Recent theories suggest that low frequency dynamic intramolecular and intermolecular motions in organic semiconductors (OSCs) are critical to determining the hole mobility. So far, however, it has not been possible to probe these motions directly experimentally and therefore no unequivocal and quantitative link exists between molecular-scale thermal disorder and macroscale hole mobility in OSCs. Here we use inelastic neutron scattering to probe thermal disorder directly by measuring the phonon spectrum in six different small molecule OSCs, which we accurately reproduce with first principles simulations. We use the simulated phonons to generate a set of electron-phonon coupling parameters. Using these parameters, the theoretical mobility is in excellent agreement with macroscopic measurements. Comparison of mobility between different materials reveals routes to improve mobility by engineering phonon and electron-phonon coupling.

Understanding the nature of dopant dynamics in the solid state is critical for improving the longevity and stability of organic electronic devices and for optimizing the doping-induced solubility control (DISC) patterning method. In this work, we use quasi-elastic neutron scattering (QENS) and fluorescence quenching techniques to develop a comprehensive picture of both the microscopic and macroscopic dynamics of the soluble p-type molecular dopant tetrafluoromethyloxycarbonyltricyanoquinodimethane (F4MCTCNQ) in the conductive polymer poly(3-hexylthiophene-2,5-diyl) (P3HT). Specifically, fast dynamics (ps-ns) of the dopant, such as the methyl and the methoxycarbonyl group rotations, are observed in QENS experiments. From confocal fluorescence microscope experiments, longer-range/slower dopant diffusion (ms-days) is captured. However, in order to fit these data, it is necessary to incorporate a Langmuir isotherm equilibrium between the neutral and ionized dopant molecules. Ionized F4MCTCNQ is strongly favored by the equilibrium, but it diffuses 3 orders of magnitude slower than neutral species. Moreover, the macroscopic diffusion is found to depend mostly on the minority concentration of neutral dopant molecules in the film. Finally, the global diffusion coefficient of the monoester-substituted dopant F4MCTCNQ is shown to be more than an order of magnitude smaller than that of the widely used dopant F4TCNQ.