Speaker
Description
Phase-dependent shear behaviour in MAB phases driven by Al layers and defects using ML potentials
Priyanshu Sorout$^{1}$, Shuyao Lin$^{1,2}$, Paul H.Mayrhofer$^{1}$, Davide G.Sangiovanni$^{2}$, Nikola Koutná$^{1,2}$
$^{1}$Institute of Materials Science and Technology, TU Wien, Vienna, Austria
$^{2}$Department of Physics, Chemistry and Biology (IFM), Linköping University, Sweden
Abstract
Layered transition-metal aluminum borides (MAB phases) are promising for high-temperature applications, combining ceramic-like oxidation resistance with metal-like fracture toughness by incorporating compliant Al layers within the MBene structure. Although their tensile properties have been extensively studied, shear deformation, especially relevant given their layered yet puckered architecture, remains largely unexplored. To address this gap, using purpose-trained ML potentials, we systematically investigate atomic and nanoscale finite-temperature shear deformation up to 10,000 atoms in MAB phases across Group IV-VI transition metals. We focus on three representative polymorphs: 212-, 222-, and 314-type. Our results reveal strongly phase-dependent ideal shear strengths, ranging from 10 GPa to 25 GPa across nine compositions. We find that mechanical instability is initiated by the gliding Al interlayer, which serves as the preferred slip plane due to the relatively weak, metallic Ti-Al bonding compared to the covalent B-Ti framework. Furthermore, introducing Al vacancies reduces the ideal shear strength and facilitates interlayer sliding, highlighting vacancy concentration as a key parameter for layer separation into MBenes. Altogether, this study establishes Al-layer-driven shear behaviour in MAB phases as a function of structural polymorph and point defects, using reliable machine-learning potentials, and provides guidance for the synthesis of future two-dimensional materials.