Thermopower across the phase diagram of the cuprate La$_{1.6-x}$Nd$_{0.4}$Sr$_x$CuO$_4$ : signatures of the pseudogap and charge-density-wave phases

TL;DR

This study measures the thermopower of La$_{1.6-x}$Nd$_{0.4}$Sr$_x$CuO$_4$ across its doping range, revealing signatures of the pseudogap and charge-density-wave phases, and clarifying their effects on carrier density and Fermi surface reconstruction.

Contribution

It provides direct thermopower evidence for the pseudogap's impact on carrier density and identifies the doping level where charge-density-wave order causes Fermi surface reconstruction.

Findings

Pseudogap phase reduces carrier density from 1+p to p below p*

Thermopower increases sharply below p*

Charge-density-wave order causes Fermi surface reconstruction below p ≈ 0.19

Abstract

The Seebeck coefficient (thermopower) $S$ of the cuprate superconductor La$_{1.6-x}$Nd$_{0.4}$Sr$_x$CuO$_4$ was measured across its doping phase diagram (from $p = 0.12$ to $p = 0.25$), at various temperatures down to $T \simeq 2$ K, in the normal state accessed by suppressing superconductivity with a magnetic field up to $H = 37.5$ T. The magnitude of $S/T$ in the $T=0$ limit is found to suddenly increase, by a factor $\simeq 5$, when the doping is reduced below $p^{\star} = 0.23$, the critical doping for the onset of the pseudogap phase. This confirms that the pseudogap phase causes a large reduction of the carrier density $n$, consistent with a drop from $n = 1 + p$ above $p^{\star}$ to $n = p$ below $p^{\star}$, as previously inferred from measurements of the Hall coefficient, resistivity and thermal conductivity. When the doping is reduced below $p = 0.19$, a qualitative change is observed whereby $S/T$ decreases as $T \to 0$, eventually to reach negative values at $T=0$. In prior work on other cuprates, negative values of $S/T$ at $T \to 0$ were shown to result from a reconstruction of the Fermi surface caused by charge-density-wave (CDW) order. We therefore identify $p_{\rm CDW} \simeq 0.19$ as the critical doping beyond which there is no CDW-induced Fermi surface reconstruction. The fact that $p_{\rm CDW}$ is well separated from $p^{\star}$ reveals that there is a doping range below $p^{\star}$ where the transport signatures of the pseudogap phase are unaffected by CDW correlations, as previously found in YBa$_2$Cu$_3$O$_y$ and La$_{2-x}$Sr$_x$CuO$_4$.