Non-adiabatic Couplings\nin Surface Hopping with Tight\nBinding Density Functional Theory: The Case of Molecular Motors

Nonadiabatic molecular dynamics (NAMD) has become an\nessential\ncomputational technique for studying the photophysical relaxation\nof molecular systems after light absorption. These phenomena require\napproximations that go beyond the Born–Oppenheimer approximation,\nand the accuracy of the results heavily depends on the electronic\nstructure theory employed. Sophisticated electronic methods, however,\nmake these techniques computationally expensive, even for medium size\nsystems. Consequently, simulations are often performed on simplified\nmodels to interpret the experimental results. In this context, a variety\nof techniques have been developed to perform NAMD using approximate\nmethods, particularly density functional tight binding (DFTB). Despite\nthe use of these techniques on large systems, where ab initio methods\nare computationally prohibitive, a comprehensive validation has been\nlacking. In this work, we present a new implementation of trajectory\nsurface hopping combined with DFTB, utilizing nonadiabatic coupling\nvectors. We selected the methaniminium cation and furan systems for\nvalidation, providing an exhaustive comparison with the higher-level\nelectronic structure methods. As a case study, we simulated a system\nfrom the class of molecular motors, which has been extensively studied\nexperimentally but remains challenging to simulate with ab initio\nmethods due to its inherent complexity. Our approach effectively captures\nthe key photophysical mechanism of dihedral rotation after the absorption\nof light. Additionally, we successfully reproduced the transition\nfrom the bright to dark states observed in the time-dependent fluorescence\nexperiments, providing valuable insights into this critical part of\nthe photophysical behavior in molecular motors.