arXiv · 2611.22505
Quantum Optimal Control of Many-Body States in an Antiferromagnetic XXZ Chain with a Single Local Magnetic Field
Abstract
We investigate the quantum optimal control of an antiferromagnetic XXZ spin chain with open boundary conditions, driven solely by a local time-dependent magnetic field applied to a single terminal spin. By optimizing the control field via Krotov's algorithm, we evaluate the feasibility of preparing multi-particle many-body states, including highly polarized Greenberger-Horne-Zeilinger (GHZ)-type states and study the mechanisms by which the control is applied. We demonstrate that global spin-flip symmetry restricts reachable state spaces, strictly confining dynamics within the parity sector of the initial ground state and requiring parity-adapted linear combinations to avoid fundamental population upper bounds. Beyond symmetry constraints, driven population transfer does not occur via direct multi-spin flips, but proceeds stepwise across intermediate magnetization sectors ($|S_{\mathrm{tot}}^z| = 0 \to 1 \to 2 \to 3$) mediated by resonant frequency components matching the internal energy gaps. To overcome optimization stagnation caused by the small overlap mechanism in gradient updates, we introduce a four-stage sequential target protocol and compared it to spectrally filtered, time-localized frequency seed pulses. Combining frequency-tailored seed fields with an extended propagation time window yields GHZ state preparation fidelities exceeding $0.927$ establishing that a single local actuator is sufficient for high-fidelity multipartite state engineering in interacting spin networks.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Nicolás Rojo, Ignacio R. Solá, Juan J. Omiste. 2026-11-17. Quantum Optimal Control of Many-Body States in an Antiferromagnetic XXZ Chain with a Single Local Magnetic Field. https://arxiv.science/abs/2611.22505
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.