STUDY OF ELECTRONIC AND MAGNETIC PROPERTIES OF BILAYER GRAPHENE NANOFLAKES AND BIMETALLIC CHALCOGENIDES USING FIRST-PRINCIPLES DENSITY FUNCTIONAL THEORY AND MACHINE LEARNING

Graphene, a one-atom-thick material, has been a wonder material since its discovery because of its superlative electronic, mechanical, and optical properties. When a layer of graphene is rotated over another layer, it exhibits many intriguing behaviors, ranging from superconductivity to the anomalous Hall effect to ferromagnetism at a magic angle of 1°, and hence the twisted bilayer graphene has been the subject of intense research in recent years. The surge in interest in this moiré structure can be attributed to the emergence of electronic flat minibands near the magic angle. Here, we studied the electronic and magnetic properties of twisted bilayer graphene nanoflakes using first-principles density functional theory. In my first project, we found that when a layer of graphene nanoflakes is rotated over another layer, the nonmagnetic energy gap of the flakes changes due to the change in interlayer interaction between the layers. When the energy gap approaches zero, an electronic instability occurs, and a ferromagnetic gap state emerges spontaneously to lower the energy. Unlike the observed ferromagnetism at a magic angle in the graphene bilayer, we noticed the ferromagnetic phase appearing aperiodically between 0 and 30° in the twisted nanoflakes. The origin of electronic instability at various twist angles is ascribed to the several higher-symmetry phases that are broken to lower the energy resulting from an aperiodic modulation of the interlayer interaction in the nanoflakes. Besides unraveling a spin-pairing mechanism for the reappearance of the nonmagnetic phase, we found orbitals at the boundary of nanoflakes contributing to ferromagnetism. Subsequently, we investigated the role of electronic pressure on the magic-twist angle. Here, using first-principles density functional theory, we demonstrated that the magic twist angle in a bilayer graphene nanoflake at which the transition from a nonmagnetic to a ferromagnetic phase occurs can be tuned by exerting uniaxial electronic pressure. Electronic pressure, which provides another route to control the inter-layer electronic coupling energy, is simulated by varying the interlayer spacing in the nanoflake. In my third project, we designed more than 4348 Fe-based bimetallic chalcogenides and studied their magnetic properties using density functional theory. Further, we designed and developed a stacked machine learning model to predict the magnetic moment in all possible configurations of Fe-based bimetallic chalcogenides made of transition metals Ni, Co, Cr, or Mn and chalcogen elements S, Se, or Te. The final stacked machine learning model predicts the magnetic moment with an R2 score of 92%