Review pubs.acs.org/CR
Bulk Heterojunction Solar Cells: Morphology and Performance Relationships Ye Huang,†,‡ Edward J. Kramer,*,‡,§ Alan J. Heeger,*,∥ and Guillermo C. Bazan*,† †
Center for Polymers and Organic Solids, Department of Chemistry & Biochemistry, ‡Department of Materials, §Department of Chemical Engineering, and ∥Department of Physics, University of California, Santa Barbara, California 93106, United States 1. INTRODUCTION With the hope of solving the present energy crisis and associated environmental issues, much research has been focused on solar cells that can harvest energy directly from sunlight to enable sustainable and green energy technology. Organic photovoltaic thin films offer considerable promise for meeting some of these needs.1 Their potential for low-cost and fast roll-to-roll production as well as their light weight and fabrication on flexible substrates could give them major advantages over traditional inorganic solar cells. The discovery of ultrafast charge transfer opened the field of bulk heterojunction (BHJ) solar cells.2 This ultrafast charge transfer implied a high efficiency for charge separation, the first step in the conversion of sunlight to electricity. Once separated, the electron (electron polaron) and hole (hole polaron) must be transported through the acceptor and donor phases to the cathode and anode. These dual requirements suggest optimization through the creation of bicontinuous interpenetrating network morphology with optimal domain size. Therefore, an interpenetrating morphology with a large interfacial area is the most widely accepted target and is referred to as the BHJ structure.3−5 The physical basis for the ultrafast charge transfer step has been the topic of substantial studies. Much of the original explanation for charge carrier photogeneration begins with an intramolecular excitation leading to a localized and bound electron−hole pair (Frenkel exiciton).6 The exciton was thought to diffuse until it encounters a donor−acceptor interface where charge transfer and charge separation are energetically favorable.6 Given the known diffusion constants in molecular solids, however, it is not possible to envision diffusion over the distances (10−20 nm) observed in the phaseseparated (nanostructured) BHJ material in times of less than 100 fs: Exciton difffusion over such distances would require hundreds of picoseconds.7 In contrast to this exciton diffusion picture, ultrafast generation of mobile carriers (t < 100 fs, in some cases
