Project title : Advanced electron transport materials for application in organic photovoltaics ( OPV )

This project is focused on discovering efficient, less expensive and tunable organicbased electron transport materials for application in organic photovoltaics (OPV) in an effort to replace more commonly used fullerene-based materials. The emphasis during the first six months of this project was more on the synthesis of new materials, with less emphasis on device fabrication that will ramp up during the next stage. Several novel molecules were designed, prepared and characterized from simple, minimal step, moderate yield, and inexpensive synthetic processes. These novel materials were prepared by chemically linking together conjugated electron deficient moieties of phenyland naphthyl-imides, dicyanoimidazoles and benzothiadiazoles. Initial thermal, optical and electronic characterization of these materials suggests they have favorable properties such as relatively hightemperature thermal transitions, favorablylocated energy levels, visible light absorption, high solubility in commonly used organic solvents, and high purity. For example, they are excellent candidates to prepare solution processed bulk heterojunction (BHJ) OPVs by combining with regio-regular poly(3-hexylthiophene), the most commonly studied donor material used in bulk heterojunction OPVs. Introduction & Background Organic photovoltaics (OPV) have recently reached power conversion efficiencies (PCE) of 8.3% and extrapolated lifetimes of >30 years bringing this area of technology closer to commercialization. Common to both of these reports are the use of fullerenes as Figure 1. Fullerene based materials used in current state-of-the-art OPVs. the electron acceptor/transport materials. In the case of Chen et al, bulk heterojunction (BHJ) solar cells are reported where PC70BM (Figure 1) is used as the acceptor together with a novel donor polymer. For the report by Schwartz et al, vacuum deposited tandem devices are prepared using the “naked” C60 molecule as the acceptor with small molecule donor materials. Despite these exciting developments for OPVs, from a materials perspective fullerenes tend to have low absorption in the visible range, produce devices with relatively low open circuit voltages (Voc), and are very difficult to synthesize and purify, which leads to higher costs. The most commonly used fullerene compounds for solution processed OPVs are shown in Figure 1. These compounds are functionalized with side chains that make them more soluble for solution processing. Of the fullerene derivatives in Figure 1, PC60BM is the most commonly used primarily because it is the least expensive ($300/g). PC70BM, PC84BM, and bis-PC70BM have been recently shown to improve device performance, but the higher prices for these materials, $800/g, $10,000/g, and $1200/g respectively, ultimately translates into higher cost OPV cells. With regard to new materials development for application in OPVs, to date most studies have focused on electron donor semiconductors, with significantly less work being reported on new electron acceptor materials. New acceptor materials that could address and solve the issues of fullerenes described above could push PCE levels to 10% or higher and reduce the cost. For example, replacing the weakly absorbing fullerenes with more strongly absorbing acceptors should lead to efficient light harvesting below 700nm (Figure 2a) and increased short circuit current densities (Jsc). Assuming a peak external quantum efficiency (EQE) of 0.75, we estimate the Jsc can be improved from 15.2 mA/cm in the record device to 17.2 mA/cm (Figure 3). Even in optimized state-of-the-art devices, a significant fraction of the incident power is lost due to the large offset in the HOMO and LUMO levels between the two materials (Figure 2b). It has been shown that with properly chosen electrodes, the difference in energy between the HOMO of the donor and the LUMO of the acceptor is typically 0.3V larger than the Voc of the device. Although a HOMO level offset (and LUMO level offset) of about 0.3V is required to split the photogenerated excitons, a large HOMO and (a) (b) Figure 2. (a). Potential improvement in EQE over the current record device from using new acceptor materials. (b) Energy levels in the record bulk heterojunction solar cell between the donor polymer PBDTTT-CF and the fullerene acceptor PCBM. LUMO offset results in a reduction in the Voc that the device can produce. If new acceptor materials can be designed with HOMO and LUMO levels that are less electronegative (red arrows in Figure 2b), the reduced HOMO and LUMO offsets would result in higher Voc. Assuming that Voc = ELUMO,Acceptor –EHOMO,Donor – 0.3V, reducing the HOMO and LUMO energy offsets to 0.3V would increase the Voc of this device to 1.2V (Figure 3). Fill factors of 0.7 have been obtained in the commonly studied system of regio-regular poly(3hexylthiophene) (P3HT) with PC60BM BHJ devices and should also be attainable in devices with new acceptors that have electron mobilites comparable to the fullerenes. Assuming a fill factor of 0.7, the improvements in the Voc (1.2V) and Jsc (17.2 mA/cm) would result in a 14.4% efficient device. Of the few non-fullerene acceptor materials that have been reported in the literature, perylene derivatives, cyano poly(phenylenevinylenes) (CN-PPV), and 4,5dicyanoimidazole (Vinazene) based materials have been the most promising, resulting in top PCEs of ~2% when combined with a donor material. As the potential for PCEs greater than 10% is a reality, there needs to be a systematic fundamental study on new acceptor materials for application in OPVs that includes chemical synthesis, materials characterization, and device physics.