Giant thermopower changes related to the resistivity maximum and colossal magnetoresistance in EuCd2P2

TL;DR

The paper investigates an extraordinary thermopower response in EuCd2P2 that coincides with a dramatic resistivity maximum in the 10–25 K range. Using temperature- and field-dependent S, the authors show two sign changes and |S| values exceeding 4000 μV/K over a span of less than 5 K, with the anomaly almost completely suppressed by small magnetic fields. They introduce a drift-diffusion picture that links S to strong resistivity gradients via the relation $S_{diff} = (k_B/q)[1 + d( ln sigma)/d( ln T)]$ with $ sigma = 1/ ho$, achieving parameter-free agreement with the observed data and arguing against phonon-drag as the primary origin. Although the resulting zT is small due to high ρ, the work demonstrates a general mechanism to realize giant thermopower through internal electronic-property gradients, suggesting routes to tunable thermoelectric responses by engineering gradients or compositional variations.

Abstract

We present the thermopower of EuCd2P2, a material which exhibits a large resistivity peak with significant magnetic field dependence in the temperature range of 10-25 K. In the same region we observe a highly unusual behavior of the thermopower with two sign changes and giant extrema. The overall variation of the thermopower exceeds 4 000 muV/K and takes place in an extremely narrow temperature region of less than 5 K. The anomaly is suppressed completely in a small magnetic field of 0.5 T. We discuss this observation using a simple drift-diffusion picture and taking into account that the temperature gradient inducing the thermopower voltage is accompanied by a gradient of the electrical resistivity. Our simple estimation yields the correct magnitude, shape, and field dependence of the thermopower anomaly observed in EuCd2P2. These results open a new route to giant thermopower values via gradients of electronic properties.

Figures (6)

• Figure 1: Comparison of thermopower and resistivity of EuCd$_2$P$_2$. a Thermopower $S$ in the full investigated temperature range from 2 K to 300 K. The inset shows the high-$T$ part of $S(T)$ on a larger scale. b Electrical resistivity $\rho(T)$ measured on the same sample and contacts in zero magnetic field and $B = \mu_0 H = 1$ T.
• Figure 2: Thermopower of EuCd$_2$P$_2$ in magnetic fields.a Temperature dependence of the thermopower $S$ of EuCd$_2$P$_2$ at different magnetic fields $B = \mu_0 H$. The anomaly is rapidly suppressed by small fields. The results from $T$ sweeps (lines with small symbols) are in good agreement with those from $B$ sweeps (asterisks). The Néel temperature $T_\mathrm{N}$ in zero field is marked by an arrow. The inset shows the position of the maxima in $\rho(T)$ and of sign changes in $S(T)$ and $S(B)$. The lines are guides to the eye. b Field dependence of the thermoelectric power $S$ of EuCd$_2$P$_2$ at different temperatures. $B$ sweeps (lines with small symbols) and $T$ sweeps (asterisks) exhibit the same behavior.
• Figure 3: Drift-diffuion model of the thermopower. Temperature dependence of the thermoelectric power $S$ of EuCd$_2$P$_2$ at different magnetic fields. Lines are calculations from the electrical resistivity based on a parameter-free drift-diffusion model as explained in the main text.
• Figure 4: Electrical resistivity of EuCd$_2$P$_2$. Comparison of the electrical resistivity $\rho$ of EuCd$_2$P$_2$ obtained on the same contacts, but with two different measurement techniques. Data points have been measured using the TTO. The lines correspond to measurements with the ACTO. Both methods yield the same results at high $T$ and $B$. However, the TTO was not able to resolve very large resistances, as observed in low fields below 20 K in EuCd$_2$P$_2$.
• Figure 5: Thermal conductivity $\kappa$ of EuCd$_2$P$_2$. Measurements in zero field and 1 T yielded very similar results indicating a negligible field dependence of $\kappa$ in this region.