Institutional Repository of Vilnius University
Aktyvaus mašininio mokymosi metodais paremtų tarpatominių sąveikos potencialų skirtų fosforo rūgšties sistemoms vystymas ir taikymai /
Abstract
dc:descriptionThis work discusses widespread computational chemistry methods, such as density functional theory (DFT), force fields (FF) and machine-learning force fields (MLFF), particularly their essence, differences and features. MLFFs is presented as a modern and superior method. Additionally, orthophosphoric acid \ce{H3PO4} features are discussed. Two MLFFs are constructed in this work: one being dedicated for describing pure \ce{H3PO4} (PPA) systems, and the other being dedicated for describing \ce{H3PO4} and \ce{H2O} (APA) systems. The process of constructing an MLFF is discussed as well as the simulation parameters necessary for constructing accurate and correct MLFFs. Additionally, this work analyses MLFFs' accuracy by comparing some of the results to experimental findings and DFT calculations. It is found that the constructed MLFFs give accurate results, which are validated by DFT computational studies and experimental findings.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Mockus, Gabrielius,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308330417&prefLang=en_US