A significant challenge in the rational design of organic thermoelectric materials is to realize simultaneously high electrical conductivity and high induced-voltage in response to a thermal gradient, which is represented by the Seebeck coefficient. Conventional wisdom posits that the polymer alone dictates thermoelectric efficiency. Herein, we show that doping — in particular, clustering of dopants within conjugated polymer films — has a profound and predictable influence on their thermoelectric properties. We correlate Seebeck coefficient and electrical conductivity of iodine-doped poly(3-hexylthiophene) and poly[2,5-bis(2-octyldodecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione-3,6-diyl)-alt-(2,2′;5′,2′′;5′′,2′′′-quaterthiophen-5,5′′′-diyl)] films with Kelvin probe force microscopy to highlight the role of the spatial distribution of dopants in determining overall charge transport. We fit the experimental data to a phonon-assisted hopping model and found that the distribution of dopants alters the distribution of the density of states and the Kang–Snyder transport parameter. These results highlight the importance of controlling dopant distribution within conjugated polymer films for thermoelectric and other electronic applications.

Organic materials have attracted recent interest as thermoelectric (TE) converters due to their low cost and ease of fabrication. We examine the effects of disorder on the TE properties of semiconducting polymers based on the Gaussian disorder model (GDM) for site energies while employing Pauli’s master equation approach to model hopping between localized sites. Our model is in good agreement with experimental results and a useful tool to study hopping transport. We show that stronger overlap between sites can improve the electrical conductivity without adversely affecting the Seebeck coefficient. We find that positional disorder aids the formation of new conduction paths with an increased probability of carriers in high energy sites, leading to an increase in electrical conductivity while leaving the Seebeck unchanged. On the other hand, energetic disorder leads to increased energy gaps between sites, hindering transport. This adversely affects conductivity while only slightly increasing Seebeck and results in lower TE power factors. Furthermore, positional correlation primarily affects conductivity, while correlation in site energies has no effect on TE properties of polymers. Our results also show that the Lorenz number increases with Seebeck coefficient, largely deviating from the Sommerfeld value, in agreement with experiments and in contrast to band conductors. We conclude that reducing energetic disorder and positional correlation, while increasing positional disorder can lead to higher TE power factors.

We present a calculation of the lattice thermal conductivity of Si-Ge nanowires (NWs), based on solving the Boltzmann transport equation by the Monte Carlo method of sampling the phonon mean free paths. We augment the previous work with the full phonon dispersion and a partially diffuse momentum-dependent specularity model for boundary roughness scattering. We find that phonon flights are comprised of a mix of long free-flights over several µm interrupted by bursts of short flights, resulting in a heavy tailed distribution of flight lengths, typically encountered in Lévy walk dynamics. Consequently, phonon transport in Si-Ge NWs is neither entirely ballistic nor diffusive; instead, it falls into an intermediate regime called superdiffusion where thermal conductivity scales with the length of the NW as κ ∝ L α with the exponent of length dependence α ≈ 0.33 over a broad range of wire lengths 10 nm< L <10 µm regardless of diameter and roughness. We conclude that thermal conductivity in Si-Ge alloy NWs is length-dependent up to 10 µm and therefore can be tuned for thermoelectric applications. arXiv:1610.08477v1 [cond-mat.mes-hall]