Polymer-based organic photovoltaics (OPV) are a promising option for light-weight, cost-effective solar cells, especially if they can be processed in solution.1 In photovoltaic devices, electron transfer occurs predominantly at the interface between two materials that differ in their electron affinities. However, exciton diffusion lengths (the distance that an excited state can travel before decaying back to the ground state) of the materials used in the photoactive layer of these devices is limited to approximately 5-10nm.2, 3 Electron donor and acceptor molecules must be in close proximity to ensure excitation results in a photocurrent before recombination of the electron-hole pairs, and so this imposes considerable restrictions on device morphology.
Research has focused on two broad architectures for creating efficient devices: a donor-acceptor bilayer, typically built through vacuum deposition of the components, and a structure in which the two materials are highly intercalated, referred to as a bulk heterojunction (BHJ). The mixed nature of the photoactive layer in BHJs greatly increases the interface between the high-affinity and low-affinity regions solving the problem of short diffusion lengths in these materials. They also offer the advantage of being able to be processed in solution in a single step. In fact, the key enabling characteristic of organic semiconductor mixtures targeted for use in solar cells is this ability to self-assemble into nanostructured morphologies. In this manner, photogenerated excitons can find donor-acceptor interfaces that promote dissociation prior to exciton decay.
The morphology of the BHJ active layer is critical for device performance. A large interface between the two components must exist and the domains of donor and acceptor regions must be approximately 10nm. In addition, each domain must remain a continuous structure since electrons travel within the acceptor phase while holes travel through the donor phase. Currently, the most promising acceptor molecule is considered to be the fullerene phenyl-C61-butyric acid methyl ester (PCBM). One of the most commonly used donors is the sulfur-containing polymer poly(3-hexylthiophene), also known as P3HT. Mixtures of P3HT and PCBM spontaneously self-assemble into BHJ morphologies. But are these morphologies optimal for an efficient solar cell architecture?
We utilized energy-filtered transmission electron microscopy (EFTEM) to unequivocally confirm the presence of an intricate morphology well-suited for organic solar cells in P3HT/PCBM mixtures (see Figure 1).4 Analysis of the thermodynamics of the mixing reaction using Flory-Huggins theory reveals that P3HT and PCBM are partially miscible when amorphous and at optimum compositions the materials can form a single phase. P3HT crystals, however, can also precipitate out of solution and create a fibrous mesostructure that extends throughout the active layer: see Figure 1(S). Thus, the fiber-like motif of P3HT crystals makes this material ideal for OPVs. Moreover, the self-limiting width of the fibers yields a robust morphology. As a result, the characteristic length scales of the mesostructure vary little, even under a variety of processing conditions. Grazing-incidence small angle x-ray scattering experiments confirm that the morphology varies by less than 10nm.4
Figure 1. Bright field (BF), sulfur elemental (S), and carbon elemental (C) maps obtained from energy-filtered electron microscopy of poly(3-hexylthiophene)/phenyl-C61-butyric acid methyl ester (P3HT/PCBM) mixtures. The light regions in the elemental maps correspond to the presence of the element of interest. The cloudy fibrous image under BF is revealed to be a matrix of crystalline P3HT (which contains sulfur) running through an amorphous layer of mixed P3HT and PCBM. The scale bar is 200nm.
The semi-crystalline nature of P3HT is significant in other ways. Since P3HT and PCBM are largely miscible, un-crystallized (amorphous) P3HT mixes with PCBM. Domain composition measurements from EFTEM confirm the presence of molecular mixing in the PCBM-rich phase: see Figure 1(C). Thus, instead of an architecture of fibrous donor molecules intercalated throughout a base of acceptors, the morphology is that of donors intercalated throughout an amorphous region of combined donors and acceptors. The role of this mixed phase, ubiquitous in P3HT/PCBM devices and likely to be present in all semi-crystalline mixtures, is unclear. Furthermore, no studies on amorphous P3HT/PCBM mixtures have been performed to date, and so their transport properties are currently unknown. The structure works, but the intermixing of P3HT into the PCBM-rich phase will likely have deleterious consequences on electron transport and may enhance charge recombination.
Nonetheless, the fact that P3HT and PCBM are at least partially miscible is important for the morphological evolution. It prevents macrophase separation and allows the P3HT crystallization to fix structural length scales near the all important exciton diffusion length. Indeed, this robustness in the morphology must be a contributing reason for the widespread interest in this material system. Determining the optimal extent of miscibility desired for organic semiconductor mixtures employed in photovoltaics is an unresolved problem. To mix or not to mix, that is the question, at least for organic solar cells. We are continuing to direct our efforts at understanding the structural parameters which affect the performance of organic solar cells with mesostructured BHJ morphologies through a combination of electron and light microscopy, x-ray and light scattering, electron diffraction, and device testing.
Enrique Gomez, Derek Kozub, Kiarash Vakhshouri, Lisa Orme
Chemical Engineering Department
Pennsylvania State University
University Park, PA
Enrique Gomez is an assistant professor of chemical engineering. His work is focused on elucidating the role of structure on macroscopic properties of soft materials. Current efforts are directed at examining the structure formation process of inhomogeneous organic semiconductors to identify the critical parameters for improving performance of organic electronics.
Cheng Wang, Alexander Hexemer
The Advanced Light Source
Lawrence Berkeley National Laboratory
1. C. J. Brabec, S. Gowrisanker, J. J. M. Halls, D. Laird, S. Jia, S. P. Williams, Polymer-fullerene bulk-heterojunction solar cells, Adv. Mater. 22, pp. 3839–3856, 2010.
2. C. Goh, S. R. Scully, M. D. McGehee, Effects of molecular interface modification in hybrid organic-inorganic photovoltaic cells, J. Appl. Phys. 101, pp. 114503, 2007.
3. B. C. Thompson, J. M. J. Fréchet, Polymer-fullerene composite solar cells, Angew. Chem. Int'l Ed. 47, pp. 58–77, 2008.
4. D. R. Kozub, K. Vakhshouri, L. M. Orme, C. Wang, A. Hexemer, E. D. Gomez, Polymer crystallization of partially miscible polythiophene/fullerene mixtures controls morphology, Macromolecules 44, pp. 5722–5726, 2011.