Search

The combination of computational chemistry and computational materials science with machine learning and artificial intelligence provides a powerful way of relating structural features of nanomaterials with functional properties. However, combining these fundamentally different scientific approaches is not as straightforward as it seems. Machine learning methods were developed for large data sets with small numbers of consistent features. Typically nanomaterials data sets are small, with high dimensionality and high variance in the feature space, and suffer from numerous destructive biases. None of the established data science or machine learning methods in widespread use today were devised with (nano)materials data sets in mind, but there are ways to overcome these challenges and use them reliably. In this review we will discuss domain-specific constraints on data-driven nanomaterials design, and explore the differences between nanomaterials simulation and nanoinformatics that can be leveraged for greater impact.

Zinc sulfide (ZnS) nanoparticles are of interest for their luminescent and catalytic properties which are being considered for the next generation of optical, electronic and photovoltaic devices. However, ZnS nanoparticles undergo reversible and irreversible phase transformations under ambient conditions, so a detailed understanding of the nanomorphology is critical in ensuring these desirable properties can be controlled and maintained. Anticipating the structure and transformations in ZnS nanoparticles experimentally is difficult, since selectivity among competing phases, shapes and sizes is intrinsically linked. Presented here are the results of first principle computer simulations and advanced theoretical modelling used to investigate the relationship between size and shape in determining the crystallinity of ZnS nanoparticles. We find that the equilibrium morphology is characterised by {220} facets, irrespective of the size of the particle, but that the presence of different high energy facets introduced kinetically may significantly influence the zinc blende to amorphous ZnS transformation size, as well as the agglomeration behaviour. In addition to this, we model the relationship between transformation size, morphology and the ratio of crystalline core to amorphous shell and show that at small sizes, a core-shell crystalline/amorphous structure is thermodynamically favourable.