Method development

We work on method development in the field of many body perturbation theory at the GW level, as well as in time dependent density functional theory (TDDFT). The latter method has become increasingly popular in quantum chemistry and is often used to predict the optical absorption spectra of compounds with hundreds of atoms. By extending the DFTB method to time dependent phenomena even larger structures are now accessible. In linear response, for weak external fields, excited state forces have been derived, which allow geometry relaxations also in excited states. This is a prerequisite for the evaluation of luminescence spectra of nanoparticles. Beyond linear response, we directly propagate the Kohn-Sham orbitals in time and perform quantum-classical molecular dynamics in the Ehrenfest approach. In this way, ultrafast chemical photoreactions which involve intersections of electronic states may be modeled. Also, vibrational motion induced by femtosecond laser pulses can be monitored. Recently, we apply TDDFT also to the electronic transport through single molecules, in order to simulate transients, alternating currents and photo-induced switching phenomena.


The GW approximation has been applied in solid state physics with great success to determine the band-structure and predict the correct band gap of semiconductors. We developed an implementation of GW on the basis of the DFTB method, which can be used to obtain the quasi-particle spectrum of rather large finite systems. An adaption for periodic systems is currently under way. In the context of molecular electronics, the GW approximation is an interesting alternative to conventional DFT based approaches and holds the promise to overcome the severe overestimation of currents obtained in the latter scheme. We already implemented an NEGF-GW scheme and are currently working on an improved and consistent description of the leads.


A somewhat related development is the recent implementation of range-separated functionals for the DFTB method. The goal is to reduce the self-interaction error in DFTB which manifests in a variety of failures, ranging from an underestimation of the band gap, the instability of polaronic defects, problems with charge-transfer excited states up to overly delocalized electron densities.


Back



Project related publications

Rigorous Excited-State Entropy in Finite-Temperature Time-Dependent Density Functional Theory
Niehaus, T.A.
J. Phys. Chem. Lett. 17, 4546 (2026)

Phonon-induced band gap renormalization by dielectric dependent global hybrid density functional tight binding
van der Heide, T., Hourahine, B., Aradi, B., Frauenheim, T., and Niehaus, T.A.
Phys. Rev. B 109, 245103 (2024)

Implementation and benchmark of a long-range corrected functional in the density functional based tight-binding method
Lutsker, V., Aradi, B. und Niehaus, T.
The Journal of Chemical Physics 143, 184107 (2015)

Accurate Hydrogen Bond Energies within the Density Functional Tight Binding Method
Dominguez, A., Niehaus, T. und Frauenheim, T.
The Journal of Physical Chemistry A 119, 3535 (2015)

A New Efficient Method for Calculation of Frenkel Exciton Parameters in Molecular Aggregates
Plötz, P., Niehaus, T. und Kühn, O.
J. Chem. Phys. 140, 174101 (2014)

Extensions of the time-dependent density functional based tight-binding approach
Dominguez, A., Aradi, B., Frauenheim, T., Lutsker, V. und Niehaus, T.
J. Chem. Theory Comput. 9, 4901 (2013)