Molecular dynamics study on heat conduction in poly(3,4-ethylenedioxythiophene)

Abstract

The poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) composite has been gaining attention as a potential organic thermoelectric material owing to its superior electrical property and low thermal conductivity. Several experiments have recently demonstrated that introduction of high-boiling-point solvents increases optimal charge carrier concentration and leads to high crystallinity, resulting in noticeable improvement of thermoelectric performance. Therefore, much attention has been paid to electrical properties. Understanding of intrinsic heat conduction in PEDOT is also important since the thermal conductivity of bulk polymers is sensitive to macroscopic alignment and transport properties of individual chains. Moreover, thermoelectric performance inversely scales with thermal conductivity. In this work, by molecular dynamics simulations, we have investigated heat conduction in PEDOT chains and evaluated the impact of adsorbed protonic acid on the thermal conductivity of PEDOT. Owing to the quasi-ballistic nature of heat conduction in PEDOT chains, we found that thermal conductivity has strong size dependence and varies from 1–10 W m−1 K−1. Furthermore, we found that the adsorption on PEDOT of toluene sulfonic acid (TSA); a monomer of PSS; suppresses contribution of the long-wavelength phonons to heat conduction. These results suggest that decreasing the length of PEDOT chains in a composite and increasing the surface density of adsorbed TSA molecules are effective methods of reducing the thermal conductivity of the composite, leading to the enhancement of thermoelectric performance.

Thermal Transport in Conductive Polymer–Based Materials

“…The analysis of the morphology of the polymer films shows that the anisotropy of thermal conductivity originates from the chain orientations. Other molecular dynamics study performed by Maeno et al [131] has also found a large anisotropy of thermal conductivity in PEDOT. Their simulation results have shown that a single PEDOT chain with a length of 2-35 nm has high thermal conductivity ranging 4-27 W m −1 K −1 along the chain direction, which is higher than the thermal conductivity of polymers with randomly twisted and entangled chains.…”

Hybrid Thermal Transport Characteristics of Doped Organic Semiconductor Poly(3,4-ethylenedioxythiophene):Tosylate

“…21,50 The along-chain thermal conductivity (κ y ) is lower than that in previous MD results. 38,39 This discrepancy is attributed to different simulation method and force field. The equilibrium MD method is widely used to simulate the thermal transport of bulk materials.…”

“…37 Recently, through molecular dynamics (MD) simulations, Shi et al reported the along-chain thermal conductivity values of PEDOT:Tos on the order of 10 Wm −1 K −1 . 38 Maeno et al 39 and Crnjar et al 40 investigated the alongchain thermal transport in PEDOT chains by MD simulations. Yet, the effect of dopants on thermal transport has not attained enough attention.…”

“…34−37 Therefore, as a base study, PEDOT:Tos crystal is preferred, due to the well-validated molecular structures and the convenience in comparison with previous simulation studies. 38,39 In addition, it is noteworthy that single-crystal PEDOT:Tos nanowires have been synthesized, 41 suggesting the promising potential of PEDOT:Tos. Experimental observations reported the doping level of PEDOT:Tos to be about one Tos molecule per four EDOT monomers.…”

Progress and Perspective: Soft Thermoelectric Materials for Wearable and Internet‐of‐Things Applications

“…According to this prediction, κ can be reduced either by reduction in the product Cλ(λ,T)ν(λ) through phonon confinement in nanomaterials and superlattices with extremely small dimensions, or by reducing L(λ ,T) through enhancement of phonon boundary and interface scattering. In order to optimize the thermal conductivity in a given material, theoretical approaches have been developed including molecular dynamics simulations, [87] firstprinciples calculations, [88] and machine learning, data-driven methods. [89,90] Experimental measurement of phonon transport is more challenging.…”