Spin-Gap Proximity Effect Mechanism of High Temperature Superconductivity

TL;DR

This paper proposes a theory where high-temperature superconductivity arises from a spin-gap proximity effect in doped antiferromagnetic insulators, involving stripe formations and local spin-charge separation, leading to pairing without large mass renormalization.

Contribution

It introduces a novel mechanism for pairing in cuprates based on spin-gap proximity effect in stripe-structured Mott insulators, supported by exact solutions and experimental relevance.

Findings

Spin gaps form in Mott-insulating regions near metallic stripes.

Pairing occurs via a magnetic proximity effect without large mass renormalization.

Long-range superconductivity emerges from Josephson coupling between stripes.

Abstract

When holes are doped into an antiferromagnetic insulator they form a slowly fluctuating array of ``topological defects'' (metallic stripes) in which the motion of the holes exhibits a self-organized quasi one-dimensional electronic character. The accompanying lateral confinement of the intervening Mott-insulating regions induces a spin gap or pseudogap in the environment of the stripes. We present a theory of underdoped high temperature superconductors and show that there is a {\it local} separation of spin and charge, and that the mobile holes on an individual stripe acquire a spin gap via pair hopping between the stripe and its environment; i.e. via a magnetic analog of the usual superconducting proximity effect. In this way a high pairing scale without a large mass renormalization is established despite the strong Coulomb repulsion between the holes. Thus the {\it mechanism} of pairing is the generation of a spin gap in spatially-confined {\it Mott-insulating} regions of the material in the proximity of the metallic stripes. At non-vanishing stripe densities, Josephson coupling between stripes produces a dimensional crossover to a state with long-range superconducting phase coherence. This picture is established by obtaining exact and well-controlled approximate solutions of a model of a one-dimensional electron gas in an active environment. An extended discussion of the experimental evidence supporting the relevance of these results to the cuprate superconductors is given.