Influence of Excess Conjugated Wrapping Polymer in Semiconducting Single-Walled Carbon Nanotube Dispersions

Single-walled carbon nanotubes (SWNTs) are promising nanomaterials for incorporation into organic electronic devices (OEDs), with the potential to fabricate flexible devices while exploiting inexpensive solution-processing techniques. As-synthesized SWNTs are insoluble and comprised of a mixture of metallic and semiconducting SWNTs (sc-SWNTs), necessitating dispersal and purification before integration into OEDs. Refinement of conjugated polymer extraction techniques has allowed for the isolation of sc-SWNTs from metallic in a reproducible and scalable manner. The availability of highly-pure sc-SWNT materials has facilitated the production of thin-film transistors (TFTs) with very high charge carrier mobilities, outperforming organic small molecule and polymer semiconductors. However, the realization of commercial OED applications of polymer-sorted sc-SWNTs have not yet been achieved, partially due to the prohibitive time and materials costs associated with purifying sc-SWNTs. Current protocols for dispersing sc-SWNTs with conjugated polymers involve three broad steps: (1) dispersion of bulk SWNT material, (2) removal of non-dispersed carbonaceous materials, and (3) removal of excess polymer through filtration or centrifugation. The final step of removal of excess polymer is time-consuming and wasteful, but viewed as necessary for preparing high-performing TFTs, as the conjugated polymer has much lower performance compared to SWNTs. In this study we performed the first systematic investigation of the effect of excess polymer on SWNT TFT performance. Three SWNT concentrations were investigated, with varying ratios of excess polymer added to each. TFT device performance was monitored using several metrics, including: mobility, threshold voltage, on/off ratios and hysteresis. Characterization of large numbers of replicate TFT devices determined that below a threshold amount of excess polymer the presence of excess polymer did not have a negative impact on device performance. Detailed analysis of the sc-SWNT films through Raman spectroscopy and atomic force microscopy (AFM) confirmed that a simple rinsing step was sufficient to remove all the unbound conjugated polymer from the substrate surface without affecting the sc-SWNT network. The volume of solvent required for the rinsing step was substantially lower than that required for filtration or centrifugation steps. Furthermore, at higher SWNT concentrations the excess polymer prevented nanotube bundling, resulting in moderate improvements in both mobility and on/off ratios. Our results were reproducible for two different conjugated polymer sc-SWNT systems, demonstrating the versatility of this procedure.