Skip to main content

Thermoelectric power generation from anomalous Nernst effect based on rare earth free hard magnetic materials

Final Report Summary - THERMO-SPIN (Thermoelectric power generation from anomalous Nernst effect based on rare earth free hard magnetic materials)

This project explored the concept of thermoelectric power generation (TEG) based on the anomalous Nernst effect (ANE). The research plan involved the construction of two experimental setups to evaluate the ANE in thin film structures made of carefully pre-selected magnetic materials
for best performance. The project considered special hard and also ferrimagnetic materials to harness electric energy from in the form of thin film laminate structures, where external magnetic field is not required for continuous operation. The investigation of the fundamental, quantum- mechanical origin that is responsible for the ANE in these materials was also the subject of the study. In the first period of the project, two measurement benches were built for accurate measurements of temperature gradients (in-plane and out-of- plane) in thin films in a large temperature (100-400K) and magnetic field (0-5T) ranges, respectively. The ambiguity of such measurements lies in the comparatively large heat capacity of the temperature sensors attached to the surface of the films. This problem seriously hinders the progress of accurate thermomagnetic analysis of magnetic films as comparative analysis of experimental results (in absolute value) is impossible. In order to resolve the problem of temperature sensing at nanoscale, a calorimetric approach was adopted where the heat flux transferred through the sample is measured, rather than the actual temperature. The technique was tested by measuring the conventional seebeck coefficient of known materials such as Al, Cu, Pt and the magnetocaloric La(FeSi)13 compound. The measurements successfully reproduced the values available in the literature for these materials after some minor corrections based on heat transfer models using a finite element approach as implemented in Comsol Multiphysics software package. The experimental apparatus was constructed in a way that allows the in-situ measurement of conventional seebeck and ANE simultaneously. In addition, in a thermally equilibrated operation, the setup (in-plane) allowed us to also measure other important physical properties (resistivity,
anomalous Hall effect) without any alternation to the actual sample arrangement and electric contacts. This is especially valuable for the understanding of the occurrence of ANE at the fundamental level.
In the second period of the project, based on our selection procedure for best suited materials, we decided to integrate the highly unusual ferrimagnetic Mn2-xRuxGa (MRG) alloy into a thin film structure. MRG is a ferrimagnet with two compensated magnetic sublatices that can be fine-tuned by both Ru concentration and substrate-induced strain. At this compensation point in temperature, the unusually large anomalous Hall effect (AHE) changes its sign. Similarly, we found the change of sign in the ANE at the compensation point in our experimental analysis. Figure 1. shows the in-situ measurements of electrical (AHE) and thermoelectric (ANE) properties of uniaxial MRG thin films. In order to simultaneously measure the electrical and thermoelectric properties, 30nm thick MRG films were prepared using magnetron sputtering technique and are subsequently patterned into Hall bars. The purpose-built sample stage is used to generate a linear in-plane temperature gradient along the Hall bar by turning on a heater at one end of the substrate. The sample environment is varied using a continuous-flow liquid helium cryostat within a superconducting magnet in the range of 77 - 400 K and 0 to 5 T, respectively. Both longitudinal and transverse thermal emf voltages (Vx and Vy) were recorded simultaneously. Seebeck (Sxx) and Nernst coefficients (Sxy) were calculated using the recorded emf voltages and temperature gradients. In addition, anomalous Hall resistivities (Rxx, Rxy) are also measured with no applied temperature gradients. Fig. 1(a) shows the temperature dependence of both AHE and ANE as a function of temperature. The dominant feature is the apparent sign change of Rxy (black line) at the ferrimagnetic compensation point (~235K). Indeed, the latter temperature coincides well with the change of sign of ANE (red line) implying the same intristic origin. However, contrary to the large observed AHE, the thermal equivalent ANE remains small that requires further analysis. The smallness of the ANE can be beneficial in devices aimed at exploiting thermal switching of net-compensated half-metals.

These simple lateral thin film structures also offer potential application to large size ( > square-meter) heat sources as the fabrication process to make a lateral thermopile structure is easier than for conventional SE-based perpendicular thermopile. Furthermore, the internal resistance of the
laminar structure can be tuned by the change in the thickness of the magnetic substances. The approach that is taken in this project by using ferrimagnetic materials close to the magnetic compensation point demonstrated that there are other viable routes to harness useful thermal
energies of waste heats, in addition to the coercivity engineering or tube shaped thermoelectric power generation available in the literature. However, the research on ANE is still at an explanatory stage and it requires further investigations in order to be competitive with TEG based on the
conventional Seebeck effect. At the present stage it is not clear why the used material (MRG) shows large AHE but small ANE. This requires further analysis at both fundamental and exploratory levels.

The harnessable electric power needs to be improved by one or two order of magnitudes in order to open up a possibility for practical usage. The fast pacing research into ANE materials could lead to more efficient TEG products in the next 5-10 years, thus it can play an important part in meeting EU targets to increase energy efficiency by 20% as set out by the European Commission.