The performance of polymer:fullerene bulk heterojunction solar cells is heavily influenced by the interpenetrating nanostructure formed by the two semiconductors because the size of the phases, the nature of the interface, and molecular packing affect exciton dissociation, recombination, and charge transport. Here, X‐ray diffraction is used to demonstrate the formation of stable, well‐ordered bimolecular crystals of fullerene intercalated between the side‐chains of the semiconducting polymer poly(2,5‐bis(3‐tetradecylthiophen‐2‐yl)thieno[3,2‐b]thiophene. It is shown that fullerene intercalation is general and is likely to occur in blends with both amorphous and semicrystalline polymers when there is enough free volume between the side‐chains to accommodate the fullerene molecule. These findings offer explanations for why luminescence is completely quenched in crystals much larger than exciton diffusion lengths, how the hole mobility of poly(2‐methoxy‐5‐(3′,7′‐dimethyloxy)‐p‐phylene vinylene) increases by over 2 orders of magnitude when blended with fullerene derivatives, and why large‐scale phase separation occurs in some polymer:fullerene blend ratios while thermodynamically stable mixing on the molecular scale occurs for others. Furthermore, it is shown that intercalation of fullerenes between side chains mostly determines the optimum polymer:fullerene blending ratios. These discoveries suggest a method of intentionally designing bimolecular crystals and tuning their properties to create novel materials for photovoltaic and other applications.

We present investigations of organic photovoltaic devices consisting of bulk heterojunction layers made from several material combinations. All of the investigated systems reveal close similarities to the behavior of classical pn-junction devices. The consequences of the pn-junction-like behavior on the device parameters and performance are presented. Furthermore, device characteristics and parameters of the pristine materials are correlated, resulting in a model that permits an identification of high potential materials, a performance prediction, and a device optimization. The resulting model is able to predict an open circuit voltage and a fill factor and their evolution with the light intensity or thickness of the active layer. It simplifies the identification of the internal morphology and therefore the choice of appropriate solvents. Necessary parameters concerning the choice of electrode materials are also provided.

No abstract