Surface diffusion in metals can be simulated with the atomistic kinetic Monte Carlo (KMC) method, where the evolution of a system is modeled by successive atomic jumps. The parametrisation of the method requires calculating the energy barriers of the different jumps that can occur in the system, which poses a limitation to its use. A promising solution to this are machine learning methods, such as artificial neural networks, which can be trained to predict barriers based on a set of pre-calculated ones. In this work, an existing neural network based parametrisation scheme is enhanced by expanding the atomic environment of the jump to include more atoms. A set of surface diffusion jumps was selected and their barriers were calculated with the nudged elastic band method. Artificial neural networks were then trained on the calculated barriers. Finally, KMC simulations of nanotip flattening were run using barriers which were predicted by the neural networks. The simulations were compared to the KMC results obtained with the existing scheme. The additional atoms in the jump environment caused significant changes to the barriers, which cannot be described by the existing model. The trained networks also showed a good prediction accuracy. However, the KMC results were in some cases more realistic or as realistic as the previous results, but often worse. The quality of the results also depended strongly on the selection of training barriers. We suggest that, for example, active learning methods can be used in the future to select the training data optimally.

Sharp nanoscale tips on the metal surfaces of electrodes enhance locally applied electric fields. Strongly enhanced electric fields trigger electron field emission and atom evaporation from the apexes of nanotips. Together, these processes may explain electric discharges in the form of small local arcs observed near metal surfaces in the presence of electric fields, even in ultra-high vacuum conditions. In the present work, we investigate the stability of nanoscale tips by means of computer simulations of surface diffusion processes on copper, the main material used in high-voltage electronics. We study the stability and lifetime of thin copper (Cu) surface nanotips at different temperatures in terms of diffusion processes. For this purpose we have developed a surface kinetic Monte Carlo (KMC) model where the jump processes are described by tabulated precalculated energy barriers. We show that tall surface features with high aspect ratios can be fairly stable at room temperature. However, the stability was found to depend strongly on the temperature: 13 nm nanotips with the major axes in the <110 > crystallographic directions were found to flatten down to half of the original height in less than 100 ns at temperatures close to the melting point, whereas no significant change in the height of these nanotips was observed after 10 mu s at room temperature. Moreover, the nanotips built up along the <110 > crystallographic directions were found to be significantly more stable than those oriented in the <100 > or <111 > crystallographic directions. The proposed KMC model has been found to be well-suited for simulating atomic surface processes and was validated against molecular dynamics simulation results via the comparison of the flattening times obtained by both methods. We also note that the KMC simulations were two orders of magnitude computationally faster than the corresponding molecular dynamics calculations.