Polymeric organic photovoltaic (OPV) cells with polymer-fullerene bulk heterojunction are promising candidates for future low-cost, high-performance energy sources, owing to their low material and processing costs and mechanical flexibility. 1,2 High efficiencies of 6-8% have been realized in polymeric OPVs by reducing both the optical bandgap and the highest occupied molecular orbital of the semiconducting polymers, and by optimizing the morphology of polymerfullerene blend film with thermal annealing and solvent annealing. [3][4][5][6][7][8][9] Although further optimization of the bandgap and highest occupied molecular orbital level of the semiconducting polymer is possible, 3 the path to increasing OPV efficiency to 15% must include recovery of significant energy loss even in the relatively high efficiency devices demonstrated so far. [10][11][12] There are five main causes of reduced efficiency in OPV devices: energy level misalignment, insufficient light trapping and absorption, low exciton diffusion lengths, and non-radiative recombination of charges or chargetransfer excitons (CTEs), which consist of electrons at the acceptor and holes at the donor bound by Coulomb attraction, and low carrier mobilities. 11,13 In many of the most efficient polymer-fullerene OPV devices, 50% or more of the energy loss is caused by the recombination of CTEs (Reference 11). For example, if we look at the most intensively studied material system, regioregular poly(3-hexylthiophene) (P3HT) and phenyl-C61-butyric acid methyl ester (PCBM) blend, the photocurrent at the maximum power output point is only 70% of what could be extracted by externally applying a large reverse bias voltage to the device. 14 The purpose of the present work is to show how to achieve greater efficiency using a large internal electrical field provided by the permanent electrical polarization of a ferroelectric (FE) polymer layer. To understand how this innovation works, we need to understand how internal electric fields affect charge extraction efficiency. Figure 1a, CTEs form right after the photoinduced electron transfer. The next step is to separate them and enable them to contribute to the photocurrent. CTEs can be treated as a precursors of free carriers, and their bandgap sets a maximum value for the open-circuit voltage (V oc ; References 10, 15-23). The CTEs can be lost by non-radiative recombination when the dissociation driving forces (temperature and electric field) are small. The non-radiative recombination of CTEs, of course, reduces the photocurrent. In addition, the non-radiative recombination of CTEs reduces the free charge concentration, and then reduces the quasi-Fermi-energy (or chemical-potential) difference between electrons and holes, resulting in a lower V oc , reducing the output power even further. A strong reduction of the non-radiative recombination is required to reduce the fundamental efficiency loss. 11,12,24 Using the detailed balance theory, which was used to calculate the theoretical efficiency limit of a p-n junction sol...
