Efficient Electronic Tunneling Governs Transport in Conducting Polymer-Insulator Blends

Abstract

Electronic transport models for conducting polymers (CPs) and blends focus on the arrangement of conjugated chains, while the contributions of the nominally insulating components to transport are largely ignored. In this work, an archetypal CP blend is used to demonstrate that the chemical structure of the non-conductive component has a substantial effect on charge carrier mobility. Upon diluting a CP with excess insulator, blends with as high as 97.4 wt % insulator can display carrier mobilities comparable to some pure CPs such as polyaniline and low regioregularity P3HT. In this work, we develop a single, multiscale transport model based on the microstructure of the CP blends, which describes the transport properties for all dilutions tested. The results show that the high carrier mobility of primarily insulator blends results from the inclusion of aromatic rings, which facilitate long-range tunneling (up to ca. 3 nm) between isolated CP chains. This tunneling mechanism calls into question the current paradigm used to design CPs, where the solubilizing or ionically conducting component is considered electronically inert. Indeed, optimizing the participation of the nominally insulating component in electronic transport may lead to enhanced electronic mobility and overall better performance in CPs.

Figure 1

Conductivity scaling of increasingly diluted PEDOT:PSS blends. (a) Chemical structures of poly(ethylene dioxythiophene) (PEDOT) and poly(styrene sulfonate) (PSS). (b) Schematic showing electronic (orange) and ionic (dark blue) transport through a representative PEDOT:PSS microstructure. (c) Conductivity of PEDOT:PSS blends with increasing PSS concentration measured at room temperature with a 4-point probe technique. (d) Schematic diagram of an organic electrochemical transistor (OECT) test structure. (e) Transfer characteristics of OECTs made using increasingly diluted PEDOT:PSS blends.

Figure 2

Characterization of charge transport in diluted CP blends. (a) Volumetric capacitance (orange) and mobility (blue) scaling for increasingly diluted PEDOT:PSS blends. (b) Temperature-dependent van der Pauw conductivity measurements for diluted PEDOT:PSS blends. (c) Activation energy E_A (blue) and mobility pre-factor μ₀ (orange) fits for the temperature dependent conductivity measurements based on a thermally activated hopping transport model (see Figure S5). (d) Comparison of the measured mobility (dashed line) with the estimated room-temperature mobility using the activated hopping model varying only E_A while holding μ₀ constant (blue) and including both E_A and μ₀ (orange).

Figure 3

Structural characterization of diluted PEDOT:PSS blends. c-AFM images for (a) 1:2.5 (pristine), (b) 1:6, and (c) 1:16.5 PEDOT:PSS blends. (d) GISAXS and (e) peak intensity for PEDOT:PSS blends. (f) Schematic demonstrating the structural effect of dilution with excess PSS; the spacing between PEDOT-rich grains increases while the PEDOT concentration in the PSS-rich matrix decreases.

Figure 4

Computational transport model in PEDOT:PSS films. (a) Schematic of diluted PEDOT:PSS films, where PEDOT chains in PSS regions facilitate charge transport (arrows) between PEDOT-rich regions (dark blue ellipsoids). (b) Simulated arrangement of PEDOT chains in the PSS-rich matrix, which are tessellated to generate the (c) molecular and meso scale charge transport (CT) model. At the molecular scale, holes depart from a PEDOT-rich region (blue sphere, bottom left) and tunnel between PEDOT chains (yellow-green molecules) through the PSS-rich matrix. k_ET is defined in eq S1. At the mesoscale, holes move between PEDOT-rich regions at an effective rate, k_ss(L), calculated with eqs S4–S6 where the grain-to-grain distance L is between 25 and 60 nm (see Section S2). (d) At the device scale, holes move through the film by traveling between randomly dispersed PEDOT-rich grains. (e) Mobility curves derived from MD simulated structures with varying tunneling attenuation coefficients, β, compared to the experimentally measured mobilities of diluted PEDOT:PSS (dashed line). (f) Mobility curves from experiment (dashed line) and transport models using various geometric estimates of PEDOT chains with β = 0.3 Å^–1. Colored traces in (e,f) are shifted vertically for clarity.

Figure 5

Conductivity scaling for p(g2T-TT):PEG mixed conducting blends. (a) Chemical structures for the semiconducting polymer p(g2T-TT) and blending polymers PEG and p(4MeOS). (b) OECT device test structure using an ion gel electrolyte. (c) OECT transfer characteristics for the diluted p(g2T-TT) blends. (d) Comparison of the mobility scaling PEDOT:PSS (blue), p(g2T-TT):PEG (purple), and p(g2T-TT):p(4MeOS) (green) diluted blends. For reference, the slopes corresponding to tunneling coefficients of β = 0.3, 0.6, and 1.0 Å^–1 are plotted (gray dashed lines). We note that the weight of the pegylated side chains of p(g2T-TT) was considered as a part of the non-conductive weight fraction.