merging data from different nanoclusters.Machine learning: hydrogen adsorption energy on nanocluster surfacesThe accuracy and efficiency of descriptors for machine learning are tested for hydrogen evolution reaction in nanocatalytic systems. A team led by Adam Foster at Aalto University analysed the performance of SOAP, MBTR and ACSF structural descriptors used to gain insight to the free energy of hydrogen adsorption on the surface of nanoclusters relevant to catalysis applications. Using atomically thin MoS2 and AuxCuy metallic alloys as test systems, the accuracy of the predictors was evaluated when used as features in kernel ridge regression. Analysis on a dataset of 91 MoS2 clusters and 24 AuxCuy clusters (x + y = 13) indicated that while none of the descriptors which had been designed for molecules and crystals was optimized for nanoclusters, SOAP performed significantly better, making it a more suitable candidate for adsorption energy prediction.