Related papers: Understanding solid nitrogen through machine learn…
Understanding material surfaces and interfaces is vital in applications like catalysis or electronics. By combining energies from electronic structure with statistical mechanics, ab initio simulations can in principle predict the structure…
Molecular dynamics simulations provide a versatile framework to study interfacial heat transport, but their accuracy remains limited by the accuracy of available interatomic potentials. In the past, researchers have adopted the use of…
Unraveling the atomistic and the electronic structure of solid-liquid interfaces is the key to the design of new materials for many important applications, from heterogeneous catalysis to battery technology. Density functional theory (DFT)…
Precise prediction of phase diagrams in molecular dynamics (MD) simulations is challenging due to the simultaneous need for long time scales, large length scales and accurate interatomic potentials. We show that thermodynamic integration…
A key problem in the modern study of AI is predicting and understanding emergent capabilities in models during training. Inspired by methods for studying reactions in quantum chemistry, we present the ``2-datapoint reduced density matrix".…
This study utilises linear-scaling density functional theory (DFT) and develops a new machine-learning potential for carbon and nitrogen (GAP-CN), based on the carbon potential (GAP20), to investigate the interaction between carbon…
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…
The crystal nucleation from liquid in most cases is too rare to be accessed within the limited timescales of the conventional molecular dynamics (MD) simulation. Here, we developed a "persistent embryo" method to facilitate crystal…
Chemical segregation and structural transitions at interfaces are important nanoscale phenomena, making them natural targets for atomistic modeling, yet interatomic potentials must be fit to secondary physical properties. To isolate the…
Gaussian Process Regression-based Gaussian Approximation Potential has been used to develop machine-learned interatomic potentials having density-functional accuracy for free sodium clusters. The training data was generated from a large…
A Spectral Neighbor Analysis (SNAP) machine learning interatomic potential (MLIP) has been developed for simulations of carbon at extreme pressures (up to 5 TPa) and temperatures (up to 20,000 K). This was achieved using a large database of…
The length and time scales of atomistic simulations are limited by the computational cost of the methods used to predict material properties. In recent years there has been great progress in the use of machine learning algorithms to develop…
We introduce a machine-learning interatomic potential for tungsten using the Gaussian Approximation Potential framework. We specifically focus on properties relevant for simulations of radiation-induced collision cascades and the damage…
This thesis can be divided into two independent parts. In the first part of this thesis, we focus on studying the kinetic pathways of nucleation in colloidal systems. In Chapter 2, we briefly introduce the relevant theory of nucleation,…
A Gaussian approximation machine learning interatomic potential for platinum is presented. It has been trained on DFT data computed for bulk, surfaces and nanostructured platinum, in particular nanoparticles. Across the range of tested…
The folding formalism for the nucleon-nucleus optical potential and inelastic form factor is applied to study elastic and inelastic proton scattering on 30-40S isotopes. A recently developed realistic density dependent M3Y interaction, well…
In the present paper, using a molecular dynamics simulation, we study a nature of melting of a two-dimensional ($2D$) system of classical particles interacting through a purely repulsive isotropic core-softened potential which is used for…
The quest for efficient and robust deep learning models for molecular systems representation is increasingly critical in scientific exploration. The advent of message passing neural networks has marked a transformative era in graph-based…
Machine learning interatomic potentials (MLIPs) are routinely used to model diverse atomistic phenomena, yet parameterizing them to accurately capture solid-state phase transformations remains difficult. We present error metrics and…
We introduce machine learning models of quantum mechanical observables of atoms in molecules. Instant out-of-sample predictions for proton and carbon nuclear chemical shifts, atomic core level excitations, and forces on atoms reach…