Explorations of boron-based materials through theoretical simulations

This thesis focuses on boron-based materials, notable for their structural complexity and unique combination of ceramic and metallic properties. These properties typically result in materials with high mechanical strength, electrical conductivity, and melting points. Among these materials are MAB phases, a family of layered materials comprised of a transition metal (M), an A-element (typically an element from Group 13-14), and boron (B). The layered nature of these materials provides a pathway towards the realization of 2D materials, coined MBenes (or boridene), through chemical exfoliation. While the potential for discovering novel materials is immense, their realization often demands extensive experimental efforts. Theoretical models may here be used as a filter by guiding experimental endeavours. The work presented herein aims to leverage theoretical models and to develop frameworks suitable for reliable thermodynamical predictions in hope of the discovery of additional boron-based materials. First-principles calculations, particularly density functional theory (DFT), have extensively been employed in this thesis to determine the ground state energy of materials and predict their stability or tendency to decompose. However, first-principles calculations typically rely on a pre-defined crystal structure which may be constructed through a priori information or obtained through the use of crystal structure prediction (CSP) frameworks. We herein explore both of these approaches by i) systematically substituting elements in known low-energy structures, and ii) deriving novel low-energy structures by combining CSP with cluster expansion (CE) models. The first approach is herein exemplified when considering the low-energy structures of V3B2 (P4/mbm) and Cr5B3 (I4/mbm). These structures are comprised of two M-sites in addition to boron and thus form the general compositions M’2M’’B2 and M’4M’’B3, respectively. In a follow-up project, this approach was refined by probing
Funding agencies: The Swedish Research Council (grant numbers 2019-05047, 2022-06099, 2022-06725 and 2018-05973) and the Knut and Alice Wallenberg (KAW) foundation (grant number KAW 2020.0033)