Data-driven simulation and characterisation of gold nanoparticle melting

Abstract

The simulation and analysis of the thermal stability of nanoparticles, a stepping stone towards their application in technological devices, require fast and accurate force fields, in conjunction with effective characterisation methods. In this work, we develop efficient, transferable, and interpretable machine learning force fields for gold nanoparticles based on data gathered from Density Functional Theory calculations. We use them to investigate the thermodynamic stability of gold nanoparticles of different sizes (1 to 6 nm), containing up to 6266 atoms, concerning a solid-liquid phase change through molecular dynamics simulations. We predict nanoparticle melting temperatures in good agreement with available experimental data. Furthermore, we characterize the solid-liquid phase change mechanism employing an unsupervised learning scheme to categorize local atomic environments. We thus provide a data-driven definition of liquid atomic arrangements in the inner and surface regions of a nanoparticle and employ it to show that melting initiates at the outer layers.

On the melting point depression, coalescence, and chemical ordering of bimetallic nanoparticles: the miscible Ni–Pt system

“…Their lower melting points are among the properties that distinguish NPs from their bulk counterparts. 2,3 Indeed, the graph showing the melting point depression of Au as a function of NP diameter regularly appears at introductory nanotechnology classes 4 as a typical example of a physical property within the scalable regime of the nanoworld. 5…”

“…Their lower melting points are among the properties that distinguish NPs from their bulk counterparts. 2,3 Indeed, the graph showing the melting point depression of Au as a function of NP diameter regularly appears at introductory nanotechnology classes 4 as a typical example of a physical property within the scalable regime of the nanoworld. 5 One of the most characteristic consequences of the melting point depression of NPs is that it enhances their coalescence, 6 an exemplar process where bulk and nano matter's behaviour differentiate not only quantitatively but also qualitatively.…”

Accelerating the theoretical study of Li‐polysulfide adsorption on single‐atom catalysts via machine learning approaches

“…We utilize the elements in the averaged p power spectrum as the features in kernel ridge regression (KRR). SOAP power spectrum coefficients [47][48][49], and local-density expansion coefficients more in general [50][51][52][53], have been largely successful features in kernel-based and linear machine learning models for structure classification and energy regression.…”

“…SOAP representations, and in general local-density representations, when coupled with kernel and linear methods alike, are proven to be capable of accurate and fast predictions of atomistic systems properties in bulk and of molecular materials with diverse chemistries, also providing fairly accurate predictions when trained on small ($10 3 configurations) datasets. [51,54,55]. Atom-density representation further encode by design the physical symmetries (rotation, permutation, translation invariance) which rule interatomic interaction, and, though not-transparent, correlate with more interpretable "classical" descriptors, for example, coordination [56][57][58].…”

Mapping the Finite-Temperature Behavior of Conformations to Their Potential Energy Barriers: Case Studies on Si6B and Si5B Clusters

“…Based on this surprising revelation, a flurry of computational simulations based on Born–Oppenheimer molecular dynamics followed in attempting to identify the origin of this unanticipated finite-temperature behavior. − Simulations that followed over the years attempted to discern the conformational dynamics followed by the various-sized clusters and the origin of their enhanced thermal stabilities. These computational studies also further revealed that the shape- and size-specific clusters of Au and Sn have enhanced thermal stability as compared to their bulk counterparts, thereby indicating it to be a more general property than thought of among the atomic clusters. , Along with these newer aspects, detailed simulations also highlighted interesting aspects such as the fluxionality of clusters and/or role of factors such as charge in enhancing the stability of a given conformation at finite temperatures. − Several studies continue to follow where the thermal stability of a given conformation or alloy is explored using first-principles-based molecular dynamics. − …”