The doublet-spin nature of radical emitters is advantageous for applications in organic light-emitting diodes, as it avoids the formation of triplet excitons that limit the electroluminescence efficiency of non-radical emitters. However, radicals generally show low optical absorption and photoluminescence yields. Here we explain the poor optical properties of radicals based on alternant hydrocarbons, and establish design rules to increase the absorption and luminescence yields for donor–acceptor-type radicals. We show that non-alternant systems are necessary to lift the degeneracy of the lowest energy orbital excitations; moreover, intensity borrowing from an intense high-lying transition by the low-energy charge-transfer excitation enhances the oscillator strength of the emitter. We apply these rules to design tris(2,4,6-trichlorophenyl)methyl–pyridoindolyl derivatives with a high photoluminescence quantum yield (>90%). Organic light-emitting diodes based on these molecules showed a pure-red emission with an over 12% external quantum efficiency. These insights may be beneficial for the rational design and discovery of highly luminescent doublet emitters.
In solar energy harvesting devices based on molecular semiconductors, such as organic photovoltaics (OPVs) and artificial photosynthetic systems, Frenkel excitons must be dissociated via charge transfer at heterojunctions to yield free charges. What controls the rate and efficiency of charge transfer and charge separation is an important question, as it determines the overall power conversion efficiency (PCE) of these systems. In bulk heterojunctions between polymer donor and fullerene acceptors, which provide a model system to understand the fundamental dynamics of electron transfer in molecular systems, it has been established that the first step of photoinduced electron transfer can be fast, of order 100 fs. But here we report the first study which correlates differences in the electron transfer rate with electronic structure and morphology, achieved with sub-20 fs time resolution pump–probe spectroscopy. We vary both the fullerene substitution and donor/fullerene ratio which allow us to control both aggregate size and the energetic driving force for charge transfer. We observe a range of electron transfer times from polymer to fullerene, from 240 fs to as short as 37 fs. Using ultrafast electro-optical pump-push-photocurrent spectroscopy, we find the yield of free versus bound charges to be weakly dependent on the energetic driving force, but to be very strongly dependent on fullerene aggregate size and packing. Our results point toward the importance of state accessibility and charge delocalization and suggest that energetic offsets between donor and acceptor levels are not an important criterion for efficient charge generation. This provides design rules for next-generation materials to minimize losses related to driving energy and boost PCE.