{"id":"ec8da583-f4d6-4591-b7ee-20a9639664a6","arxiv_id":"2607.28702","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A measurement-feedback quantum engine in a trapped 40Ca+ ion reaches 59.1% efficiency, above the 50% Otto limit, using measurement-induced coherence and a positive inner-friction contribution.","lead":"In a trapped-ion experiment, researchers ran a quantum information engine in which a measurement replaces the hot reservoir, and the resulting coherence and 'inner friction' are used to push efficiency above the standard Otto limit. The work claims that friction, normally a loss, can become a resource that lets the engine get high efficiency and high power at once.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Super-Otto claim rests on idealized energy bookkeeping: Eq. (3) computes Landauer erasure from the thermalized system entropy rather than the measurement record, and measurement input ΔE_mea is an accounting residual, not a measured external cost.","rationale":"The reader's weakest_assumption already identified the idealized energy accounting, including the non-implemented erasure cost, as the main soft spot. My analysis agrees and sharpens it: Eq. (3) is not just unimplemented but appears conceptually mismatched with Landauer's principle, since it charges the entropy of the thermalized system rather than the information content of the measurement record. This is a load-bearing concern because the reported efficiency exceeds the Otto limit by a modest margin, and the missing or mis-specified erasure/measurement costs could flip the sign of the advantage. However, the reader's verdict of CONDITIONAL already reflects this uncertainty, and the paper's companion manuscript [55] and Supplementary Material [54] might contain a derivation that justifies Eq. (3) in a specific information-theoretic framework. I therefore do not see a reason to change the verdict; the same conditions (provide the supplement, clarify the energy accounting) remain necessary. I say 'partial' rather than 'agree' because I go further than the reader by identifying a specific formal flaw in the erasure expression, while the reader framed it more generally as 'idealized and not physically implemented.'","tokens_in":10226,"tokens_out":11950,"duration_ms":125075,"concrete_test":"Recompute the stable-cycle efficiency in Fig. 3(b) with a physically grounded memory model: append a one-bit memory that records a±, use the standard Landauer cost W_er = β_c^{−1} H(p_+, p_−) (or the entropy of the joint system–memory state after feedback) in place of Eq. (3), and re-evaluate η over the θ range where the paper claims η>η_Otto. If the corrected efficiency no longer exceeds 1−ω_c/ω_h, the super-Otto claim is an artifact of the erasure accounting. As a secondary check, estimate the energy supplied by the 397 nm detection beams and microwave pulses from the reported Rabi frequencies and pulse areas and add it to the denominator.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the MFQIE exceeds the Otto limit and breaks the efficiency–power trade-off depends entirely on the denominator of η. Two of its three input terms are not measured external energy supplies. First, ⟨ΔE_mea⟩=⟨w⟩−⟨q⟩ is an energy-balance identity for the working medium; it is not the energy actually drawn from the microwave source, 397 nm detection lasers, or feedback electronics. Second, Eq. (3) states the Landauer erasure cost as ⟨qer⟩=−p_−β_c^{−1}Tr[ρ_{t1^-} ln ρ_{t1^-}], i.e., the entropy of the conditional thermalized state weighted by p_−. For a binary measurement record, Landauer's bound is β_c^{−1} times the Shannon entropy of the outcome distribution (or the entropy of the memory state), not the von Neumann entropy of one branch. The a+ outcome is also recorded and must be erased, so its contribution is missing. The efficiency advantage is narrow (59.1(1.6)% vs 50%); even a modest correction to the erasure term or inclusion of the actual measurement-laser/microwave energy could erase the claimed advantage. The paper's own text admits the erasure step is 'not implemented as an additional physical stroke', so the reported efficiency is not a demonstrated physical energy balance but an idealized bookkeeping result.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a trapped-ion experiment implementing a two-level measurement-feedback quantum information engine (MFQIE). The cycle consists of a projective measurement in a rotated basis, followed by feedback: outcome a+ leads to a finite-time unitary compression-expansion stroke, outcome a− leads to thermalization with an engineered cold reservoir. Using quantum-state tomography, the authors reconstruct the conditional states and compute extracted work, heat, measurement-energy input, and a Landauer erasure cost, defining efficiency as work divided by the sum of these costs. They find convergence to a stable cycle and report an efficiency of 59.1(1.6)% against an Otto limit of 50%, with a duration-dependent positive 'inner-friction' contribution that they argue enhances work output and can break the efficiency-power trade-off.","tokens_in":10679,"tokens_out":9809,"duration_ms":94471,"significance":"The experiment addresses a genuinely open question: whether nonadiabaticity in a finite-time quantum engine can be exploited as a resource rather than a penalty. Its strengths are the direct state-tomographic characterization, the demonstration of convergence to a stable operating regime, and the presentation of a clear quantitative result (59.1(1.6)%) with simulation agreement. If the thermodynamic accounting convention is accepted, the result would be an important experimental step. However, the accounting is the main weakness: two of the three terms in the efficiency denominator are not measured external energy inputs, and the erasure term follows a nonstandard convention that is not physically implemented. The paper is therefore scientifically credible only with substantial clarification and re-analysis under standard cost conventions.","major_comments":[{"comment":"Eq. (3) evaluates the Landauer erasure cost as -p_- beta_c^{-1} Tr[rho_{t1^-} ln rho_{t1^-}], i.e., from the von Neumann entropy of the thermalizing branch. For a binary measurement record, the standard Landauer cost is determined by the entropy of the measurement record (at least H(p_+,p_-)), not by the entropy of one branch. The a+ outcome is also recorded and must be reset. The text explicitly states that the erasure stroke is not implemented as a physical stroke. Consequently the denominator of eta is an idealized bookkeeping term, and the reported 59.1% > 50% comparison to the Otto limit is not a demonstrated physical energy balance. Please either implement/measure the erasure or recompute with the standard Landauer term and report the sensitivity of the super-Otto claim.","section":"Model, Eq. (3) and efficiency definition"},{"comment":"The measurement input is defined by ⟨Delta E_mea⟩ = ⟨w⟩ − ⟨q⟩, an identity from energy conservation for the working medium. It is not the measured energy drawn from the 397 nm detection lasers, microwave source, or feedback electronics. The phrase 'fully resolved energetic balance' (p. 4) therefore overstates what is measured. The super-Otto conclusion depends on this residual being the true external cost. Please state this limitation explicitly and, ideally, estimate or bound the actual metrological energy cost.","section":"Model, 'energy cost of measurement'"},{"comment":"The abstract claims that quantum inner friction can break the traditional efficiency-power trade-off to synchronously achieve high efficiency and large power. Yet Fig. 4(b) shows power decreasing at small tau_c for tau_e = 0.1 tau_0 while efficiency increases—the conventional trade-off. The authors acknowledge this in one sentence but do not reconcile it with the global claim. Please define the precise parameter domain and provide an efficiency-power characteristic relative to a standard Otto/reference cycle, rather than making a global trade-off-breaking claim.","section":"Fig. 4(b) and abstract"},{"comment":"The paper calls the term p_+ epsilon_0 cosθ (xi_c + xi_e − 2 xi_c xi_e) the quantum inner-friction contribution and states it is always positive. This is an algebraic term in Eq. (1), not the standard nonadiabatic excess work, which is nonnegative and describes dissipated energy. The connection between this term and conventional inner friction needs a derivation and comparison with the standard definition; otherwise the claim that friction is exploited as a resource is a matter of convention rather than a demonstrated physical mechanism.","section":"Eq. (1) and 'inner-friction contribution'"}],"minor_comments":[{"comment":"The main text relies on Supplemental Material [54] for the derivations of Eqs. (1)-(3), but [54] is only cited as a URL; a referee cannot verify the derivation. The central formulas should be self-contained or at least accompanied by a brief derivation in the main text.","section":"General"},{"comment":"The effective temperature T_c = 0.26 T_0 is quoted without defining the relationship between T_0 and the dissipative rates. Define T_0 clearly in the caption or text, and avoid potential confusion with the compression/expansion durations tau_c, tau_e.","section":"Fig. 1(c)"},{"comment":"The companion paper [55] is by the same group and is listed as submitted. Since it is used to support the central interpretation, the main text should either summarize its results or cite it as a preprint with full archival status. The paper also lacks a data availability statement.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The core experimental data appear to be taken and analyzed carefully, and the stable-cycle convergence is a nice result. The main risk is the thermodynamic accounting: if the referee or readers adopt the standard Landauer erasure cost (based on the measurement record rather than the thermalizing branch), the reported 59.1% efficiency may not exceed the Otto limit. The authors should be asked to re-analyze with standard conventions and to be explicit that the efficiency is an idealized theoretical cost, not a wall-plug efficiency. If the conclusion survives such re-analysis, the paper could be a valuable contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis is a real experiment with a questionable headline. The trapped-ion setup is competent: they run the measurement-feedback cycle, tomograph the states, show convergence to a steady cycle, and get good agreement between data and simulation. That part deserves credit. What they have not done is measure the energy input. The denominator of their efficiency is built from two bookkeeping terms: the measurement energy ⟨ΔE_mea⟩, which is defined as a residual from energy conservation, and the Landauer erasure cost, which they explicitly say is not implemented as a physical stroke. Eq. (3) gives the erasure cost as the von Neumann entropy of the thermalized branch weighted by p_−. That is not the standard Landauer cost for resetting a binary measurement record; the standard cost is the Shannon entropy of the outcome distribution (or the memory state), and the a+ branch also has to be erased. If they correct this term, the 59.1% efficiency probably drops. The claimed advantage over the 50% Otto limit is only nine percentage points, so the correction matters.\n\nThe 'break the efficiency-power trade-off' claim is also stronger than the data. In Fig. 4(b), at τ_e = 0.1τ_0 and very small τ_c, the power is actually suppressed. So the trade-off is not broken across the whole parameter range; it's broken in a window. They should say that.\n\nI'm also bothered by how much is in the missing supplement and the companion paper. The main text leans on those for the derivations and the dephased comparison. For a PRL-style claim, that's a problem.\n\nThe citation pattern is concentrated on their own companion papers, but that's not a deep flaw if the theory is actually in those papers. The experimental data itself looks honest.\n\nMy bottom line: this is a serious experimental platform and the paper is worth a referee, but the referee should push hard on the energy accounting. The authors need to either implement the erasure, or at minimum show the efficiency under alternative Landauer costings (Shannon entropy, full memory), and report the actual microwave/laser energy if they want to claim a 'demonstrated' physical engine. As it stands, the super-Otto number is a conditional, idealized accounting result, not a metered one.","headline":"Real trapped-ion experiment, but the super-Otto efficiency is an accounting construct, not a metered one.","tokens_in":11093,"tokens_out":3986,"would_cite":false,"duration_ms":39282,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"By replacing the hot reservoir with a projective measurement and routing feedback through fast unitary strokes, a trapped-ion engine shows that quantum inner friction can push efficiency above the Otto limit and break the efficiency-power t","keywords":["quantum information engine","measurement-feedback control","quantum inner friction","Otto limit","Landauer erasure","trapped ion","quantum thermodynamics","efficiency-power trade-off"],"falsifier":"Equip the measurement and feedback apparatus with energy meters and measure the total power draw over a complete cycle, including the microwave pulses, laser cooling, and any electronic control. If the measured external energy input exceeds the bookkeeping input ⟨ΔE_mea⟩ + ⟨q_er⟩ (with uncertainties), the corrected efficiency drops below the Otto limit, which would falsify the claim of a genuine measurement-powered super-Otto engine.","tokens_in":10133,"feed_emoji":"⚛️","tokens_out":8644,"duration_ms":71059,"temperature":0.7,"pith_summary":"This paper reports an experimental measurement-feedback quantum information engine built from a single trapped calcium ion. Instead of a hot reservoir, the engine uses a projective measurement along a tilted spin axis to inject nonthermal energy, and the measurement outcome determines whether the ion undergoes a rapid compression-expansion stroke or is coupled to a cold bath. The authors show that after a few cycles the engine settles into a stable operating point with a complete energy balance, and that by tuning the measurement angle and the stroke durations, both the coherence the measurement creates and the 'inner friction' of fast non-adiabatic driving become beneficial: the engine's efficiency exceeds the Otto limit, and the inner-friction contribution lets it produce high efficiency and high power simultaneously, breaking the usual efficiency-power trade-off.","feed_headline":"Quantum inner friction boosts engine beyond Otto limit","feed_subtitle":"A trapped-ion measurement-feedback cycle turns finite-time losses into work, beating the efficiency-power trade-off.","key_machinery":"The load-bearing object is the feedback-routed cycle: a projective measurement M(θ) replaces the hot bath, and the outcome selects one of two conditional strokes. The cycle's performance is quantified by two dynamical parameters—the non-adiabatic transition probability ξ (the inner-friction strength) and the coherence-dependent parameter ζ—which enter the work formula ⟨w⟩ = p_+ ε0[2(ζ_c + ζ_e − 2ξ_e ζ_c) + (ξ_c + ξ_e − 2ξ_e ξ_c) cosθ]. The inner-friction contribution ⟨w⟩_fri = p_+ ε0 cosθ(ξ_c + ξ_e − 2ξ_c ξ_e) is the term that turns fast driving into a resource, and the Landauer erasure cost closes the cycle's energy bookkeeping.","core_discovery":"The central discovery is that finite-time irreversibility—normally a loss mechanism—can serve as a work-producing resource in a measurement-feedback engine. In the experiment, the measurement of M(θ) = cosθ σ_z + sinθ σ_x collapses the spin onto a coherent superposition; for a+ outcomes the ion is driven through a time-dependent compression-expansion unitary, while a− outcomes send it toward thermal equilibrium with a cold reservoir. The extracted work in the unitary branch contains an inner-friction term proportional to cosθ(ξ_c + ξ_e − 2 ξ_c ξ_e), which is non-negative for |θ| < π/2, so rapid driving contributes positive work. The engine reaches a stable cycle with efficiency 59.1(1.6)% ve","pith_inferences":["If the energy-accounting assumption is right, the same positive-friction mechanism could be imported into other finite-time quantum machines, such as refrigerators or batteries, where non-adiabatic driving is currently treated purely as a cost.","A testable implication is that the size of the efficiency-power trade-off breaking should scale with how much the measurement basis is tilted relative to the energy basis (θ); scanning θ and measuring efficiency and power would map the anticipated phase boundary.","The paper leaves the measurement apparatus outside the thermodynamic bookkeeping; a natural next test is to build a version where the demon's memory is reset inside the device and measure the total external energy input, which would confirm whether the super-Otto efficiency is fully extractable.","Connecting this engine to a quantum load or battery, as the authors suggest, would test whether the claimed advantage survives when work is stored rather than measured theoretically."],"forward_implications":["If correct, the stable operating regime means a measurement-feedback engine can run continuously from arbitrary initial states without external state preparation.","Efficiency beyond the Otto limit by about 18% (relative) is achieved experimentally, so the protocol offers a concrete route to outperform standard cycle efficiency.","The positive inner-friction contribution implies that faster strokes can enhance both efficiency and power, overturning the assumption that finite-time driving always degrades engine performance.","Tuning the measurement angle and stroke duration gives two independent control knobs for the engine's efficiency and power, which could be used in practical engineering of quantum engines.","The comparison with a fully dephased reference cycle indicates that measurement-induced coherence is a genuine resource, not an artifact of the energy accounting."],"fun_headline_variants":["Friction as fuel: quantum engine beats Otto","Trapped-ion engine exploits quantum friction","Engine breaks efficiency-power trade-off via quantum friction","Measurement-feedback engine turns losses into work"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The cycle's energy ledger assumes that the projective measurement costs exactly the change in the ion's average energy and that erasing the measurement record costs exactly the Landauer minimum—yet the real measurement apparatus and controller are not metered; if their actual energy consumption is higher, the engine's efficiency advantage over the Otto cycle shrinks or disappears.","fun_headline_variants_meta":{"raw":{"variants":["Friction as fuel: quantum engine beats Otto","Trapped-ion engine exploits quantum friction","Engine breaks efficiency-power trade-off via quantum friction","Measurement-feedback engine turns losses into work"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000959,"raw_usage":{"total_tokens":3908,"prompt_tokens":714,"completion_tokens":3194,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":458,"completion_tokens_details":{"reasoning_tokens":3138}},"tokens_in":458,"tokens_out":3194,"duration_ms":22282,"temperature":1.0,"reasoning_tokens":3138,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T00:36:27.198042+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Equip the measurement and feedback apparatus with energy meters and measure the total power draw over a complete cycle, including the microwave pulses, laser cooling, and any electronic control. If the measured external energy input exceeds the bookkeeping input ⟨ΔE_mea⟩ + ⟨q_er⟩ (with uncertainties), the corrected efficiency drops below the Otto limit, which would falsify the claim of a genuine measurement-powered super-Otto engine.","supporting_citations":[],"review_version":1}