Thiemann, Fabian Lukas; (2022) Properties of Low-dimensional Materials Explored with Machine Learning Potentials. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text flthiemann_final.pdf - Accepted Version Download (45MB) | Preview

Abstract

The wonder material graphene and its inorganic analogue, hexagonal boron nitride, have attracted widespread interest due to their fascinating and technologically relevant properties. Being two-dimensional and merely one-atom thick, their properties, however, strongly depend on the atomic morphology with many physical phenomena being evoked by the wrinkled and corrugated topology. Controlling and modifying the intrinsic ripples, therefore, could allow the tailor-made design of nanodevices. In the first part of this thesis, we explore routes to nanoengineer the morphology of both materials by performing machine learning-driven atomistic simulations. In chapter 3, we develop a novel machine learning potential for various phases of hexagonal boron nitride investigating how thermally and mechanically induced ripples compare between the two materials. As an alternative to introducing strain, in chapter 4 we show how defects can be used to significantly alter and control graphene’s atomic structure. We reveal that the extent of the induced morphological changes is directly related to the preferred geometrical orientation and the interactions between defects. The second part of this work is dedicated to the one-dimensional counterparts of graphene and hexagonal boron nitride, carbon and boron nitride nanotubes, which show great promise for desalination and blue energy harvesting. In chapter 5, we aim to address and understand the puzzling experiments observing a contrasting flow behaviour of water on both materials. Our large-scale simulations of the water transport in single-wall nanotubes provide an accurate benchmark and reveal that the fast water transport on carbon is governed by facile oxygen motion, whereas the higher friction on boron nitride arises from specific hydrogen-nitrogen interactions. Overall, this thesis illustrates how subtle details such as lattice imperfections, atomic vibrations, and differences in the electronic structure can significantly affect the material properties offering great potential to control the behaviour of matter at the nanoscale.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item