Tutorial 7: Input parameters and output variables for time-dependent density-functional theory

The interface for defining and changing input parameters, and recovering output results, in time-dependent density-functional theory (TDDFT) simulations is similar to the Schrödinger equation case. Here we summarize the main features of TDDFT simulations and refer to its documentation page for more details. The documentation page also includes examples for each set of observable that can be calculated on the fly while the propagation is performed.

• External driving field: driving fields are taken in the dipole approximation and the propagation can be done using the length or velocity gauge.
• Absorbing boundaries: both complex absorbing potentials and mask functions can be used as absorbers.
• Propagators: currently, propagation is only available for grid-based models (not basis-set discretization), with orders 2, 4, or 6 in time.
• Propagation results: besides the propagation of the wave function(s), propagators can calculate and save the dipole signal, energy, ionization, wave function, as well as the result of a user-defined installable function. Each resulting output is stored in a separate structure in the TDDFT-propagator object. Each requested output can independently define its own time sampling. For field-driven calculations, the exact values for the driving field used in the simulation at the time each output is saved can be added to each output structure.
• Restart file: a restart file can be generated, from which the propagation can be restarted in case it gets stopped before the end of the propagation (without having to start over from the beginning).