Characterization

The Raman spectra (Renishaw, λ = 514 nm laser, 10% power) were collected using the extended recording mode from 0 to 3500 cm⁻¹ and the integration time of 10 s to increase the signal to noise ratio. For every sample, the measurement was repeated in at least three sample locations using 3 accumulations. The Raman spectra were used to determine the influence of the doping agent on the electronic characteristics of CNT films and to verify whether their addition causes chemical modification of the material.

The thermal stability of the CNTs doped with halide compounds was characterized by a thermogravimetric analyzer (Mettler Toledo TGA/DSC 1 STAR) in the temperature range from 25 to 1000 °C in a flow of air (30 mL/min) and with the heating rate of 10 °C/min.

The changes in electrical conductivity of the samples after doping with halide compounds were measured using the 4-point method. The tests were conducted using the Keithley 2450 SourceMeter, and the electrical resistivity was measured under 100 mA electric current. The absolute value of electrical conductivity was calculated based on the dimensions of samples measured with a 4″ digital caliper (Hi-Tech diamond) for width and length, as well as a Multi-Anvil Micrometer (Electronic Universal IP54, Linear) for thickness. The electrical conductivity of all the samples was initially measured at room temperature. For the most promising dopants, measurements were also taken at higher temperatures (40 °C, 70 °C, and 100 °C) by heating the samples on a hot plate while recording the sample temperatures in a non-contact mode by Infrared Thermometer Scan Temp 380 (TFA).

The Seebeck coefficient was determined with the use of a custom-made setup (SeebCam 2018, LBR, Lublin, Poland). It was measured in the range of temperatures from 30 to 100 °C for the samples of the size of 2 × 50 mm. The sample was placed on a board maintained in a sealed chamber to eliminate the convection effect (Fig. ^S8). Both ends of the sample had appropriate contact with temperature sensors and resistive heaters. To ensure a suitable electrical and thermal interface between the sample and the equipment, silver paint SCP Electrolube was used. The electric potential difference between the sample ends was measured (Keithley 2128A) at the temperature gradient of 5 °C in the temperature range between 30 °C and 100 °C. The temperatures mentioned in the text refer to the temperatures of the hot end. Each sample was subject to at least 5 measurements. The results were averaged, and the statistical error was calculated.

The Power Factor value was calculated from the measured values of electrical conductivity and Seebeck coefficient.

Scanning Electron Microscope (SEM, FEI Quanta 250 FEG) running at the acceleration voltage of 15 kV was used to study the microstructure of the selected materials. The samples were not sputtered with metal because of their high electrical conductivity.

X-ray photoelectron spectroscopy (XPS) measurements were performed in an ultra-high vacuum system (base pressure ca. 8 × 10^–9 Pa) with a PREVAC EA-15 hemispherical electron energy analyzer equipped with the 2D-MCP detector. In order to provide the best possible energy resolution, the Mg Kα energy line (1253.60 eV; PREVAC XR-40B dual-anode source) was used as well as the curved analyzer slit was applied. The pass energy was set to 200 eV for the survey spectra, with a step of 0.9 eV and set to 100 eV for individual energy regions, with an energy step of 0.05 eV. The binding energy (BE) scale of the analyzer was calibrated to Ag 3d 5_/2 (368.2 eV)⁷¹. The energy region decomposition was performed with the use of CASA XPS software. Each peak was represented by a sum of Gaussian (70%) and Lorentzian (30%) lines. For the background subtraction, the Shirley function was utilized. The full width at half maximum (FWHM) of the components at the same energy region were allowed to vary within a narrow range.

Optical absorption spectroscopy was used to estimate the ratio of metallic to semiconducting CNTs. CNTs (1 mg/mL) in 1%wt aqueous sodium cholate solution were homogenized by tip-sonication (1 h, Hielscher UP200St), and then ultracentrifuged (1 h, 15,314×g, Eppendorf Centrifuge 5804 R) to remove the non-individualized fraction. The supernatant was investigated in the wavelength range of 400–1100 nm using the Hitachi U-2910 spectrophotometer.