arXiv:2610.15748
Variational Minimization of Almost Strong Zero Modes in Parafermionic Wires
Niklas Tausendpfund, Matteo Rizzi
Abstract
Strongly interacting fermionic chains obeying a $Z_n$ symmetry can host topological phases with highly degenerate ground-state manifolds. The different ground states are connected by strongly bound fractionalized edge modes called parafermionic zero modes. While exactly preserved under time evolution at zero temperature, interactions lead to a destabilization of these modes at finite energy densities, and, given a generic Hamiltonian, obtaining the lifetime of these modes is often impossible as it requires knowledge of the full spectrum of the Hamiltonian or long time evolutions. Here, we propose a novel quantum-inspired algorithm that can be used to variationally search for these almost strong zero modes in generic parafermionic chain models. We demonstrate analytically and numerically that our ansatz faithfully reproduces the exact results and finds strongly bound parafermionic operators in the non-integrable Kitaev-Hubbard chain and chiral Potts model. Our results further allow us to verify the existence of a true strong zero mode in the symmetrical limit of the chiral Potts model for $Z_3$-parafermions.
AI Summary
The paper addresses how to construct edge-localized strong and almost strong zero modes in interacting $\mathbb{Z}_n$ parafermionic chains without full-spectrum calculations or long-time evolution. It formulates the search as minimization of the normalized Frobenius commutator cost $\mathcal{F}(\Gamma)=n^{-N}\operatorname{Tr}([H,\Gamma]^\dagger[H,\Gamma])$ subject to parafermionic algebra, unitarity, and symmetry-charge constraints. The variational ansatz dresses a single-site edge operator with a symmetry-preserving ladder circuit of overlapping $k$-site unitaries, optimized using Riemannian methods and automatic differentiation. Unitary conjugation preserves the algebraic constraints exactly. For the $n=2$ Kitaev–Hubbard chain, explicit two-site circuits reproduce known strong zero modes on both integrable lines. Numerical optimization recovers exponentially decreasing commutator costs and decay exponents consistent with the exact expressions $-\log(\mu/2w)$ and $-\tfrac12\log(U/w)$. Away from integrability, costs saturate with chain length and block size; at $\mu=0.2w$, $U=0.4w$, the reported limiting value is approximately $\mathcal{F}/w^2=10^{-3}$. For the $\mathbb{Z}_3$ chiral Potts chain, calculations find edge-overlapping operators with finite residual costs, approximately $10^{-2}w^2$ at $\mu=w$, $\varphi=\pi/12$. Near the special integrable angle $\varphi=\pi/6$, costs reach the reported numerical threshold for weak coupling, interpreted as evidence for a true strong zero mode rather than an analytic proof. The paper states that more comprehensive convergence studies, broader parameter scans, and alternative circuit architectures are needed for conclusive nonintegrable results. Larger block sizes are computationally expensive; representative nonintegrable calculations use $N=16$, $k=6$ for $n=2$ and $N=12$, $k=4$ for $n=3$.
Central claim may not hold: The claimed verification of a true Z3 strong zero mode rests on a small finite-system cost near an optimization threshold, without demonstrating the required asymptotic decay. The ladder circuit does not guarantee exponential edge localization merely because its first gate has a nonzero identity component, so an essential zero-mode constraint is not automatically enforced. Saturation of variational upper bounds within the tested circuits cannot establish a strictly positive optimal commutator cost or exclude a true strong zero mode. The squared commutator cost determines initial autocorrelation curvature, not a general inverse lifetime, leaving the paper's quantitative lifetime estimates unsupported. The stated parafermionic parity operator contradicts both the common shift relation and its claimed product-of-tau spin representation for n=3.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 14 pages, 8 Figures