Experimental Demonstration of a Measurement-Feedback Quantum Information Engine

AI summary

Overview

• Research area: Quantum thermodynamics, specifically measurement-powered quantum information engines using trapped-ion quantum hardware.
• Technical level: Intermediate (assumes familiarity with two-level systems, projective measurement, and basic thermodynamic accounting such as work, heat, and efficiency).
• Scope: An experimental realization of a measurement-feedback quantum information engine in a single trapped ⁴⁰Ca⁺ ion, showing that measurement-induced coherence and finite-time nonadiabatic transitions ("quantum inner friction") can increase extracted work and push efficiency above the corresponding Otto benchmark.

What This Paper Is About

Conventional quantum heat engines suffer from "quantum inner friction" — irreversible work lost when a stroke is driven too quickly for the system to follow adiabatically — and the standard strategy is to suppress it. This paper instead builds an engine in which a projective measurement replaces the hot reservoir as the energy source, and feedback based on the measurement outcome sends the system down either a unitary compression–expansion branch or a thermalization branch. The goal is to show experimentally that the nonadiabatic transitions normally treated as a penalty can instead be turned into a resource that boosts work extraction and efficiency.

Key Contributions

1. A new engine architecture, experimentally realized. The authors introduce and demonstrate a measurement-feedback quantum information engine (MFQIE) in which a projective measurement of the observable M(θ) = cos θ σ_z + sin θ σ_x replaces the hot reservoir of an Otto-type cycle, and the outcome conditionally steers the working medium into either a unitary compression–expansion stroke or a thermalization stroke with a cold reservoir.
2. A resolved energetic ledger for a single-ion engine. Using quantum-state tomography, the authors account for extracted work, measurement-induced energy input, heat exchange with the engineered cold reservoir, and a conditional Landauer reset charge assigned to the outcome record (the reset itself is not implemented as a physical stroke in this experiment), yielding an operational efficiency.
3. Demonstration that measurement-induced coherence raises performance. By tuning the measurement angle θ, the authors show that the engine crosses from a conventional regime (θ < 0.04π, efficiency below the Otto benchmark) into a measurement-enhanced regime above the benchmark, and corroborate the coherence contribution with a fully dephased reference-cycle comparison.
4. First experimental evidence that quantum inner friction can help. By varying the compression and expansion stroke durations, the authors show that the stroke-resolved inner-friction contribution to extracted work remains nonnegative for |θ| < π/2 and enhances both efficiency and intrinsic cycle power over finite ranges of stroke durations.

Main Findings

• Convergence to a stable cycle: Starting from the nonstationary initial state |g⟩, the repeated feedback map drives the branch probabilities as p⁽ⁱ⁺¹⁾ = P p⁽ⁱ⁾, which exponentially approach a stationary distribution, giving fixed thermodynamic quantities and efficiency after several iterations.
• Efficiency above the Otto benchmark: In the stable cycle with full thermalization, the measured efficiency stabilizes at 59.1(1.6)%, consistent with the theoretical value of 59.9% and surpassing the corresponding Otto benchmark η_OttO = 1 − ω_c/ω_h by 18.2%.
• Two operating regimes controlled by measurement angle: For small θ < 0.04π the device works as a conventional heat engine (work from heat absorbed from the cold bath, efficiency below the Otto benchmark); as θ increases, measurement-induced energy input supports work extraction while heat is released to the cold bath, and efficiency exceeds the benchmark.
• Coherence is not monotonic in efficiency: Efficiency does not strictly follow the monotonic increase of the relative-entropy-of-coherence measure C(θ), because changing θ also alters the post-measurement state evolution, and hence inner friction and heat dissipation.
• Dephased reference comparison: A self-consistent comparison against a fully dephased reference cycle (same post-measurement populations, energy-basis coherence removed, each engine iterated to its own steady cycle) shows that measurement-induced coherence yields higher steady-state work output and efficiency in the relevant parameter regime.
• Thermalization duration dependence: Efficiency increases with the thermalization duration τ_d and then saturates, following from the redistribution of steady-cycle branch probabilities and the rebalancing of work, measurement energy, heat, and erasure cost.
• Inner friction as a positive contribution: The quantum inner-friction contribution to work, ⟨w⟩_fri = p₊ε₀ cos θ (ξ_c + ξ_e − 2ξ_cξ_e), remains nonnegative whenever |θ| < π/2, so friction enhances rather than degrades the engine whenever it is nonzero.
• Damped-oscillatory duration dependence: Efficiency and power show duration-dependent damped-oscillatory behavior; at small τ_c the efficiency rises quickly because of a larger positive inner-friction contribution, and the shorter cycle also raises power. Performance is governed by the combined finite-time dynamics of both unitary strokes rather than either stroke alone.
• Efficiency–power trade-off mitigated: Over finite ranges of stroke durations, efficiency and intrinsic cycle power (⟨P⟩ = ⟨w⟩/τ_cycle with τ_cycle = p₊(τ_c + τ_e) + p₋τ_d) can increase simultaneously. An exception is noted at very small τ_c for τ_e = 0.1τ₀, where power is suppressed because the reduction in extracted work outweighs the shorter cycle duration while thermalization duration stays fixed.
• Adiabatic limit gives zero work: In the adiabatic limit (ξ_c,e, ζ_c,e → 0), the compression–expansion stroke runs quasistatically and results in zero average work output.

Methodology in Plain English

The engine is a single ⁴⁰Ca⁺ ion held at the center of a five-segment linear Paul trap with secular frequencies (ω_x, ω_y, ω_z)/2π = (1.89, 1.85, 0.36) MHz. The working medium is encoded in two Zeeman sub-levels of the ground state, |g⟩ = |4S₁/₂, m_s = −1/2⟩ and |e⟩ = |4S₁/₂, m_s = +1/2⟩, split by ω_s/2π = 26.55 MHz using a 9.485 G magnetic field oriented at 45° to the ẑ axis.

A microwave drive with controllable Rabi frequency and detuning executes the unitary compression–expansion stroke, with splittings fixed at ω_c/2π = 4 kHz and ω_h/2π = 8 kHz and a reference time τ₀ = π/2ω_c. The cold reservoir is engineered with two counter-propagating circularly polarized 397 nm laser beams that couple |g⟩ to |P₂⟩ and |e⟩ to |P₁⟩, giving effective dissipation rates γ_g = Ω₁²/3Γ and γ_e = Ω₂²/3Γ; this produces a steady-state population P_g = γ_e/(γ_g + γ_e) = 0.979(1) and an effective temperature T_c = 0.26 T₀, with T₀ = ℏω_c/k_B.

Because direct fluorescence readout heats the ion, the measurement-conditioned feedback is implemented at the ensemble level: measurement statistics are acquired first, then the ion is recooled and reprepared in the corresponding post-measurement state before the selected feedback branch is applied. State evolution is tracked by quantum-state tomography, and the cycle is repeated until it converges to a stable operating point. Experimental data points are compared throughout with numerical simulations.

Why This Matters

This work supplies the first experimental evidence that finite-time nonadiabatic dynamics — usually treated purely as an efficiency penalty in quantum engines — can be deliberately converted into a performance-enhancing resource in a measurement-powered engine. It also gives a fully resolved energetic accounting for a single-ion information engine, including the conditional reset charge, which is the kind of bookkeeping needed to test information-thermodynamic bounds experimentally.

Potential application areas (directions rather than demonstrated uses in this paper):

• Benchmarking quantum thermodynamic hardware: providing an experimentally validated testbed for fluctuation theorems, work quasiprobabilities, and coherence-transition interference in real devices.
• Quantum energy storage: the authors suggest combining measurement-feedback control with a quantum load or quantum battery to connect information-to-work conversion with quantum energy storage.
• Autonomous nanoscale engines: implementing real-time or autonomous feedback, where the measurement record is processed and reset within the same physical device, would bring the system closer to a complete Maxwell-demon cycle.
• Device-level design guidance: the finding that efficiency and power can rise together over finite stroke-duration ranges is directly relevant to engineering fast-operating cycles rather than quasi-static ones.

Industry relevance: The near-term relevance is primarily to quantum-technology R&D — trapped-ion and other quantum-computing platforms, quantum sensing and metrology, and the emerging quantum-batteries field — where understanding and controlling heat, coherence, and irreversibility at the level of individual quantum systems matters for device performance. The paper does not report applied or industry-partnered demonstrations.

Future Directions

1. Real-time and autonomous feedback: processing and resetting the measurement record inside the same physical device, rather than the current ensemble-level protocol that requires recooling and repreparation between measurement and feedback.
2. Full theoretical treatment of the feedback architecture: the companion theoretical work [20] develops the symmetric-FCS work quasiprobability, coherence-transition interference, correlated finite-cycle and fixed-time work statistics, and an entropy-rate formulation of the reset cost — extending these experimentally is an open route.
3. Coupling to a quantum load or quantum battery, linking information-to-work conversion with quantum energy storage.
4. Resource-resolved optimization: combining coherence and finite-time nonadiabaticity analyses toward engines with controllable efficiency, power, and irreversibility — noting that the paper does not report which specific parameter regions would be optimal, as this is left as an open direction.

Target Audience

Researchers and graduate students in quantum thermodynamics, quantum information, and atomic physics; experimentalists working with trapped ions or other quantum platforms interested in thermodynamic control; and theorists studying Maxwell-demon architectures, measurement backaction, and coherence as a thermodynamic resource. Readers seeking a beginner-level introduction will need background in two-level systems and thermodynamic accounting, but the paper's experimental narrative is accessible to physicists outside the immediate subfield.

Authors’ abstract

Harnessing finite-time nonadiabatic transitions that are conventionally associated with quantum inner friction for useful work extraction remains an open experimental challenge. Here we address this issue by introducing and experimentally realizing an innovative measurement-feedback quantum information engine model in the trapped 40Ca+ ion system, in which the projective measurement replaces the hot reservoir as a nonthermal energy source and the feedback control conditionally steers the system through either unitary compression-expansion strokes or thermalization. We experimentally show that the repeated feedback cycle converges to a stable operating regime with a resolved energetic balance, and, by controlling the measurement angle and stroke duration, measurement-induced coherence and the finite-time nonadiabatic contribution can enhance work extraction and raise the efficiency above the corresponding Otto benchmark. The experimental results further show that, over finite ranges of stroke durations, the efficiency and intrinsic cycle power can increase simultaneously. Our experiment establishes a route toward information-to-work quantum engines that convert finite-time irreversibility into performance-enhancing resources.