Eliminating resistance measurement error due to thermoelectric effects in micro four-point probe measurements

Abstract

The micro four-point probe (M4PP) technique has become a well-established method for characterizing the electrical properties of materials. However, extra attention must be paid when measuring the resistivity of thermoelectric materials due to the possibility of an additional Seebeck voltage. This issue vanishes when measuring at a sufficiently high frequency, but the threshold frequency is substantial due to the small separation between the pins of the probes. Typical M4PP measurements are far from reaching this frequency, and their accuracy on thermoelectric materials is severely compromised. In this work, we explain the experimental conditions needed to measure reliably the electrical conductivity of thermoelectric materials and present a new method for measuring this property that reduces the frequency requirements by two orders of magnitude. The method is proven using two skutterudites and bismuth telluride material. It is also found that the resistance overestimation in the bismuth telluride sample is larger than that in the skutterudites due to its superior thermoelectric properties; the overestimate reaching 35%. The advances reported here enable the M4PP technique to be used for the measurement of the electrical conductivity of thermoelectric materials.