(451f) Improved Methods for Assigning Point Charges, Bond Orders, and Magnetic Moments In Complex Materials

Net atomic charges (NACs), atomic spin moments (ASMs), and effective bond orders (EBOs) are important concepts for understanding many chemical systems. Although natural population analysis (NPA) provides reliable NACs, ASMs, and EBOs for nonperiodic materials, it is not yet available for periodic materials and does not specifically optimize NACs to reproduce the electrostatic potential V(r). Herein, we describe the development of density derived electrostatic and chemical (DDEC) population analysis for computing NACs, ASMs, and EBOs in both periodic and nonperiodic materials. This method analyzes a material’s total and spin densities computed via density functional theory, coupled-cluster theory, or another ab initio method. To produce NACs suitable for constructing force-fields used in atomistic simulations and for understanding electron transfer during chemical reactions, the atomic charge distributions are simultaneously optimized to reproduce the chemical states of atoms and V(r) outside a material’s electron distribution. Our method can be used to compute ASMs for both collinear and non-collinear magnetic materials. The EBO includes both traditional covalent bonding (e.g., H₂) and dispersion bonding (e.g., Kr₂). In this talk, we will describe the basic theory behind the DDEC population analysis method and use it to study (a) a spin-charge-ordering reconstruction anti-phase domain boundary on the magnetite 001 surface, (b) the non-collinear single molecule magnet Fe₄C₄₀H₅₂N₄O₁₂, (c) CO₂ adsorption in metal organic frameworks for carbon capture applications, and (d) other periodic and nonperiodic materials.