Tully's fewest-switches surface hopping (FSSH) algorithm is probably the most popular modern approach for modeling photo-excited nonadiabatic dynamics, and yet we know that the FSSH model has one large shortcoming, its failure to treat decoherence correctly. In previous research, we shown that FSSH's overcoherence problem is manifested at long times if nuclei visit regions of derivative coupling more than once. In this poster, we will show a very troubling ramification of this same problem. Our focus is the spin-boson model for the case of a small diabatic coupling parameter (V ), where nuclei should cross a region of nonadiabatic coupling repeatedly before an electron is transferred. Using FSSH dynamics, we calculate the transition rates between diabatic surfaces� ideally, FSSH should yield the celebrated Marcus rates for electron transfer. Instead, we show that FSSH yields an incorrect scaling with diabatic coupling (~ V ). To recover the correct scaling (~ V 2), we must modify FSSH to include decoherence events, thus proving the absolute need for decoherence when modeling electron-transfer dynamics and highlighting the strengths of our new (decoherence-included) augmented FSSH (A-FSSH) approach.

Surface hopping predicts the incorrect scaling for Marcus's electron transfer rate: Why decoherence saves the day

244th National Meeting (2012)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.