Search

An analytical energy gradient formula for the density-matrix-based linear-scaling divide-and-conquer (DC) self-consistent field (SCF) method was proposed in a previous paper by Yang and Lee (YL) [J. Chem. Phys. 103, 5674 (1995)]. Since the formula by YL does not correspond to the exact gradient of the DC-SCF energy, we derive the exact formula by direct differentiation, which requires solving the coupled-perturbed equations while including the inter-subsystem coupling terms. Next, we present an alternative formula for approximately evaluating the DC-SCF energy gradient, assuming the variational condition for the subsystem density matrices. Numerical assessments confirmed that the DC-SCF energy gradient values obtained by the present formula are in reasonable agreement with the conventional SCF values when adopting a reliable buffer region. Furthermore, the performance of the present method was found to be better than that of the YL method. [ABSTRACT FROM AUTHOR]

We propose a new implementation of Ehrenfest molecular dynamics based on the configuration interaction theory using configuration state functions (CSF) as basis set originally proposed by Amano and Takatsuka (J. Chem. Phys. 2005, 122, 084113). Our development consists of two independet new features. The first one deals with the problem on how to 'identify' the molecular orbitals at a simulation time step in terms of those at the previous time step. By giving an exact expression of CSF Ehrenfest method, this problem naturaly vanishes. To actually perform this method, the concept of molecular orbital connection which allows the MOs to be noncanonical is necessary. The second feature of our method is aimed to reduce the computational cost. We propose an approximaion to effectively perform the time propagation of the electron wavefunction. Due to the analogy to the locally diabatic representation method, we name our method locally quasi-diabatic representation method. In the present work, these two new features were combined and employed to perform test computations. © 2016 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]