Experimental Investigation of the Nernst Effect for Thermomagnetic Applications

Thermal to electrical energy conversion using thermoelectric devices built based on the Seebeck effect has been intensely explored over the past two centuries. These devices not only convert heat into electricity but also operate in reverse, acting as solid-state refrigerators or coolers by actively pumping heat. Thermomagnetic devices are alternatives to thermoelectric devices and are built based on the Nernst effect. When temperature gradient and magnetic field are applied perpendicular to each other in a given material, a transverse voltage develops due to the Nernst effect in the direction normal to the plane containing the magnetic field and temperature gradient vectors. The Nernst coefficient is then defined as the ratio of the transverse electric field to the longitudinal temperature gradient, normalized to the applied magnetic field. The material parameters influencing the response of the Nernst effects are relatively unexplored compared to the well-known Seebeck effect.

This PhD thesis presents several significant findings. Firstly, an evaluation of the applicability of Moreau’s relation in predicting the magnitude and trend of the Nernst coefficient in the standard thermoelectric material, Bi2Te3, a narrow-band gap semiconductor, is undertaken. According to this relation, the Nernst coefficient of a material can be written as the negative of the product of its Thompson coefficient (defined as the product of the temperature of the sample and the first derivative of its Seebeck coefficient with respect to the temperature), Hall coefficient, and electrical conductivity. Moreau’s relation is found to predict reasonably well the magnitude of Nernst coefficient and the temperature at which it changes sign. However, it could not explain the non-linear relationship between the Nernst coefficient and the applied magnetic field observed at low temperatures. As predicted by the relation, an increase in the Nernst coefficient magnitude is observed as the mobility of the carriers increases at lower temperatures.

Secondly, a larger Nernst coefficient is observed in the high-temperature phase of 1T-TiSe2 where the mobility is the lowest. 1T-TiSe2 is a layered semimetal representing a distinctive class of materials that exhibits slight overlap between their valence and the conduction band, accompanied by a high degree of carrier compensation. We attribute this increase to an abrupt reconstruction of the Fermi surface driving a rapid increase in the size of electron-hole pockets which leads to an increased bipolar effect and enhanced Nernst coefficient.

Thirdly, an enhanced Nernst signal peak in MoTe2 is observed. MoTe2 is a topological Weyl semimetal that exhibits both quadratic and linear dispersion in its band structure. Analyzing the transport data, we attribute this to the movement of the Fermi level in crossing the Dirac point. A similar study was carried out on PdTe2, a Dirac semimetal. Even after changing the temperature from 2 − 400 K and the magnetic field from −9 T to +9 T, such band crossing was not observed. This highlights the importance of EDirac point−EF as an important selection parameter in recognizing potential topological semimetals for thermomagnetic applications.

Fourthly, a sign change in the Nernst coefficient during the antiferromagnetic (AF) - ferromagnetic (FM) transition of FeRh/Al2O3, a metal, is recorded. The differing signs in the two phases were attributed to an anisotropic mobility of spin-up and spin-down electrons in the AF phase. Additionally, despite the AF phase displaying higher mobility, the magnitude of the Nernst coefficient was higher in the FM phase owing to the contribution from its internal magnetization. A hysteresis loop was evident in the temperature-dependent Nernst coefficient response with a shape similar to that observed in the Seebeck coefficient but with a narrower width.

Lastly, a significant Nernst coefficient of 11.2 μVK−1T−1 at 80 K is observed in the F4TCNQ-doped P3HT polymer which is comparable to the response in many inorganic materials. The basis for this is not yet clear due to the polymers exhibiting much more complex molecular structures than the inorganic materials. As of now, this remains an open question and is a potential starting point for future work.

The practical implications of this PhD thesis lie in identifying efficient thermomagnetic materials that can be used in building environmentally friendly waste-heat to useful energy converters.