Energy gaps in high-$T_c$ superconductors: BCS after all?

Abstract

A major impediment to solving the problem of high-$T_c$ superconductivity is the ongoing confusion about the magnitude, structure and doping dependence of the superconducting gap, $Δ_0$, and of the mysterious pseudogap found in underdoped samples\cite{TallonLoram}. The pseudogap opens around the ($π$,0) antinodes below a temperature $T^*$ leaving Fermi arcs across the remnant Fermi surface\cite{Kanigel} on which the superconducting gap forms at $T_c$. One thing that seems agreed is that the ratio $2Δ_0/k_BT_c$ well exceeds the BCS value and grows with underdoping\cite{Miyakawa1,Miyakawa2}, suggesting unconventional, non-BCS superconductivity. Here we re-examine data from many spectroscopies, especially Raman $B_{1g}$ and $B_{2g}$ scattering\cite{Sacuto,Guyard}, and reconcile them all within a two-gap scenario showing that the points of disagreement are an artefact of spectral-weight loss arising from the pseudogap. Crucially, we find that $Δ_0(p)$, or more generally the order parameter, now scales with the mean-field $T_c$ value, adopting the weak-coupling BCS ratio across the entire phase diagram.