Articial neural network potentials for studying structure and properties of bare, doped, and protected Au nanoclusters

Abstract

Atomistic simulations of metal nanoclusters critically depends upon the accurate description of potential energy surface. First principle methods such as density functional theory (DFT) can describe atomic interactions accurately but unfortunately the computations are not a ordable for experimentally relevant sizes. In this thesis it is demonstrated that arti cial neural network (ANN) based interatomic potentials are accurate in describing interatomic interactions and also make the calculations a ordable. In terms of computational speed, ANN's are very fast and hence it allows us to run molecular dynamic simulations upto a time scales of 3 ns on a single CPU machine for a medium sized nanocluster. ANN potentials are constructed by training a set of DFT data of nanoclusters. The accuracy and applicability of ANN potentials are described based on the Nan nanoclusters (0:4 n 1:2 nm) as a benchmark example. Further, ANN potentials are explored for bare (Aun), doped ((AgAu)55) and protected Au nanoclusters (AunSHm) to study the e ect of size, composition on structure and dynamical properties.This thesis presents a detailed study on the most probable structure of Au nanoclusters (Aun, 17 n 58), Ag-Au nanoalloys ((AgAu)55) and thio protected Au nanoclusters (AunSHm, 15 n 38) at zero kelvin temperature. In case of bare Au nanoclusters, molecular dynamics simulations were performed in canonical (NVT) and microcanonical (NVE) ensembles on Au17, Au34, Au58. The study shows that there is a dynamical coexistence of solidlike and liquidlike phases near melting transition. We estimate the probability at nite temperatures for set of isomers lying below 0.5 eV from the global minimum structure. In case of Au17 and Au58, the properties can be estimated using global minimum structure at room temperature, while for Au34, global minimum structure is not a dominant structure even at low temperatures. In case of (AgAu)55, ANN based interatomic potential was used for understanding the structure, dynamics and thermal stability. Knowledge of composition - temperature (c-T) phase diagram is essential due to explicit dependence of properties on composition and temperature. Experimentally generating the phase diagrams are very much challenging and therefore theoretical insight is necessary. The present work based on c-T phase diagram, surface area, surface charge, probability of isomers and landau free energies, supports the enhancement of catalyticproperty of Ag-Au nanoalloys by incorporation of Ag up to 24 % by composition in Au nanoparticles as experimentally proved. The phase diagram shows that there is a coexistence temperature range of 70 K for Ag28Au27 compared to all other compositions. In addition, powerspectrum coe cients derived from spherical harmonics as an order parameter was proposed to calculate landau free energies. For thio protected Au nanoclusters, e ect of composition of SH for di erent sizes of nanocluster was studied using ANN based interatomic potential. The probability study at nite temperature shows that there is a increase in number of isomers at room temperature if we decrease SH percentage. UV-visible spectra was utilized to probe the structure of nanoclusters.