Molecular self-ordering in organic solar cell materials
Solar energy is expected to become one of the world's main energy sources, one that does not significantly raise the concentration of greenhouse gases in the atmosphere. Thin light-absorbing, semiconducting organic films may help lower the economic barrier to generating solar electricity because they are flexible, lightweight, and can be fabricated over large areas using low-cost, solution-based processing.
In the most common implementation of an organic solar cell, light is absorbed in a bulk heterojunction (BHJ) layer. The heterojunction between the electron donor (typically, a conjugated, light-absorbing polymer) and acceptor (a fullerene) is formed from a bicontinuous network of the two phases, with phase separation on the nanometer length-scale.1 Such cells have recently achieved certified power conversion efficiency above 6%.2,3 Over the last decade, scientists increased efficiency through materials design and morphology control. Yet, until recently, researchers did not understand why some polymer:fullerene systems reach optimal morphology and efficiency at equal parts of each material, while other systems optimize near 80% fullerene content (by weight). Little was known about the material properties that determined the optimal blend ratio.
We recently investigated a polymer that reaches optimal solar cell efficiency at 80% fullerene content. Synchrotron x-ray diffraction measurements revealed a highly ordered bimolecular crystal, in which the two distinct chemical species order on the same lattice.4 These ordered structures form when the fullerene molecules are intercalated between the side chains of the semi-crystalline polymer: see Figure 1(c). We observed this intercalation in several polymer:fullerene blends deposited from solution. It can explain the origin of the optimal 1:4 weight ratio in these blends. An excess of the fullerene yields a two-phase film composed of the bimolecular crystal and a pure fullerene phase with approximately equal volumes. For a ratio ∼1:1, only the pure phase of the bimolecular crystal forms.
The specular high-resolution x-ray scans in Figure 2(a) show the phase behavior. The peaks of the pure polymer phase, (h00)p, disappear when we add enough fullerene derivative to the blend. At a 3:1 ratio, the intercalated (h00)i and the pure polymer phase coexist, whereas [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) contents above 1:1 only form the intercalated phase.
In a few polymers, we showed that the lattice spacing increases by about 9Â upon blending with a soluble fullerene derivative, e.g., phenyl-C71-butyric acid methyl ester (PC71BM). The one with the highest crystallinity is poly(2,5-bis(3-tetradecyllthiophen-2-yl)thieno[3,2-b]thiophene) (PBTTT), a conjugated polymer with a very high hole mobility: see Figure 1(a).5 We associate this lattice spacing increase—approximately the fullerene diameter—with the intercalation of a fullerene between the polymer side chains.
The solar cell efficiency in PBTTT:PCBM blends improves substantially from 0.16% to 2.35% for 1:1 to the 1:4 ratio, respectively.6 This improvement may be an effect of the dependence of the charge transport on the blend ratio: see Figure 2(b). While the hole mobility (measured in field-effect transistors of the blend) decreased by less than an order of magnitude from the pure polymer phase to 80% PC71BM, the electron mobility decreases by more than an order of magnitude from 1:4 to 1:1, and then becomes immeasurable for lower PCBM ratios. The low mobility at the 1:1 ratio accounts for the impeded electron transport from the absorption layer to the respective electrode, due to the absence of an electron-conducting phase (pure PCBM).
Another loss mechanism in solar cells, photoluminescence, is the emissive recombination of bound electron-hole pairs that are excited by the absorbed light. In our intercalated structure, photoluminescence is completely quenched because a PCBM molecule is available in the vicinity of every exciton generated in the polymer, and may act as an ‘interface’ on the molecular scale: see Figure 2(c).
It is not yet clear if intercalation is desirable for solar cell devices. Nonetheless, our findings suggest that we should develop separate recombination models for cells forming with and without intercalation. In order to increase efficiencies, the design of polymers for BHJ solar cells must consider intercalation. The discovery suggests a way to intentionally design bimolecular crystals and tune their properties to create novel organic materials for photovoltaics and other applications such as LEDs, lasers, and biosensors. In the future, our group will study the impact of intercalation on solar cell performance, recombination, and charge transport in this and other material combinations
Roman Gysel received a PhD from the Swiss Federal Institute of Technology in Lausanne, Switzerland. He is working on bulk heterojunction solar cells, with an emphasis on structural analysis of the absorbing layers. He was awarded a Swiss National Science Foundation fellowship.
Nichole Cates is a PhD candidate studying structure-property relationships in organic solar cells.
Michael McGehee is an associate professor and the director of the Center for Advanced Molecular Photovoltaics. He received his PhD from the University of California at Santa Barbara, where he did research on polymer lasers in Nobel Laureate Alan Heeger's laboratory. Among other awards, he has won the 2007 Materials Research Society Outstanding Young Investigator Award.
Alex Mayer received a PhD in applied physics from Cornell University. His PhD was followed by a postdoc at Stanford University focusing on polymer-based solar cells. He is currently a research scientist.
Michael Toney is a senior staff scientist and has partial responsibility for the x-ray scattering program. He leads a research team of five post-doctoral researchers who do fundamental research on the atomic structure and morphology of thin films, nanoparticles, and other advanced materials.
