Electron Transfer Keywords in the Jaguar Input File
The etransfer keyword allows you to calculate the coupling matrix element for charge transfer between two structures that are in a specified geometric arrangement. This capability is important for charge mobility calculations on materials. Setting etransfer=1 requests calculation of the electronic coupling matrix elements.
The coupling is calculated between an initial state and a final wave state, according to the following expression:
where Hii, Hff, and Hif are matrix elements of the system Hamiltonian, and Sif is the overlap matrix element. As different wave functions are used for the initial and final states, the orbitals for these states are not orthogonal, so the Hamiltonian and overlap matrix elements for the transfer (Hif and Sif) are calculated in a biorthogonal set, generated by singular value decomposition of the overlap matrix between the initial and final orbital sets.
The calculation of the coupling matrix element requires input wave functions for the initial state, with the charge localized in one part of the composite system, and the final state, with the charge localized on the other part of the composite system. You can either perform calculations on each structure, both neutral and charged, or use constrained DFT (CDFT) to perform calculations on the composite system, one with the charge constrained on one structure and one with the charge constrained on the other. The first method does not include the effect of polarization of the uncharged structure by the charged structure, as they are treated independently; the second method does include polarization effects.
The wave functions are taken from restart files (jobname.01.in), which are specified with directives, placed at the beginning of the input file. Four restart files are needed if calculations on the individual structures are used, specified with the ET1A, ET2A, ET1D and ET2D directives, and two restart files are needed if CDFT calculations on the composite system are used, specified with the ET1DA and ET2DA directives. The directives are given in Table 1. The calculations on the initial and final state structures must use the same coordinates, i.e. the structure is frozen. The input file for the charge transfer calculation must be set up for the composite system, and if separate calculations are done on each structure, the coordinates of the donor must be placed before the coordinates of the acceptor in the input file.