Resumen de Advanced characterization of thermoelectric materials and devices by impedance spectroscopy
It is estimated that around 60% of the global energy used is lost as waste heat. Thermoelectric (TE) devices can directly convert heat into electricity (energy generation mode) or use electricity to create a temperature difference (cooling/heating mode). For this reason, they can be a suitable energy harvesting technology and contribute to the current energy crisis. However, they are not widely spread currently due to their low efficiency. The development of new, more efficient materials is typically based on the optimization of the dimensionless figure of merit (zT=S^2σT/λ), which is determined by three material properties: the Seebeck coefficient (S), the electrical conductivity (σ) and the thermal conductivity (λ), and also the temperature (T). Hence, the determination of these properties as a function of temperature is a necessary step in the development of any new material.
Regrettably, the characterization of all these parameters is quite lengthy and tedious, typically requiring the use of at least two different apparatuses. In addition, there is not a standard equipment established and homemade apparatuses are frequently employed, which makes difficult inter-laboratory correlation. Moreover, a high uncertainty is typically associated with zT, since it adds the errors of the three properties that define it.
Impedance spectroscopy (IS) is successfully used as a standard characterization technique in many different research fields (batteries, capacitors, coatings, photovoltaics, etc.) and could be also very useful for TEs. However, only a few studies showed the potential of this technique to characterize TE materials and devices at the beginning of this thesis. For this reason, the main objective of this work is to advance the application of IS in the TE field in order to potentially establish it as a standard method in thermoelectricity.
Our results have shown the possibility to determine σ, λ and zT in TE materials with good accuracy at temperatures up to 250 ºC using IS, even for a material with modest TE properties. Moreover, we developed a new method capable of determining all these properties and the Seebeck coefficient (complete characterization) in a single IS measurement. We also extended the initial ideal impedance models (equivalent circuits) for the characterization of TE modules to include the convection effect at the outer ceramic surfaces for suspended TE devices, and the effect of the thermal contact resistance between the TE device and ideal heat sinks for sandwiched configurations. Finally, we developed a more comprehensive equivalent circuit, which includes all the key phenomena that could take place in a TE device. This model included the metallic strips that connect the TE legs in a TE device, the thermal contact resistances inside the TE module (TE legs/metallic strips and metallic strips/ceramic layers), and the spreading-constriction effects.
All these developments establish the impedance method as a powerful technique in thermoelectricity, allowing the complete characterization of TE materials from a single measurement, opening up the possibility of using this technique as a tool to quantify and monitor relevant thermal parameters, such as the convection heat transfer coefficient (h) and thermal contact resistances. Furthermore, our results also open the possibility of using IS as a quality control tool to detect and monitor in great detail issues in TE devices.
