(Invited) Evaluating Chemical Doping and Charge-Carrier Transport in Carbon Nanotube Dispersions

Abstract

Chemical charge-transfer, or redox, doping has long been recognized as a powerful approach to tune the charge carrier density in pi-conjugated semiconductors. Despite this, and the extensive effort that has been put into developing novel redox dopants, several challenges remain to understand (i) the doping mechanism/efficiency, (ii) the interactions between the injected charge carrier and the associated counterion, and (iii) the impact that these have on charge carrier transport. This is complicated by the fact that it is very difficult to accurately measure charge carrier densities in pi-conjugated semiconductors and to assess carrier-counterion interactions. Here we employ dispersions of enriched semiconducting single-walled carbon nanotubes (s-SWCNT) as a prototypical pi-conjugated semiconductor, since the rigid structure and strong optical transitions offer some advantages over typically more disordered conjugated polymers. We employ a suite of spectroscopic techniques to probe charge carrier doping and transport in isolated polymer-wrapped s-SWCNT in a low dielectric constant solution-phase environment. We compare solution-phase s-SWCNT samples in toluene that are p-type doped by either 2,3,5,6-tetrafluoro-7,7,8,8-tetracyano-quinodimethane (F4-TCNQ) or (2,3,4,5,6-pentafluoro)benzyloxy-substituted icosahedral dodecaborane cluster (DDB-F60). We use a combined approach of steady state optical spectroscopy with 19F NMR studies. Optical spectroscopy allows us to track the bleach of the electronic transitions due to excitons and the appearance of signature due to charge carrier-induced trions, which we can correlate to the concentration of dopant counterions extracted from NMR experiments. This combined methodology allows for quantitative determination of a molar extinction coefficient for free carriers for the SWCNT dispersion. We subsequently show that DDB-F60 is a more potent charge-carrier dopant than F4-TCNQ at low concentrations but that the ultimate injected hole charge-carrier density is limited by the large size of the DDB-F60 counteranion, due to saturation of the carbon nanotube surface area that is accessible to the dopant molecules. Using a non-contact spectroscopy technique in the microwave frequency region (ca. 9 GHz), we show that the holes injected by DDB-F60 are more delocalized and mobile, consistent with calculations that suggest a smaller coulombic binding energy between the hole on the carbon nanotube and the electron on the DDB-F60 counteranion. However, samples doped by F4-TCNQ reach a higher peak conductance than is achieved by DDB-F60 doping, where the conductance in the latter case is limited by the maximum achievable hole charge-carrier density and hole scattering processes that reduce the carrier mobility at high hole densities. Our observations point to the need to understand the subtle balance between the carrier-counterion interactions and the injected charge-carrier density to optimize charge-carrier transport in chemically doped s-SWCNT systems.