University of Twente Student Theses
Modelling and simulation of thermoelectric materials with a superionic phase transition using on-the-fly machine learning force fields.
Hoeven, Rick van der (2022) Modelling and simulation of thermoelectric materials with a superionic phase transition using on-the-fly machine learning force fields.
PDF
7MB |
Abstract: | Superionic conductors (SIC’s) posses liquid like diffusivity in solid state. Mate rials possessing a superionic phase transition can be researched using first principles calculations. These take long for for bigger system sizes however, and can become impracticable quickly. There is a new method though which can improve on this; on-the-fly machine learning force fields. This method uses first principle calculations to then use these to quicken calculations. It is however not obvious if this method will work; in the superionic phase of the SIC, diffusivity is high and the local environment of atoms varies a lot. In this BSc thesis, we will show that machine learning force fields can indeed model a superionic phase transition for the semiconductor KAg3Se2. Although the experimental superionic structure is not achieved, we will show that we do indeed observe a superionic phase and a phase transition. Using this, we will convey that we can also get accurate values for diffusion, which are comparable to lit erature. Finally, we will show that the error of the force field is in reasonable bounds, compared to first principles calculations. |
Item Type: | Essay (Bachelor) |
Faculty: | TNW: Science and Technology |
Subject: | 33 physics, 51 materials science |
Programme: | Applied Physics BSc (56962) |
Link to this item: | https://purl.utwente.nl/essays/92459 |
Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Repository Staff Only: item control page