Theory of Polarized Neutron Scattering in the Loop Ordered Phase of Cuprates

TL;DR

This paper develops a quantum-mechanical theory for polarized neutron scattering in the loop-ordered phase of cuprates, explaining the observed transverse magnetic components and providing estimates consistent with experimental data.

Contribution

It introduces a novel quantum-mechanical framework for analyzing polarized neutron scattering in cuprates' loop-ordered phase, revealing the origin of transverse magnetic components.

Findings

Finite transverse magnetic component is of quantum origin.

The model explains the tilt of magnetic moments observed experimentally.

Reasonable agreement with experimental dispersion data is achieved.

Abstract

The collective modes observed in the loop-current ordered state in under-doped cuprates by polarized neutron scattering require that the ground state is a linear combination in each unit-cell of the four basis states which are the possible classical magnetic moment configurations in each unit-cell. The direction of such moments is in the c-axis of the crystals. The basis states are connected by both time-reversal as well as spatial rotations about the center of the unit-cells. Several new features arise in the theory of polarized neutron scattering cross-section in this situation which appear not to have been encountered before. An important consequence of these is that a finite component transverse to the classical magnetic moment directions is detected in the experiments. We show that this transverse component is of purely quantum-mechanical origin and that its direction in the plane normal to the c-axis is not detectable, even in principle, in experiments, at least in the quantum-mechanical model we have adopted. We estimate the direction of the "tilt" in the moment, i.e. the ratio of the transverse component to the c-axis component, using parameters of the ground state obtained by fitting to the observed dispersion of the collective modes in the ordered state. We can obtain reasonable agreement with experiments but only by introducing a parameter for which only an approximate magnitude can be estimated. Approximate calculations of the form-factors are also provided.