{"id":"c813c342-15be-42f7-a845-92d8542a41ee","arxiv_id":"2412.20194","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"On an NMR processor, a shortcut-to-adiabaticity-driven quantum Otto engine delivers more power at shorter drive times than a non-adiabatic engine, once the counter-adiabatic driving cost is included.","lead":"An NMR experiment shows that a quantum Otto engine driven with counter-adiabatic shortcuts to adiabaticity outperforms the same engine without shortcuts, even after the energy cost of the shortcut is included. The paper also compares two ways of counting that cost and argues which one fits the NMR platform.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central STA-vs-NA comparison uses the ideal counter-adiabatic cost (Eq. 12), not the energy actually delivered by the GRAPE pulses; without measuring the real pulse power, the claimed 'after-spending-extra-energy' advantage is not established.","rationale":"The reader's weakest assumption is identical to the load-bearing concern I identify: the STA cost is theoretical, not measured. Since the experimental claim is explicitly about performance 'even after spending an extra amount of energy' (Sec. IV), the energy accounting must use the actual pulse energy for the comparison to be an experimental result. The concern is correctable—reprocessing with measured pulse powers—so the paper should not be rejected outright, but the quantitative conclusions in Figs. 4–5 and the choice of efficiency metric cannot be accepted as-is. Additional internal inconsistencies (e.g., the sign of the cost in Eq. 10 and the text's contradictory references to Eq. 9 vs. Eq. 10) reinforce the need for revision but are secondary to the missing energy measurement. I therefore maintain the reader's CONDITIONAL verdict.","tokens_in":10690,"tokens_out":5417,"duration_ms":56870,"concrete_test":"Extract the actual GRAPE pulse waveforms used for U_e and U_c (amplitude and phase vs. time, as generated by the optimization and delivered by the spectrometer). Integrate the instantaneous RF power (e.g., |Ω(t)|^2 or the calibrated pulse amplitude) over the pulse durations to obtain an experimentally measured STA cost; recompute η1_STA, η2_STA, and P_STA in Fig. 4 with this measured cost replacing Eq. 12. If STA no longer dominates NA at τ ≈ 1550/1210 μs, or if the second efficiency definition no longer gives zero efficiency when power is zero, the central claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's key quantitative claim is that, after accounting for the STA cost, the STA engine delivers more power than the NA engine at shorter driving times. But the STA cost in Eqs. (9)-(12) is computed theoretically from the ideal counter-adiabatic Hamiltonian H_CD(t), while the implemented unitaries U_e and U_c are GRAPE-optimized pulses (Sec. III B). No measurement or estimate of the RF power delivered by these pulses is reported. GRAPE pulses are time-discretized control sequences optimized for state/unitary fidelity; their amplitude envelope, discretization, and spectrometer power calibration are not constrained to equal ∫<H_CD>dt. If the true energy expenditure differs, the efficiency and power curves in Fig. 4 shift, the turnover times (1550/1210 μs) move, and the claimed superiority of the STA engine, and the conclusion that Eq. 10 is the appropriate efficiency metric, are not supported by the experimental data as presented. This is a testable resource-accounting gap, not a disagreement with the standard STA framework.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an experimental study of a finite-time quantum Otto heat engine implemented on a two-qubit NMR processor, using counter-adiabatic (shortcuts-to-adiabaticity, STA) driving in a spin-1/2 Landau-Zener model. The authors measure efficiency and output power as functions of the driving time for two hot-reservoir spin temperatures, incorporate the cost of the STA drive through two efficiency definitions (Eqs. 9 and 10), and compare the STA engine with a non-adiabatic (NA) engine. They conclude that the STA engine outperforms the NA engine at shorter driving times and that the second efficiency definition (Eq. 10) is the appropriate descriptor for their platform.","tokens_in":11043,"tokens_out":6406,"duration_ms":59551,"significance":"If the claims are correct, the paper provides a useful experimental demonstration of STA in a quantum heat engine and a concrete comparison of cost-inclusive efficiency metrics, which is of interest for quantum thermodynamics and quantum control communities. The experimental data are independent measurements, the theoretical curves are based on a standard Hamiltonian model, and the state tomography plus pulse fidelities are reported. The main value is experimental rather than conceptual, and the conclusions depend critically on the correct accounting of the STA resource cost and on the consistency of the reported reservoir temperatures.","major_comments":[{"comment":"The reservoir spin temperatures are internally inconsistent across the text and figure captions. Section III B and Fig. 4 state kBTC = 11.94 peV and hot temperatures kBT_H1 = 40.54 peV and kBT_H2 = 53.11 peV, while Section IV and Fig. 5 state kBTC = 1.9 peV and hot temperatures 6.45 peV and 8.45 peV; the two sets are related by a common factor of about 6.28. More seriously, the Section III B combination (cold 11.94 peV with hot 6.45/8.45 peV) violates the engine working condition of Eq. (19), since TH/TC is about 0.54, below νf/νi ≈ 2.69, whereas the Section IV combination satisfies the condition. Because the efficiency and power curves, and the maximum-power times (1550/1210 μs), all depend on the actual temperatures, this inconsistency must be resolved before the experimental claims can be evaluated.","section":"§III B, §IV, Fig. 4 caption, Fig. 5 caption"},{"comment":"There is a dimensional inconsistency in the definition and use of the STA cost. Equation (12) defines ⟨Ḣ_i⟩_{STA} as an integral of ⟨Ḣ_CD(t)⟩ over time, which has units of energy. However, Eqs. (9) and (10) add ⟨Ḣ_i⟩_{STA} τ to energy quantities, giving energy×time in the denominator and numerator, and Eq. (11) subtracts the same product. If ⟨Ḣ_i⟩ is intended as a time-averaged power, the right-hand side of Eq. (12) needs an explicit 1/τ prefactor; if it is intended as an energy, the extra τ factors in Eqs. (9)–(11) should be removed. This is not a typographical nuance: the numerical values of η_STA and P_STA, and therefore the comparison with the NA engine, change with the correct accounting.","section":"Eqs. (9)–(12)"},{"comment":"The STA cost in Eq. (12) and in the efficiency/power formulas is computed theoretically from the ideal counter-adiabatic Hamiltonian H_CD(t), but the actual expansion and compression unitaries are GRAPE-optimized RF pulses whose energy delivery is not measured or calibrated. The paper reports pulse fidelities but no estimate of the actual RF power consumed by these pulses. Since the central claim is that the STA engine performs better 'even after spending an extra amount of energy,' the resource accounting should include the experimentally delivered STA cost, or at least a quantitative justification that the GRAPE pulse energy matches the ideal ∫⟨H_CD⟩dt. As it stands, the claimed advantage and the stated maximum-power times rest on an unverified equivalence between the theoretical control Hamiltonian and the physical pulse power.","section":"§III B, §IV, Fig. 5"},{"comment":"The discussion of which efficiency definition is appropriate contains a contradictory citation and a potential swap. The text says 'if we use the second definition of efficiency (Eq.9)' when Eq. (9) is the first definition, and it also says the second definition makes both efficiency and power zero at low driving times, after first saying the first definition gives positive efficiency at low times. This confusion directly affects the paper's conclusion that Eq. (10) is the appropriate metric, so the statements should be carefully corrected and aligned with the curves in Fig. 4.","section":"§IV"}],"minor_comments":[{"comment":"The relation between the driving time τ (varied from 200 to 2250 μs) and the stated GRAPE pulse duration (600 to 6000 μs) is unclear; the unitary strokes should have duration τ, so the discrepancy should be explained.","section":"§III B"},{"comment":"The maximum-power times for the NA engine are given in the text (2000 and 1500 μs) but are not marked in Fig. 4; adding these lines or explaining their omission would make the comparison easier to assess.","section":"Fig. 4 caption and §IV"},{"comment":"The sentence 'This could be due to the fact that TH2 has lower population difference as compared to TH2' contains a typo; the second TH2 should likely be TH1.","section":"§IV"},{"comment":"The sentence reporting relaxation times is grammatically ambiguous: 'The measured T1 and T2 relaxation times the for 13C1 and 13C2 spins are 28.57 s and 3.28 s and are 1.84 s and 1.25 s, respectively.' It should clearly state which value is T1 and which is T2 for each spin.","section":"§III A"},{"comment":"The stray block of text beginning 'Hz 15N 13C 1H ...' appears in both figure captions and should be removed, as it looks like an unintended insertion from the figure-generation software.","section":"Figs. 4 and 5 captions"}],"recommendation":"major_revision","confidential_remarks":"The temperature inconsistency looks like a unit-conversion slip, but it is load-bearing because the engine condition and all quantitative results depend on the actual reservoir temperatures. The STA-cost measurement gap is the deeper substantive issue: without actual pulse-power accounting, the headline claim about superior performance after spending extra energy is not fully established. If the authors can correct the temperature values, fix the dimensional inconsistency in Eqs. (9)–(12), and provide a resource-accounting justification for using the ideal H_CD cost, the paper could become a solid experimental contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a real experimental implementation of a counter-adiabatic STA quantum Otto engine on NMR, and the qualitative story — STA gives useful power at shorter driving times than the non-adiabatic engine — is plausible and supported by the data *if* the energy-cost accounting is right. But the cost in Eqs. (9)-(11) comes from the theoretical H_CD, not from the energy the GRAPE pulses actually deliver, and no measurement of the real pulse power is reported. That gap leaves the central quantitative claim unproven. The manuscript also has internal temperature inconsistencies (cold value given as 11.94 peV and 1.9 peV; hot values differ by the same factor) and a typo in the efficiency-definition discussion.\n\nWhat's new: putting STA via counter-adiabatic driving into a two-qubit NMR Otto engine and comparing two cost-accounted efficiency metrics on the same platform. Earlier work had theoretical STA-engine analysis and separate NMR engine experiments; this combined them. The experimental work is careful: state tomography, error bars, GRAPE pulses with high fidelity, and the theoretical curves track the data reasonably well.\n\nSoft spots, in order:\n1. Resource-accounting gap. Eq. (12) computes the STA cost from the ideal counter-adiabatic Hamiltonian, but the implemented unitaries are GRAPE pulses optimized for fidelity, not for matching that energy. The efficiency and power curves in Fig. 4 therefore rely on an assumption. The claim \"even after spending an extra amount of energy\" is not validated for the real pulses. Fixable by measuring or bounding the RF power, but as written it is load-bearing.\n2. Temperature inconsistencies. Sec. III B and Sec. IV give different cold temperatures (11.94 vs 1.9 peV), and the Fig. 4 caption quotes hot temperatures roughly six times larger than those in the text. The reader cannot tell which is correct.\n3. The discussion of the two efficiency definitions is muddled, with Eq. 9 called the \"second definition\" when it is the first. The argument that one metric is more appropriate rests on a consistency condition, which is fine, but the presentation needs cleaning.\n\nThe physics is standard and the qualitative conclusion probably survives a corrected cost accounting. As written, the numbers are not trustworthy. Send it to peer review — it deserves referee time — but referees should demand the actual pulse-energy accounting and consistent temperatures.","headline":"A plausible and useful NMR demonstration of an STA quantum Otto engine, but the key quantitative claim rests on an unmeasured theoretical energy cost and the temperature values are internally inconsistent.","tokens_in":11453,"tokens_out":3638,"would_cite":true,"duration_ms":37288,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"An NMR experiment demonstrates that driving a quantum Otto heat engine with counter-adiabatic shortcuts to adiabaticity yields more output power at shorter driving times than the non-adiabatic engine, even after subtracting the energy…","keywords":["quantum Otto heat engine","shortcuts to adiabaticity","counter-adiabatic driving","Landau-Zener model","NMR quantum processor","spin-1/2 system","engine efficiency with cost","output power"],"falsifier":"Measure the actual radio-frequency power delivered by the GRAPE pulses that implement $H_0 + H_{\\text{CD}}$ during a stroke and compare the time-integrated power with the theoretical $\\langle \\dot{H}_{\\text{CD}}\\rangle \\tau$. If the measured integrated power disagrees with the theoretical cost by more than the experimental uncertainty, the reported STA efficiency and power curves would shift, and the conclusion that the second efficiency metric is the appropriate one would need to be re-examined.","tokens_in":10491,"feed_emoji":"⚙️","tokens_out":11430,"duration_ms":94356,"temperature":0.7,"pith_summary":"The paper tries to show that the usual finite-time penalty of a quantum Otto heat engine—the trade-off between efficiency and power caused by non-adiabatic transitions—can be sidestepped by counter-adiabatic driving, a shortcut-to-adiabaticity protocol that pins the working spin to its instantaneous energy eigenstates. Using spin-1/2 nuclei on an NMR quantum processor, the authors implement a Landau-Zener Otto cycle with shortcut strokes at several driving times and two hot-bath temperatures, and they compare the shortcut engine with the non-adiabatic engine after charging the shortcut's energy cost against the work output. They report that the shortcut engine delivers more output power at shorter driving times in both temperature settings. They also use the comparison to adjudicate between two published efficiency metrics that include the shortcut cost, concluding that the metric which subtracts the shortcut energy from the numerator is the appropriate one for the NMR platform.","feed_headline":"Counter-adiabatic strokes boost a quantum engine's power","feed_subtitle":"On an NMR processor, the shortcut engine beats the non-adiabatic one at short driving times, even paying the shortcut's energy cost.","key_machinery":"The load-bearing object is the counter-adiabatic (CD) Hamiltonian $H_{\\text{CD}}(t) = -\\frac{b_x \\dot{b}_z(t)}{2[b_x^2 + b_z(t)^2]}\\,\\sigma_y$, which is added to the Landau-Zener Hamiltonian $H_0(t) = b_x\\sigma_x + b_z(t)\\sigma_z$ so that the spin follows the instantaneous eigenstates of $H_0$ in finite time. The profile $b_z(t)$ is a polynomial in $t/\\tau$ that makes the CD term vanish at the start and end of each stroke. The shortcut cost is defined as $\\langle \\dot{H}_i\\rangle_{\\text{STA}} = \\int_0^\\tau \\langle \\dot{H}_{\\text{CD}}(t)\\rangle_{\\text{STA}}\\,dt$ per stroke, and the two candidate efficiency definitions place this cost either in the denominator (Eq. 9) or in the numerator (Eq. 10) of the efficiency expression; the experimental power data are used to decide which placement is physically correct for this platform.","core_discovery":"The paper claims that in a spin-1/2 Landau-Zener quantum Otto engine implemented on an NMR quantum processor, replacing the finite-time unitary strokes with counter-adiabatic shortcuts to adiabaticity improves the engine's performance relative to naively fast (non-adiabatic) driving, and that the improvement survives once the energy consumed by the shortcut is deducted from the work output. The paper further claims that, between the two published ways of incorporating the shortcut cost into efficiency, the metric that subtracts the shortcut energy from the work numerator (Eq. 10) is the one that correctly describes the NMR engine, because the other metric (Eq. 9) reports positive efficiency at driving times where the engine is actually consuming net energy and producing no power.","pith_inferences":["The quantitative efficiency and power values rest on the theoretical shortcut cost; measuring the actual power delivered by the GRAPE pulses would provide a direct check of Eq. (12), and any discrepancy would shift the numerical comparisons without necessarily overturning the qualitative STA-versus-NA ordering.","The choice between the two cost-accounting metrics is likely to be platform-specific: in platforms where the shortcut is implemented by different physical drives (e.g., a second microwave tone rather than the same RF channel), the 'appropriate' efficiency definition may differ from the one chosen here.","For working media larger than a single spin-1/2, the counter-adiabatic term becomes multiqubit and its experimental cost may scale differently with system size, so the demonstrated benefit of shortcut driving may not transfer directly to many-body quantum engines."],"forward_implications":["At short driving times the shortcut engine produces positive output power while the non-adiabatic engine still consumes net energy; the maximum-power stroke time is 1550 µs and 1210 µs for the shortcut engine, versus 2000 µs and 1500 µs for the non-adiabatic engine, at the two hot temperatures studied.","The efficiency of the shortcut engine lies above that of the non-adiabatic engine at all finite driving times, and both merge at the Otto limit (≈0.629) for long cycle times.","The second efficiency definition (subtracting shortcut energy from the numerator) is the appropriate descriptor for NMR engines, since the first definition gives positive efficiency in a regime where no net work is produced.","The shortcut cost for the expansion stroke is higher than for compression, and the compression cost decreases as the hot spin temperature increases; hence the performance of the STA engine depends on stroke direction and bath temperatures."],"supporting_citations":[{"why":"Supplies the exact counter-adiabatic driving term $H_{\\text{CD}}(t)$ used to build the shortcut strokes.","marker":"[24]"},{"why":"Defines the STA cost in Eq. (12) and the boundary-conditioned polynomial profile $b_j(t)$ of Eq. (7) used for the driving fields.","marker":"[33]"},{"why":"Gives the first STA efficiency definition (Eq. 9) that adds the shortcut cost to the heat input.","marker":"[21]"},{"why":"Gives the second STA efficiency definition (Eq. 10) that subtracts the shortcut cost from the work output; the paper adjudicates between the two.","marker":"[35]"},{"why":"Supplies the NMR method for preparing pseudo-spin temperatures and the SWAP-based isochoric thermalization stroke used in the experiment.","marker":"[4]"},{"why":"Supplies the GRAPE optimal-control technique used to implement the unitary strokes and the combined $H_0+H_{\\text{CD}}$ evolution.","marker":"[38]"},{"why":"Supplies the working condition $T_H/T_C > \\nu_f/\\nu_i$ that guarantees positive work, used to choose the spin temperatures and energy gaps.","marker":"[42]"}],"fun_headline_variants":["Counter-adiabatic strokes boost a quantum engine's power","NMR quantum engine beats fast driving with adiabatic shortcut","Shortcut to adiabaticity: quantum engine gains despite cost","Quantum Otto engine: counter-adiabatic strokes win out","Even paying for it, shortcut driving boosts quantum engine"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The cost of the shortcut is taken from the ideal counter-adiabatic Hamiltonian's time-integral (Eq. 12), not from a measurement of the power actually supplied by the GRAPE-optimized radio-frequency pulses.","fun_headline_variants_meta":{"raw":{"variants":["Counter-adiabatic strokes boost a quantum engine's power","NMR quantum engine beats fast driving with adiabatic shortcut","Shortcut to adiabaticity: quantum engine gains despite cost","Quantum Otto engine: counter-adiabatic strokes win out","Even paying for it, shortcut driving boosts quantum engine"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000154,"raw_usage":{"total_tokens":1174,"prompt_tokens":869,"completion_tokens":305,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":485,"completion_tokens_details":{"reasoning_tokens":222}},"tokens_in":485,"tokens_out":305,"duration_ms":3721,"temperature":1.0,"reasoning_tokens":222,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T23:26:59.784823+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the actual radio-frequency power delivered by the GRAPE pulses that implement $H_0 + H_{\\text{CD}}$ during a stroke and compare the time-integrated power with the theoretical $\\langle \\dot{H}_{\\text{CD}}\\rangle \\tau$. If the measured integrated power disagrees with the theoretical cost by more than the experimental uncertainty, the reported STA efficiency and power curves would shift, and the conclusion that the second efficiency metric is the appropriate one would need to be re-examined.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the STA cost in Eq. (12) and the boundary-conditioned polynomial profile $b_j(t)$ of Eq. (7) used for the driving fields."},{"cited_title":"Abah and E","cited_arxiv_id":null,"evidence_quote":"Gives the first STA efficiency definition (Eq. 9) that adds the shortcut cost to the heat input."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the second STA efficiency definition (Eq. 10) that subtracts the shortcut cost from the work output; the paper adjudicates between the two."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the NMR method for preparing pseudo-spin temperatures and the SWAP-based isochoric thermalization stroke used in the experiment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the working condition $T_H/T_C > \\nu_f/\\nu_i$ that guarantees positive work, used to choose the spin temperatures and energy gaps."}],"review_version":1}