Thermogalvanic cells: A side-by-side comparison of measurement methods

Thermogalvanic cells (or thermocells) are being increasingly investigated due to their ability to directly convert modest temperature gradients into electricity, using redox chemistry. However, this has resulted in a diverse range of cells, electrolytes, temperature gradients, etc. being reported. Recently, so-called temperature-difference ‘normalised’ power outputs have been reported; the power output is divided by the square of the temperature difference to yield temperature-insensitive power output values. The validity of this procedure is quantitatively assessed here, and found to be far from accurate. Additionally, numerous different experimental procedures for measuring the current and power output from thermocells can be found in the literature, covering different approaches, vastly different time scales, and employing a diverse range of measuring instruments. A range of experimental methodologies covering a sequence of constant resistances (or resistive loads), a sequence of constant currents, a sequence of constant potentials (or voltage), chronoamperometry and linear sweep voltammetry (LSV) were evaluated in this study, using source measure units, potentiostats and voltammeters/multimeters. While most methodologies were consistent, the measurement time and number of electrodes were found to be highly influential, with LSV and 3-electrode assemblies especially overestimating the current and power output from thermocells.