Exploring the helical nanofilament liquid crystal phase for organic photovoltaics
Liquid crystals (LCs) revolutionized information display. Currently, considerable effort in the field is spent targeting non-display applications in various optical, photonic, and electro-optic spheres, including organic photovoltaics (OPVs).1 Nonetheless, the efficiency of current OPVs remains at about 10%. There are countless approaches to increasing OPV efficiency, many of which focus on bulk heterojunction (BHJ) solar cells. Efficiency increases have been held back by a lack of control of morphology and nanostructure.
Our work focuses on the helical nanofilament (HNF) liquid crystal phase, also known as the B4 banana phase, one of the most complex and interesting of the new LCs. We consider the HNF phase to provide a new approach that could increase OPV efficiency. Key to this is our discovery that the HNF phase forms nanoconfined composites with many materials and also allows unprecedented nanostructural and morphological control.
We published the HNF phase structure in 2009.2 The phase is dramatically chiral (i.e., left- or right-handed), exhibiting an extremely large rotatory power (circular birefringence) for visible light, which is easily observed in films about 10μm thick, even though the molecules are achiral. The phase, composed of bent-core LC molecules (mesogens), represents a unique and beautiful hierarchical self-assembly (see Figure 1). HNF molecules self-assemble into layers, ∼50Å thick, to form a kind of lamellar (smectic) phase. Due to their natural curvature, the layers grow no wider than about 30nm, while the number of layers in stacks self-limits at five to eight. This gives HNFs that are nearly cylindrical (∼30nm in diameter) with no
inherent constraint on their length. These HNFs then self-assemble to form a kind of liquid crystalline superstructure where the twist of the HNFs is coherent over large length scales. In addition, the chirality of the phase is coherent, in the sense that very large unichiral optically active domains (in the order of tens of microns) form spontaneously. Further adding to the novelty of the system, our solid-state NMR data demonstrates unusually slow conformational dynamics, suggesting that individual HNF layers are effectively single crystal-like, although there is no correlation of molecular positions across layers (i.e., the HNFs are not true nanocrystals).3
Early experiments on the HNF phase suggested a layered structure. However, miscibility tests with known smectics or other liquid crystal phases, a classic way of identifying phase structure, were confusing. It was as if on cooling an isotropic melted mixture, the HNF phase was simply ignoring the second component, which seemingly disappeared. These studies ultimately led us to unveil an additional remarkable characteristic of this phase: its ability to entrain up to 50% by weight of a large variety of organic molecules, including polymers such as P3HT—poly(3-hexylthiophene-2,5-diyl)—and fullerenes such as phenyl-C60-butyric acid methyl esters (PCBMs), frequently used together in plastic solar cells.4
In the case of PCBMs, our x-ray scattering data give no signature of fullerene crystallization, suggesting nanoconfinement of amorphous fullerenes (see Figure 2). This remarkable property is due to the system's geometry, which precludes tight packing of the HNFs, making it inherently nanoporous.
The HNF phase represents an organic, ordered nanoporous nanoparticle system capable of self-assembly into aligned nanostructured composites. This combination of properties has led us to investigate the phase as the basis of BHJ solar cells. In current OPV solar cells, phases that absorb light and those that accept excited electrons are present as amorphous domains. The basic idea behind the HNF approach is that the nanostructured heterojunction, with proper control of morphology, should be much better at generating charge (by splitting excitons) and at conducting the charge than standard BHJs.
Our initial research has focused on determining whether HNF-acceptor heterojunctions are capable of photocharge generation (exciton splitting). Recent results obtained in collaboration with scientists at the National Renewable Energy Laboratories in Colorado, using flash photolysis time-resolved microwave conductivity, are consistent with efficient exciton splitting from photoexcited fullerenes, both in planar heterojunctions with C60 and drop-cast NOBOW (1,3-phenylene bis[4-(4-nonyloxyphenyliminomethyl)benzoate]) HNF/PCBM BHJs.5 While the precise nature of this charge-generation process is still under investigation, data analysis suggests very efficient exciton splitting.
In summary, we have shown that it is possible to obtain nanostructured heterojunctions with HNF composites. The lack of fullerene crystallization in the composite is evidence of the nanosegregation of the HNF and fullerene domains. Although other donor–acceptor systems are also efficient at splitting excitons, they cannot be processed to give the same degree of structural control that we can get with the HNF phase composites. Due in part to exciting potential applications suggested by these results, we are collaborating with Dong Ki Yoon's group at Korea Advanced Institute of Science and Technology to develop approaches to gain additional control of the HNF phase's supramolecular structure and properties. For example, we are researching uniform alignment of HNFs over large areas (similar to the critical alignment of LCs in televisions). We are also exploring the HNF structure space aimed at providing HNF mesogens capable of absorbing visible light, as well as working on HNF-organic composites with small-molecule acceptors other than fullerenes. Meanwhile, we are investigating whether charge carrier mobility is larger along the long axis of HNF ‘rods,’ than from rod to rod, which would improve the efficiency of an OPV device.
We would like to acknowledge the National Science Foundation funded University of Colorado Boulder Materials Research Science and Engineering Center for supporting the project (NSF project DMR-1420736).
David Walba is professor of chemistry. He completed his PhD at Caltech with Robert Ireland in 1975, then went on to postdoctoral work with Donald Cram at the University of California, Los Angeles, before beginning his academic career at UCB.
