arXiv:2609.35655
Optimal performance in a chaotic Floquet quantum battery
Lakshmija N K, Uma Divakaran
Abstract
We study a quantum battery consisting of $N$ qubits interacting uniformly with each other. Starting from its ground state, the battery is charged by introducing periodic kicks in the transverse direction along with adding disorder in the interaction direction. Although the battery Hamiltonian along with the initial state preserves the permutation symmetry, the dynamics in presence of disorder during the charging phase takes the system out of the permutation symmetric subspace, resulting to a larger amount of stored energy as compared to that without disorder. In the limit of large disorder, this interplay brings in a possibility of a chaotic charging phase when stored energy can be exactly calculated using random matrix theory. More importantly, we also find that the energy fluctuations decrease and become negligible in the limit of large disorder, which is also when the energy stored is maximum, a condition highly desirable for the optimum functioning of the battery. We complete the battery cycle by considering the storage phase in the presence of dephasing noise. We find that for larger disorder strengths, the ergotropy remains close to the stored energy, indicating nearly complete energy extractability with an efficiency approaching unity.
AI Summary
The paper asks whether disorder-induced symmetry breaking in a collectively interacting Floquet system can improve quantum-battery charging and subsequent work extraction. The battery has Hamiltonian $H_B=-(k/N)J_x^2$ and initially occupies the polarized ground state. Charging combines Gaussian-disordered all-to-all $xx$ interactions with periodic transverse kicks, fixing the kick angle to $p=\pi/2$. Disorder breaks permutation symmetry, allowing dynamics beyond the $(N+1)$-dimensional symmetric subspace. Finite-size simulations, including $N=14$, average over 100 disorder realizations and use kicks 600–1000 for long-time averages. Increasing disorder raises the long-time stored energy and suppresses oscillations and energy variance. At strong disorder, these quantities approach full-Hilbert-space random-state benchmarks: stored energy $k(N-1)/4$ and energy variance $k^2(N-1)/(8N)$. These follow from trace calculations of $J_x^2$ and $J_x^4$; the absolute variance remains finite as $N$ grows. Floquet diagonal-ensemble calculations reproduce the saturated energy. For $k=1$, finite-size variance curves cross near disorder strength $w\sim3$, interpreted as an ergodic crossover. Storage is modeled by collective energy dephasing, $\dot\rho=-i[H_B,\rho]-\gamma[H_B,[H_B,\rho]]$. Starting after 800 charging kicks, simulations with $\gamma=0.02$ through storage time 2000 show conserved stored energy and weak ergotropy loss; the extractable fraction approaches unity at strong disorder. The analysis is restricted to numerically accessible finite sizes. Noise during charging and cycle reinitialization are not modeled; investigating open-system charging is identified as future work.
Central claim may not hold: The predicted large-disorder quantum energy variance remains k^2(N-1)/(8N), so the abstract's claim that energy fluctuations become negligible with increasing disorder requires a relative-fluctuation interpretation that is not stated. The first kick already attains the quoted chaotic stored-energy plateau and its energy variance at every disorder strength, so these benchmarks establish improved sustained charging rather than a disorder-enabled maximum energy or charging power. The appendix's purported exact Haar-pure-state variance benchmark omits the factor D/(D+1), although this correction is numerically small in the full Hilbert space.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 12 pages, 9 figures