Making Organic Semiconductors Plastic

Plastic. The very word implies deformability, the ability to bend and flex without damage in response to stress. In applications from biomedical sensors to solar cells, the potential advantages of organic semiconductors depend almost entirely on their deformability—are they flexible enough for inexpensive roll-to-roll processing? Able to tolerate flexion in use? Able to do without the bulky and expensive packaging that protects inorganic semiconductor devices?

Yet polymer materials as a group span an enormous range of properties. Rigid, brittle polycarbonate offers a Young’s modulus around 1 GPa, while the modulus of silicone rubber is only about 1 MPa. Polyethylenes, ubiquitous in disposable bottles and consumer packaging, are probably the most familiar plastics for non-specialists, with Young’s modulus ranging from 0.11 to 2.7 GPa.

In organic semiconductors, though, the need for efficient charge transport conflicts with the need for mechanical compliance. While the specific tradeoffs vary among material systems, carriers generally move more efficiently along a rigid molecular backbone, or between closely packed small molecules. The rigidity of the structure usually means these materials are not especially flexible. However, as Darren Lipomi of UC San Diego explained at April’s Materials Research Society spring meeting, the properties of individual organic semiconductor molecules give only limited insight into the properties of thin films or bulk samples made from those molecules. Process conditions define such critical parameters as molecular weight and weight distribution, the crystallinity of the sample, and the interdigitation of side chains attached to the molecular backbone. A bulk heterojunction, in which an electron acceptor material is uniformly dispersed throughout an electron donor material, will be less flexible than an material with the same composition which is allowed to self-aggregate. In most cases, simulations are inadequate, and the mechanical properties can only be determined experimentally. This is a significant obstacle for many candidate materials, which may not be available in sufficient quantity.

Even when data does exist, interpreting it can be a challenge in its own right. For example, NIST’s Chad Snyder pointed out that x-ray diffraction alone cannot distinguish between low, but uniform, crystallinity throughout a sample, and a random dispersion of highly ordered crystallites. Crystallization kinetics, determined for instance by calorimetry, will define the material that is actually obtained. Variability between batches of the material can change the phase diagram, leading to different results from the same process recipe. Phase behavior ultimately depends on measurable properties like molecular weight and melting temperature, but it’s essential that manufacturers and researchers actually do the measurements to characterize the material they’re working with.

Four decades after the first identification of organic semiconductors, many fundamental mechanical properties have yet to be investigated in any kind of systematic way. Though these materials have the potential to support a vast array of light, flexible applications, that potential will not be realized without a better understanding of the relationships between processing, mechanical structure, and electrical properties.