{"id":"f5c0b990-9542-49c1-94b0-e8557bbccdea","arxiv_id":"2411.18167","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A tiny transverse field triggers fast, temperature-independent quantum annealing in the frustrated Ising magnet α-CoV2O6, unlike the frozen metastable state in zero field.","lead":"A frustrated magnet called α-CoV2O6 stays frozen in a disordered state for over 15 hours below 1 K, but relaxes in about 10 seconds when a tiny transverse magnetic field is applied. The result suggests that real, clean magnetic crystals can perform quantum annealing, a faster-than-thermal way to find the lowest-energy state.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim hinges on the single-ion estimate that a 2 T transverse field creates only Γ≈3.5 mK of non-commuting field; this mapping is not directly measured, and a modest g_x or a different CEF parametrization would put Γ in the 10–100 mK range, changing the quantitative claim.","rationale":"The raw experimental data—frozen at H_x=0 for 15 h and rapid, nearly temperature-independent relaxation in a transverse field—are compelling and would be hard to explain by simple thermal activation. The heat-transport anisotropy adds a useful consistency check, and the paper honestly discloses that QMC at Γ<0.02 K is computationally inaccessible and that defects may play a role. My concern is not that the experiment is wrong, but that the quantitative mapping from applied field to Γ=3.5 mK is the least secure link in the argument. If g_x is even 0.01, the effective transverse coupling is several times larger than claimed; if g_x is exactly zero, the splitting depends on higher-order single-ion physics that is not independently verified. The reader's conditional verdict is therefore appropriate, and no change to that verdict is needed.","tokens_in":16440,"tokens_out":18240,"duration_ms":186063,"concrete_test":"Measure the low-temperature torque or transverse magnetization M_x(H_x) at fixed H_z=2 T and T≈0.1 K on the same crystals, and extract the effective transverse coupling from dM_x/dH_x. Compare with the single-ion prediction underlying Γ=3.5 mK. As a cross-check, measure τ⁻¹(H_x) at T=0.5 K over angles θ=0°,15°,30°,45°,60° and test whether the rates collapse when plotted against the calculated Γ(H_x). If the extracted Γ at 2 T exceeds ~10 mK, the headline quantitative claim needs revision; if the rates do not collapse against Γ, the single-ion mapping is not the control parameter.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing link is the calibration of the applied transverse field to Γ in Eq. (1). The main text asserts g_x≈0 and derives Γ≈3.5 mK at μ₀H_x≈2 T from a single-ion Hamiltonian (Supplementary Note 2), but this is a calculation with crystal-field parameters, not a direct measurement. The phrase 'g_x ≈ 0' is not a bound: if g_x is as large as 0.01–0.1, the linear Zeeman term g_x μ_B H_x Sˣ contributes 7–70 mK at H_x=2 T, changing the 'tiny field' by one to two orders of magnitude. The T_N(H_x) data in Fig. 1c can fit the calculated Γ, but they do not uniquely determine it, and the QMC simulations are run only at Γ≥0.02 K—the paper explicitly states that Γ<0.02 K is computationally inaccessible. Thus the microscopic model is never tested at the claimed experimental Γ. The temperature-independent relaxation rate is a genuine quantum signature, but the quantitative claim 'Γ≈3.5 mK triggers QA' is not as secure as the experimental observation of fast, temperature-independent relaxation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magnetization and heat transport measurements on single crystals of α-CoV2O6, an Ising-like frustrated magnet. The authors find that below ~1 K in zero transverse field the longitudinal magnetization remains frozen in a metastable state for at least 15 hours, whereas applying a transverse field of about 2 T (claimed to create an effective transverse field Γ≈3.5 mK) yields relaxation toward the 1/3-plateau value with a nearly temperature-independent time constant τ~10 s. The temperature-independent relaxation is interpreted as quantum annealing driven by a tiny non-commuting field, and the authors support this with SSE-QMC simulations of the fitted spin Hamiltonian at Γ≥0.02 K. The paper argues this is the first observation of many-body quantum annealing in a structurally clean frustrated magnet.","tokens_in":16696,"tokens_out":5277,"duration_ms":43320,"significance":"The experimental observation of a drastic, nearly temperature-independent speed-up of relaxation under a transverse field, contrasted with a diverging Arrhenius relaxation time at zero field, is a significant and potentially influential result. If the interpretation holds, it extends quantum annealing phenomena beyond the disordered LiHoxY1-xF4 system to a structurally clean frustrated magnet and provides a concrete material platform for further studies. The paper is commendable for including extensive experimental data, openly describing limitations of the numerical simulations, and making explicit statements about the incompleteness of annealing and the possible role of defects. The central experimental result is robust and warrants publication, but the quantitative calibration of the effective transverse field is not fully established, and the numerical support is qualitative rather than at the claimed experimental parameter.","major_comments":[{"comment":"The calibration of the applied transverse field to the effective Γ is the load-bearing link of the paper. The main text states gx∼gy∼0 and Γ≈3.5 mK at μ0Hx≈2 T based on a single-ion Hamiltonian in Supplementary Note 2, but gx≈0 is not presented as a measured bound. If gx were as large as 0.01–0.1, the linear Zeeman term gx μB Hx Sx would contribute 7–70 mK, changing the claimed 'tiny field' by one to two orders of magnitude. The agreement of TN(Hx) with the calculated Γ (Fig. 1c) does not uniquely determine Γ, and the authors should either provide an experimental bound on gx or quantify the uncertainty in the single-ion mapping. Without this, the central quantitative claim that Γ≈3.5 mK triggers quantum annealing is not secure.","section":"Spin Hamiltonian, Eq. (1) and Fig. 1c"},{"comment":"The numerical simulations are carried out only at Γ≥0.02 K, whereas the experimental claim is made for Γ≈3.5 mK, an order of magnitude smaller. The paper explicitly acknowledges that 'observing clear QA effects at 0 < Γ < 0.02 K remains extremely challenging' and that the numerical results 'only seek to provide a qualitative interpretation.' Consequently, the microscopic model is never directly tested at the experimentally claimed parameter. The qualitative support is welcome, but the extrapolation over a factor of roughly six in Γ is not demonstrated. The authors should either extend the simulations to lower Γ using more efficient or specialized methods, or clearly state that the quantitative relation between Γ and the relaxation process is not directly verified.","section":"Quantum Monte Carlo simulations, Fig. 4"},{"comment":"The authors state that the microscopic model 'may lack high precision in simulating the slow spin dynamics observed in α-CoV2O6, especially at smaller transverse fields (0 < Γ < 0.02 K).' Combined with the fact that the exchange parameters J0–J3 are fitted to quasi-equilibrium thermodynamic data of the same material, this limits the predictive value of the simulations for the out-of-equilibrium annealing dynamics. The experimental observation stands on its own, but the interpretation as quantum annealing with a specific Γ would be strengthened by an independent constraint on Γ and by an explicit discussion of whether classical mechanisms (e.g., a nonzero gx) could reproduce the temperature-independent relaxation.","section":"Quantum Monte Carlo simulations / Discussion"}],"minor_comments":[{"comment":"The statement 'gx ∼ gy ∼ 0' should be replaced by an explicit upper bound (e.g., |gx| < 0.01 or a similar measured limit) with a reference to the underlying determination, so that readers can assess the error in Γ.","section":"Spin Hamiltonian"},{"comment":"The temperature range and goodness-of-fit for the constant red-line fit to τ^{-1} at low temperatures are not stated in the main text; please add these details.","section":"Fig. 3b"},{"comment":"There is a typo in the Results section: 'the reduce of annealing timeτ' should read 'the reduction of annealing time τ'; also, 'slight derivations from the T^3 law' should read 'slight deviations'.","section":"Results"},{"comment":"The term 'inner gap' for Γ is unconventional; consider using 'tunnel splitting' or 'transverse-field-induced splitting' to avoid confusion with the crystal-field gap E3−E1.","section":"Fig. 1b and related text"},{"comment":"The molecular-field estimate of W↓/W↑ is a single-site approximation; the text should clarify that this is not a many-body result and that the SSE-QMC results are the relevant many-body evidence.","section":"Fig. 4c"},{"comment":"The paper should state explicitly the measurement uncertainty in the transverse-field angle θ and how it propagates to Hx and Hz, since the claimed Γ≈3.5 mK is sensitive to small misalignments.","section":"Methods"}],"recommendation":"major_revision","confidential_remarks":"The manuscript presents a striking experimental observation that is likely to attract broad interest. The main concern for acceptance is the uncalibrated mapping from applied field to effective Γ and the order-of-magnitude gap between the experimental Γ and the simulations. I believe these are fixable through a revised manuscript that clearly bounds gx, discusses the uncertainty in Γ, and tempers the claims about quantitative agreement. The experimental data and the discussion of limitations are of high quality. I encourage the editor to send the paper back for a major revision rather than reject it."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is the first clean many-body magnet to show quantum annealing, and the experimental signature—temperature-independent relaxation in a transverse field—looks real. The weak link is the calibration of that transverse field to an effective Γ: it's a single-ion calculation, not a direct measurement, and the QMC is never run at the claimed Γ.\n\nWhat's new: Brooke et al. established QA in the disordered LiHoxY1-xF4; this paper extends it to a structurally clean frustrated Ising magnet, α-CoV2O6. The main evidence is convincing: below 1 K, the longitudinally polarized initial state stays frozen for 15+ hours at zero transverse field, but relaxes with τ ~10 s when a 2 T transverse field is applied, with a temperature-independent rate that rules out classical activation. The heat transport data give an independent handle, showing similar quantum tunneling of domain walls. The framing in terms of KT topological defects is speculative but a reasonable and potentially useful way to think about the metastability.\n\nThe soft spots, in order of seriousness:\n\n1. The mapping from μ0H_x ≈ 2 T to Γ ≈ 3.5 mK is not directly measured. It comes from a single-ion Hamiltonian with g_x ≈ 0. If g_x were as small as 0.01, the linear Zeeman term would already contribute ~13 mK at 2 T, and the whole 'tiny field' narrative shifts. The TN vs H data in Fig. 1c fit the Γ(H) curve, but they don't uniquely pin down Γ, especially at the low-field end. This needs error bars or a bound on g_x.\n\n2. The QMC simulations at Γ ≥ 0.02 K are an order of magnitude above the experimental Γ ≈ 3.5 mK. The paper acknowledges this, but then says the simulations 'unequivocally demonstrate' QA—a bit strong for a result that is 'qualitative' and not at the literal parameter value.\n\n3. The model parameters J0-J3 are fitted to the same compound's thermodynamics, so the QMC 'support' is partly self-consistent rather than independent. This doesn't invalidate the experimental discovery, but it does mean the theory isn't a separate confirmation.\n\nBottom line: the raw data are strong and the physics is interesting. This is a paper for anyone working on quantum annealing in real materials, frustrated magnetism, or slow dynamics in Ising systems. A serious referee should engage with it, but the authors need to tighten the Γ calibration and tone down the QMC language. I'd cite it and bring it to the reading group.","headline":"First clean magnet showing quantum annealing; the experimental relaxation data are convincing, but the calibration of the tiny transverse field is a calculation, and the QMC sits above the claimed value.","tokens_in":17303,"tokens_out":6179,"would_cite":true,"duration_ms":52964,"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":"A tiny transverse field of about 3.5 mK drives a clean frustrated magnet to near its ground state in seconds, where thermal relaxation would take more than 15 hours.","keywords":["quantum annealing","frustrated magnet","Ising model","transverse field","Kosterlitz-Thouless phase","alpha-CoV2O6","quantum Monte Carlo","spin relaxation"],"falsifier":"Measure the Co$^{2+}$ ground Kramers-doublet splitting of $\\alpha$-CoV$_2$O$_6$ at a transverse field of $\\sim 2$ T with high-field ESR or inelastic neutron scattering: if the measured $\\Gamma$ is not close to $\\sim 3.5$ mK, or if the transverse $g$-factor proves non-negligible, the claimed quantum annealing could instead be classical relaxation.","tokens_in":16168,"feed_emoji":"🧲","tokens_out":14880,"duration_ms":117720,"temperature":0.7,"pith_summary":"This paper claims that the clean frustrated magnet $\\alpha$-CoV$_2$O$_6$ exhibits real many-body quantum annealing. Below $\\sim 1$ K and in zero transverse field, the Ising spin system stays trapped in metastable Kosterlitz-Thouless phases for at least 15 hours; applying a tiny effective transverse field $\\Gamma \\approx 3.5$ mK makes it relax toward the lowest-energy state in about 10 seconds, with a nearly temperature-independent relaxation time. The authors argue this is quantum annealing triggered by time-reversal symmetry breaking across the Co$^{2+}$ ground Kramers doublet, not thermal activation, and they reproduce the speed-up qualitatively with stochastic series expansion quantum Monte Carlo simulations of the fitted spin Hamiltonian. If right, it would be the first demonstration of many-body quantum annealing in a structurally clean magnetic material, extending the phenomenon beyond the disordered spin glass LiHo$_x$Y$_{1-x}$F$_4$.","feed_headline":"Tiny 3.5-mK field triggers quantum annealing in a frustrated magnet","feed_subtitle":"Below 1 K the spin system stays frozen for 15 hours without the field; with it, relaxation takes about 10 seconds.","key_machinery":"The machinery is the transverse-field Ising Hamiltonian of $\\alpha$-CoV$_2$O$_6$, $$H = J_0 \\sum_{\\langle i,i_0\\rangle} S_i^z S_{i_0}^z + J_1 \\sum_{\\langle i,i_1\\rangle} S_i^z S_{i_1}^z + J_2 \\sum_{\\langle i,i_2\\rangle} S_i^z S_{i_2}^z + J_3 \\sum_{\\langle i,i_3\\rangle} S_i^z S_{i_3}^z - \\mu_0 \\mu_B H_z g_z \\sum_i S_i^z - \\Gamma \\sum_i S_i^x,$$ whose non-commuting term $\\Gamma \\sum_i S_i^x$ is the quantum agent. It converts the classical Ising dynamics into a tunnelling problem, and the metastable Kosterlitz-Thouless phases with vortices and antivortices around domain walls are what the tunnelling must escape. The fitted couplings and the single-ion relation between applied transverse field and $\\Gamma$ give the model predictive power for both equilibrium and annealing dynamics.","core_discovery":"On its own terms, the paper's central discovery is that a transverse field of only $\\Gamma \\approx 3.5$ mK can anneal a frustrated Ising magnet that would otherwise be frozen: in zero transverse field the longitudinal magnetization shows no approach to the $g_z/6$ plateau even after 15 hours below 1 K, while at the same temperatures a $\\sim 2$ T transverse field (whose effective spin-1/2 action is a 3.5 mK transverse term) drives relaxation with $\\tau \\approx 10$ s. The authors trace this to the transverse-field term $H_{\\rm TF} = -\\Gamma \\sum_i S_i^x$, which breaks time-reversal symmetry, splits the Kramers doublet, and lets individual Co$^{2+}$ spins tunnel between $S_i^z = \\pm 1/2$. They fit a spatially anisotropic triangular-lattice Ising Hamiltonian with couplings $J_0 = -30.73$ K, $J_1 = 3.60$ K, $J_2 = 14.21$ K, $J_3 = 2.55$ K, verify its zero-field phases against neutron diffraction, and show with SSE-QMC that tiny transverse fields (0.02--0.1 K) substantially accelerate relaxation at $T \\approx 1$ K while changing equilibrium properties negligibly. The conclusion is that $\\alpha$-CoV$_2$O$_6$ provides the first example of many-body quantum annealing in a clean frustrated magnet, with the metastable KT phases and their topological defects as the barriers that the tunnelling overcomes.","pith_inferences":["Beyond the paper: a direct measurement of the Co$^{2+}$ Kramers-doublet splitting at a transverse field of about 2 T, for example by high-field ESR or inelastic neutron scattering, would independently verify the 3.5 mK scale and separate the quantum interpretation from a classical-field artifact.","Beyond the paper: because the zero-field magnet sits in KT-like metastable states, $\\alpha$-CoV$_2$O$_6$ could serve as a physical testbed for the out-of-equilibrium topological-defect dynamics studied in programmable quantum annealers, bridging real-materials and simulator results.","Beyond the paper: the incomplete annealing after hours suggests that protocol design, such as field cycling or a temperature quench before applying the transverse field, might push the system closer to the exact ground state; that is a testable prediction about the role of history in annealing efficiency.","Beyond the paper: a systematic comparison of the ratio $\\Gamma/\\Delta E$ across clean frustrated Ising magnets could identify a practical rule for which compounds show observable many-body quantum annealing at achievable transverse fields."],"forward_implications":["A transverse field of a few millikelvin bypasses a roughly 15 K thermal barrier, implying that quantum annealing can outperform thermal annealing by many orders of magnitude in a clean magnet.","The relaxation rate at $\\Gamma \\approx 3.5$ mK is nearly independent of temperature below 1 K, in contrast to the Arrhenius law at zero transverse field, giving a sharp experimental signature that separates quantum from classical relaxation.","The same fitted Hamiltonian, with couplings checked against neutron diffraction and thermodynamic data, can be used to predict how other frustrated Ising magnets respond to transverse fields.","Heat transport along the frustrated plane is measurably suppressed by transverse fields below 1 K, providing an independent phonon-scattering probe of the domain walls and topological defects released by quantum tunnelling.","Even with quantum annealing, the magnetization does not fully reach the exact $g_z/6$ ground-state plateau, showing that residual topological defects or dilute disorder limit complete annealing in a real material."],"supporting_citations":[{"why":"supplies the prior demonstration of quantum annealing in a disordered magnet that this clean-material claim is built against","marker":"[9]"},{"why":"provides the only previous many-body QA material and the barrier-energy and classical-quantum crossover benchmarks","marker":"[16]"},{"why":"shows the structurally clean frustrated magnet Ca3Co2O6 shows no measurable QA, establishing the contrast for this first clean example","marker":"[19]"},{"why":"determines the ground-state magnetic configurations by neutron diffraction, defining the annealing targets","marker":"[20]"},{"why":"gives the density-functional calculation of exchange couplings that justifies truncating the Hamiltonian at fourth-nearest neighbors","marker":"[21]"},{"why":"supplies the single-crystal growth and characterization underlying the sample quality","marker":"[22]"},{"why":"provides the linear TN-versus-Gamma relation used to calibrate the single-ion mapping from applied transverse field to effective transverse coupling","marker":"[26]"},{"why":"introduces the stochastic series expansion quantum Monte Carlo method used for the annealing simulations","marker":"[28]"}],"fun_headline_variants":["3.5-mK field wakes frozen spin system in 10 seconds","Quantum annealing from a 3.5-mK transverse field","Frustrated magnet unthaws with millikelvin quantum field","Tiny field turns 15-hour freeze into 10-second annealing","Millikelvin transverse field triggers quantum annealing"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire quantum-annealing interpretation rests on the single-ion mapping that a $\\sim 2$ T transverse laboratory field produces only $\\Gamma \\approx 3.5$ mK of effective transverse coupling in the spin-1/2 Ising model; if the actual transverse $g$-factor or the crystal-field parameters are larger than assumed, the observed fast relaxation could be classical.","fun_headline_variants_meta":{"raw":{"variants":["3.5-mK field wakes frozen spin system in 10 seconds","Quantum annealing from a 3.5-mK transverse field","Frustrated magnet unthaws with millikelvin quantum field","Tiny field turns 15-hour freeze into 10-second annealing","Millikelvin transverse field triggers quantum annealing"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000233,"raw_usage":{"total_tokens":1553,"prompt_tokens":1065,"completion_tokens":488,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":681,"completion_tokens_details":{"reasoning_tokens":402}},"tokens_in":681,"tokens_out":488,"duration_ms":4843,"temperature":1.0,"reasoning_tokens":402,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:27:01.408765+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the Co$^{2+}$ ground Kramers-doublet splitting of $\\alpha$-CoV$_2$O$_6$ at a transverse field of $\\sim 2$ T with high-field ESR or inelastic neutron scattering: if the measured $\\Gamma$ is not close to $\\sim 3.5$ mK, or if the transverse $g$-factor proves non-negligible, the claimed quantum annealing could instead be classical relaxation.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the only previous many-body QA material and the barrier-energy and classical-quantum crossover benchmarks"},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the linear TN-versus-Gamma relation used to calibrate the single-ion mapping from applied transverse field to effective transverse coupling"}],"review_version":1}