How Long Do Energy and Charge Transfer Take in a Model Buckminsterfullerene/Pentacene Heterojunction? Assessment of Fewest Switches Time-Dependent Density-Functional Tight-Binding with and without Long-Range Correction

Tully-type mixed time-dependent long-range corrected density-functional tight-binding/classical surface-hopping photodynamics is used to investigate the nature of and time scales for energy and charge transfer in the simplest model of an organic photovoltaic heterojunction, namely a single molecule of buckminsterfullerene (C60) together with a single molecule of pentacene. The distinction between energy and charge transfer is more difficult to make in practical calculations than might at first seem to be the case, but several criteria are used to make a clear distinction between these two phenomena. It is found that the excitation fluctuates from one molecule to the other, with the first change within about 20 fs. However it is only after 188(+/-28) fs that real charge transfer occurs. This is commensurate with what is known from experiment and very different from the severe underestimate obtained when the same calculation is repeated without a long-range correction. The long-range charge separation is not feasible to simulate in this model due to lack off appropriate charge collection sinks. We believe that these encouraging results obtained with time-dependent long-range corrected density-functional tight-binding/classical surface-hopping photodynamics opens the way, because of their intrinsic computational efficiency compared with time-dependent long-range corrected density-functional theory/classical surface-hopping photodynamics, to investigating a larger variety of increasingly realistic model organic photovoltaic heterojunctions.