Enhanced charging in open Rabi and Dicke quantum batteries

AI summary

Overview

Research area: Quantum thermodynamics and open quantum systems, specifically quantum batteries (QBs) built from cavity quantum electrodynamics architectures — the quantum Rabi model (one two-level system, TLS) and its Dicke extension (N TLSs sharing one cavity mode).

Technical level: Advanced. The paper works with Lindblad master equations, non-Hermitian effective generators for conditioned no-jump dynamics, ergotropy and passive-state constructions, and finite-size scaling scans.

Scope (one sentence): The paper asks whether direct incoherent pumping of the battery degrees of freedom can stabilize and improve charging in Rabi and finite Dicke quantum batteries relative to purely Hermitian (unitary) charging protocols.

Note on the source text: The provided content is truncated partway through the description of Fig. 7. Details of the later Dicke subsections (V.3, V.4) and the summary section (Sec. VI) are therefore only partially available, and this summary flags wherever the paper's values are not reported in the available text.

What This Paper Is About

A quantum battery stores energy in quantum degrees of freedom and should ideally hold that energy long enough and predictably enough to be useful. In cavity-based batteries the battery exchanges energy coherently with a cavity "charger" mode, which means the energy sloshes back and forth: the best charging moments are sharp, transient, and sensitive to exactly when you switch the interaction off, and to frequency detuning. This paper tests a different strategy — continuously and incoherently pumping the two-level systems themselves — and asks whether that open-system drive turns fragile, spiky charging into a broad, robust high-energy plateau, for a single Rabi battery and for Dicke batteries of finite size.

Key Contributions

1. A full open-system charging benchmark for the Rabi battery. The authors compare an unpumped Hermitian reference against Lindblad dynamics with direct incoherent pumping, using three couplings (g/ω₀ = 0.1, 0.5, 1) and a dense scan of 41 calculated pump strengths (γ/ω₀ = 0, 0.05, …, 2).

2. A detuning-robustness map. They compute the maximum stored energy across Δ/ω₀ ∈ [−1, 1] at the reference coupling g/ω₀ = 0.1 for γ/ω₀ = 0.5, 1, 2, showing that pumping replaces the narrow Hermitian resonance with a broad high-energy region.

3. An observable-level "usage map" for the conditioned non-Hermitian description. Rather than claiming the no-jump generator is a general approximation, they quantify where it agrees with the full Lindblad dynamics for two different observables, using a peak-energy discrepancy δE_max and a time-averaged reduced-state infidelity I_B, and they check the statistical weight of the no-jump sector.

4. A finite-size Dicke scan plus an equal-input architecture comparison. They study a representative N = 10 Dicke battery, then scan N = 4, 6, …, 20 at four pump strengths, and finally compare a common-cavity Dicke battery against N independent Rabi batteries under an accumulated pump-input bound.

Main Findings

• The Hermitian reference has a built-in tradeoff. In the unpumped Rabi battery, the first-cycle charging energy decreases with increasing g/ω₀, while the first-cycle average power increases. Timing robustness R₀.₉₅ also decreases with increasing g, so the faster unitary cycles are the more fragile ones. Because g/ω₀ = 0.1 gives the largest first-cycle stored energy in the scan, it is adopted as the reference coupling.

• Pumping suppresses coherent backflow and creates a plateau. At the reference coupling, comparisons between the Hermitian dynamics and the pumped Lindblad dynamics show that pumping removes the pronounced recurrent discharge, produces a broad high-energy plateau, and increases the maximum average charging power — with the caveat that pump jumps themselves contribute to the Lindblad energy gain, so this power is a total open-system charging rate, not a purely cavity-mediated transfer rate.

• Stronger coupling limits the benefit. At g/ω₀ = 0.5 and g/ω₀ = 1, pumping at γ/ω₀ = 1 still strongly reduces recurrent energy return, but the peak stored energies remain below the corresponding unpumped maxima. Timing robustness improves from 0.355 to 0.500 at g/ω₀ = 0.5 and from 0.200 to 0.231 at g/ω₀ = 1, so the robustness gain becomes more modest as g grows.

• Detuning tolerance is substantially broadened. The Hermitian reference shows a narrow resonance peak, slightly shifted from exact resonance by the Bloch–Siegert correction arising from the counter-rotating terms of the full Rabi model. The pumped Lindblad curves instead stay near a broad high-energy plateau over a much wider detuning window.

• The no-jump description is accurate only in observable-dependent regions. Agreement between the conditioned no-jump evolution and the full Lindblad evolution depends on which quantity is examined and on the parameter plane. In detuning maps the low-discrepancy region broadens rapidly with pumping, whereas coupling maps retain nonmonotonic higher-discrepancy structures at larger g. The markers at δE_max = 0.05 ℏω₀ and I_B = 0.05 are explicitly described as guides, not universal validity boundaries.

• A rare no-jump sector can still look correct. At γ/ω₀ = 1 the no-jump probability is only P_nj = 0.0125 at the normalized no-jump energy maximum, yet the fidelity between the no-jump and jump-conditioned reduced battery states is 0.9985. Both sectors approach highly excited reduced battery states, so reduced-state agreement does not make no-jump histories dominant in the Lindblad ensemble, nor does it establish equivalence of the full cavity–battery states.

• Pumping increases extractable work at the studied Rabi operating points. In the separate ergotropy comparison at g/ω₀ = 0.5, pumped protocols accelerate charging and, at their ergotropy-optimal times, retain a larger extractable fraction than the Hermitian reference. At γ/ω₀ = 2 the ergotropy nearly reaches the stored energy at the marked ergotropy peak time. Accumulated inputs at that time are Q_in^sys/ℏω₀ = 1.043 (γ/ω₀ = 1) and 1.084 (γ/ω₀ = 2).

• The Dicke battery responds qualitatively like the Rabi battery. At N = 10 and g/ω₀ = 0.1, the Hermitian Dicke battery shows recurrent charging and discharging with substantial backflow after the first maximum. Pumping at γ/ω₀ = 0.2 mainly reshapes the oscillations, while the traces at γ/ω₀ = 0.5 and 1 approach a stable high-energy plateau. Peak stored energy per cell initially changes little from its γ = 0 value and then rises above the Hermitian value as pumping increases; maximum average charging power per cell increases more strongly; timing robustness improves at γ/ω₀ = 0.5 and 1.

• Power gains must be attributed carefully. The energy-flow decomposition shows that the finite-time power maximum at γ/ω₀ = 0.1 contains comparable cavity and pump contributions, whereas the maxima at γ/ω₀ = 0.5 and 1 are set by the initial pump input. The reported values include the analytical t → 0⁺ limit ΔP/(Nℏω₀²) = 2γ/ω₀. The authors state that the increased maximum average charging power therefore does not by itself establish an intrinsic cavity-mediated or collective charging advantage.

• Finite-size behavior is pump-dependent (partially available). According to the abstract and introduction, the finite-size scan shows observable-dependent tradeoffs at the two lower tested pump strengths, whereas at the largest tested value (γ/ω₀ = 1) the stored energy, ergotropy, and timing robustness stay above the Hermitian values throughout the tested size range (N = 4 to 20).

• Architecture ordering depends on the input accounting (partially available). The equal-input comparisons show that the relative performance of common-cavity Dicke batteries and independent Rabi batteries varies with the pump-input bound; the specific numerical crossover is not reported in the available text.

Methodology in Plain English

The authors model the battery as a cavity mode coupled to one two-level system (Rabi) or to N two-level systems sharing one cavity (Dicke). Energy is stored in the two-level systems; the cavity acts as the charger that shuttles energy in and out. In the "reference" protocol nothing else happens, so energy sloshes back and forth indefinitely.

The new ingredient is a continuous, noisy drive that randomly flips each two-level system from ground to excited at rate set by γ. This is written as a Lindblad master equation, the standard tool for describing a quantum system that is being driven and simultaneously losing coherence into an environment. Because the drive constantly pushes population upward, it competes with the coherent hand-back of energy from battery to cavity.

To keep the comparison fair, the authors define a consistent scorecard: charging energy (energy gained by the two-level systems), average charging power (charging energy divided by time), the peak values of each, a "timing robustness" score that measures how wide a window stays within 95% of a curve's own maximum, and ergotropy, which measures how much of the stored energy can actually be extracted as work rather than being locked in a passive state. For the ergotropy comparison they also adopt an explicit bookkeeping rule: the unpumped reference is powered by one photon preloaded in the cavity, while pumped protocols start with an empty cavity and are powered only by the pump, with cumulative pump input tracked.

Separately, they derive a non-Hermitian "no-jump" generator, which describes how the system drifts if you condition on the pump having produced no quantum jump. This is used only as an interpretive tool: agreement with the full Lindblad answer is mapped out for two different observables, and the no-jump probability is computed to check whether that sector is statistically meaningful. An approximate reduction to a non-Hermitian Jaynes–Cummings model gives analytic insight near resonance, valid when g ≪ ω₀, ω_c and, at exact resonance, when γ > 2g.

Why This Matters

Impact on research. The paper reframes quantum-battery charging as an open-system design problem rather than a purely unitary one. Its most transferable methodological point is negative but important: agreement between a non-Hermitian effective model and a full Lindblad simulation can hold for one observable, in one parameter region, while the underlying conditional sector carries almost no probability weight (P_nj = 0.0125 in the reported case). That is a caution for the wider literature that uses non-Hermitian generators as shortcuts.

Potential applications (implied by the paper's cavity-QED framing, not demonstrated in it):

• Nanoscale and mesoscale energy storage engineered in superconducting circuit QED, where cavity modes and artificial two-level systems are already controllable.
• Quantum thermal machines and micromasers, where incoherent pumping is a standard resource and stable high-energy operating points matter more than peak transients.
• Charge-stability engineering for repeated operation, since broadening the high-energy window reduces the need for precisely timed decoupling operations, which is the dominant practical weakness of unitary charging.
• Detuning-tolerant device design, relevant when fabrication spread or frequency drift makes exact resonance impractical.

Industry relevance. The results speak to platforms where cavity-mediated light–matter coupling is a fabrication reality — superconducting circuits, circuit QED architectures used in quantum computing, and cavity-embedded solid-state emitters. The architecture question the paper raises, common-cavity collective charging versus N independent units under the same input budget, is directly relevant to anyone deciding whether to scale a device by adding emitters to one resonator or by replicating modules. The paper does not claim an unconditional advantage for either choice; the ordering depends on how input is accounted for, which is exactly the kind of question a device engineer would need answered.

Future Directions

• Establish whether there is an intrinsic collective advantage. The authors state explicitly that the increased maximum average charging power includes pump-assisted jump contributions and does not by itself demonstrate a cavity-mediated or collective advantage. A decomposition that isolates the cavity channel at fixed total input is the natural next test.

• Complete the finite-size and equal-input picture at larger N and other couplings. The scan covers N = 4 to 20 at g/ω₀ = 0.1 with four pump strengths (γ/ω₀ = 0, 0.1, 0.5, 1). Whether the largest-pump advantage survives at N beyond 20 and at stronger coupling is untested here.

• Extend the non-Hermitian validity map into the strong-coupling regime. The analytic Jaynes–Cummings reduction requires g ≪ ω₀, ω_c and, at exact resonance, γ > 2g; the authors explicitly decline to extrapolate it to the strong-coupling results at g/ω₀ = 0.5 and 1. A theory covering that regime is open.

• Turn the input-bookkeeping convention into a full thermodynamic accounting. The authors describe their yield η_in^W as a diagnostic of the selected operating point, not a complete thermodynamic efficiency of the pump reservoir. A reservoir-level treatment, including the cost of maintaining the pump, remains to be done.

Target Audience

Researchers in quantum thermodynamics and quantum information who work on quantum batteries, charging protocols, and ergotropy. Also well suited to open-quantum-systems theorists interested in when non-Hermitian effective descriptions are trustworthy, since the paper's observable-level validity mapping and its no-jump probability check are directly reusable. Experimentalists in cavity QED and superconducting circuits will find the architecture comparison and the timing- and detuning-robustness framing relevant, though the paper is theoretical throughout. Readers need comfort with Lindblad master equations, partial traces, and passive-state definitions; the physical narrative is clear, but the quantitative content assumes graduate-level quantum optics.

Authors’ abstract

We investigate open-system charging protocols for Rabi and Dicke quantum batteries under direct incoherent pumping. The charging dynamics follows a Lindblad master equation, while effective non-Hermitian generators describe the conditioned no-jump mechanism. For a single Rabi battery, this pumping suppresses coherent energy backflow, broadens the high-energy charging window, and improves robustness against both timing errors and detuning. At the selected Rabi operating points, pumping also increases ergotropy and the extractable fraction of the stored energy. For finite Dicke batteries, direct incoherent pumping likewise stabilizes the charging dynamics. Across the finite-size scan, the strongest pump considered improves the peak stored energy and ergotropy per cell and near-peak timing robustness relative to the Hermitian reference. The maximum average charging power also increases, although this metric includes pump-assisted jump contributions. Equal-input comparisons show that the relative performance of common-cavity Dicke and independent Rabi batteries varies with the pump-input bound. Taken together, these results show that direct incoherent pumping provides a route to stabilized charging in Rabi and finite Dicke quantum batteries.