On the nonexistence of degenerate phase-shift multibreathers in a zigzag Klein-Gordon model

TL;DR

This paper investigates the existence of low amplitude multibreathers in zigzag Klein-Gordon chains with extended interactions, demonstrating nonexistence of certain degenerate phase-shift solutions through bifurcation analysis and NL-dNLS comparisons.

Contribution

It introduces a novel approach using Lyapunov-Schmidt decomposition and perturbation analysis to prove the nonexistence of degenerate phase-shift multibreathers in zigzag KG models.

Findings

Nonexistence of degenerate phase-shift multibreathers in zigzag KG chains.

Bifurcation equation approximated by a perturbed NL-dNLS model.

Standard methods are ineffective for square lattice-like models due to higher degeneracy.

Abstract

In this work, we study the existence of low amplitude four-site phase-shift multibreathers for small values of the coupling $\epsilon$ in Klein-Gordon (KG) chains with interactions longer than the classical nearest-neighbour ones. In the proper parameter regimes, the considered lattices bear connections to models beyond one spatial dimension, namely the so-called zigzag lattice, as well as the two-dimensional square lattice. We examine initially the persistence conditions of the system, in order to seek for vortex-like waveforms. Although this approach provides useful insights, due to the degeneracy of these solutions, it does not allow us to determine if they constitute true solutions of our system. In order to overcome this obstacle, we follow a different route. In the case of the zigzag configuration, by means of a Lyapunov-Schmidt decomposition, we are able to establish that the bifurcation equation for our model can be considered, in the small energy and small coupling regime, as a perturbation of a corresponding non-local discrete nonlinear Schr\"odinger (NL-dNLS) equation. There, nonexistence results of degenerate phase-shift discrete solitons can be demonstrated by exploiting the expansion of a suitable density current of the NL-dNLS, obtained in recent literature. Finally, briefly considering a one-dimensional model bearing similarities to the square lattice, we conclude that the above strategy is not efficient for the proof of the existence or nonexistence of vortices due to the higher degeneracy of this configuration.