# --- Title -----

Data underlying the publication:

Kang, K. et al. Investigation of the Thermoelectric Response in Conducting Polymers Doped by Solid-State Diffusion. Materials Today Physics (2019). doi:10.1016/j.mtphys.2019.02.004


# --- Authors ---

Keehoon Kang, Sam Schott, Deepak Venkateshvaran, Katharina Broch, Guillaume Schweicher, David Harkin, Cameron Jellett, Christian B. Nielsen, Iain Mcculloch, Henning Sirringhaus


# --- Contents --

1) Fig1_conductivity_UVVis.opj

Data underlying Figure 1 of the above paper: UV-Vis absorption spectra of successively dedoped PBTTT/F4-TCNQ, P3HT/F4-TCNQ and CDT-BTZ/F4-TCNQ films. Dedoping has been carried out by annealing the films and the respective annealing temperatures are given as column comments.

Additionally, conductivity data of successively dedoped PBTTT/F4-TCNQ is provided.

2) Fig2_GIWAXS.opj

Data underlying Figure 2 of the above paper: X-ray diffraction (XRD) and grazing incidence X-ray diffraction (GID) patterns in PBTTT/F4-TCNQ in the de-doping process. Extracted lamellar and pi-pi stacking distances as well as crystallite sizes (stacking coherence lengths) are reported as well.

3) Fig3_Fig5_Seebeck_power_factor.opj

Data underlying Figures 3 and 5 of the above paper: Thermoelectric measurements of conducting polymers doped by solid-state doping. Seebeck coefficients (S)􏰇, conductivities (sigma), and thermoelectric power factors􏰉 (S^2*sigma) are given for for PBTTT, CDT-BTZ and P3HT doped with F4-TCNQ with a solid-state doping method. The above values were measured at each de-doping step on the same device.

The variation of 􏰇S vs sigma􏰉 is best described by a energy-dependent mobility model by Kang et al. (Nature Materials, 16(2):252–257, February 2017). Fits to this model are also given in the attached file. Resulting parameters for PBTTT are 􏰘s = 3 and 􏰉􏰖􏰯􏰉􏰖􏰯sigma_(E_0) = 3×10􏰌􏰊^-2 S/cm with an error bound of 􏰉􏰖􏰯sigma_(E_0) = 2×10􏰌􏰊^(-2) S/cm and sigma_(E_0)􏰉􏰖􏰯 = 4×10􏰌􏰊^-2 S/cm. The s􏰘 = 0 fit is given for comparison. The s􏰘 = 3 model also produces good fits for the two other polymers with different values of sigma_(E_0)􏰉􏰖􏰯 as 1.5×10^-2􏰌􏰊 S/cm and 1.0×10^-3􏰌􏰐 S/cm for CDT-BTZ and P3HT, respectively. The errors represent the measurement error due to the device variation which was only significant for CDT-BTZ.

4) Fig3_Fig5_Seebeck_power_factor.opj


Data underlying Figure 4 of the above paper: Density of states calculation and mobility variation with doping.

This file gives the density of unpaired spins-1/2 in PBTTT/F4-TCNQ for different electrochemical potentials eta, as determined from electron spin resonance (ESR) measurements which presumably corresponds to the charge carrier density N.

Two different density of states (DOS) models were used to fit the dependence of N on eta􏰳: The first model assumes a constant DOS (g ~ constant) above a transport edge E_t and the second model assumes a square-root-dependence of the DOS (g ~ E^1/2) above E_t. Both models assume an exponential tail of localized trap states for energies E < E_t with a continuous transition of the DOS at E = E_t.

This file shows both DOS models, the corresponding N vs. eta fitting curves, and the predicted charge density dependence of the mobility from either model.