{"id":"b81b233b-4793-4361-b9ab-208426c3eb44","arxiv_id":"2412.17774","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"After a thermal quench, visons in the gapless Kitaev spin liquid show diffusive annihilation, attraction-driven annihilation, and freezing into metastable crystals depending on temperature.","lead":"This paper simulates how vison defects in a honeycomb Kitaev spin liquid relax after a sudden temperature drop. It identifies three dynamical regimes, including a frozen state made of metastable vison crystals, which could affect how experiments interpret slow relaxation in Kitaev materials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The kMC transition rates use internal Majorana energy rather than the free energy that should control slow-vison equilibrium, so the simulated barriers and freezing scales are not yet grounded in the Kitaev thermal distribution.","rationale":"The central claim rests entirely on the kMC model; there is no alternative analytic derivation of the three regimes. The reader's concern about time-scale separation is legitimate, but I find a more specific and more decisive internal inconsistency: the rates use internal Majorana energy where the adiabatic reduction requires the Majorana free energy. This is not a matter of disagreeing with the field's consensus; it follows from the authors' own equilibrium assumption. SM S2 provides useful finite-size checks and the parity treatment is thoughtful, but those checks do not address the thermodynamic potential used in the Glauber rates. The proposed F-based re-run is computationally feasible with the same exact-diagonalization machinery and would settle whether the three regimes survive. Because the qualitative picture may still survive, the manuscript should remain conditional rather than be rejected.","tokens_in":25226,"tokens_out":11981,"duration_ms":130025,"concrete_test":"Re-run the kMC with the same protocol but replace every Delta E in the Glauber factor (SM Eq. S13) by Delta F = -T ln[Z_M(C')/Z_M(C)], computing Z_M(C) = Product_k 2 cosh(eps_k/2T) from the same Majorana eigenvalues, with the parity restriction used in SM S1. Compare rho(t) and snapshots at T/|J| = 0.1, 0.01, and 0.001. If the t^-1 ln t, t^-2, and freezing regimes survive quantitatively, the concern is not load-bearing; if the regime boundaries shift or any regime disappears, the central claim needs revision.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Even granting the time-scale separations (tau_M, tau_2 << tau_1) invoked in 'Simulation method', the kMC rates are not the ones dictated by the thermalized Majorana sector. SM Eq. (S6) defines the vison-configuration energy as the internal energy E(C) = -Sum_k (eps_k/2) tanh(eps_k/2T), and SM Eq. (S13) uses the weight exp(-E_k/T). The Markov chain therefore has stationary distribution proportional to exp[-E(C)/T]. However, if the fast Majorana fermions are in equilibrium at each instantaneous C, the correct reduced equilibrium weight for slow flux configurations is the Majorana partition function, P_eq(C) proportional to Z_M(C) = Tr_M exp[-H_M(C)/T] = exp[-F(C)/T], with F(C) = -T ln Z_M(C). The difference is the configuration-dependent Majorana entropy, which is not negligible at the simulated temperatures (up to T = 0.1|J|) since visons modify the Dirac spectrum. In a finite-temperature Born-Oppenheimer treatment the potential of mean force is F, not <H>. Consequently the barriers (e.g., the 0.035|J| barrier quoted for Tfreezing), the scales Tint and Tfreezing, and the frozen-crystal scenario are not yet derived from the Kitaev thermal distribution. The qualitative regimes might survive the correction, but the paper does not show that.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the thermal quench dynamics of visons in the isotropic, gapless Kitaev honeycomb model by means of kinetic Monte Carlo (kMC) simulations. The authors start from a random high-temperature vison configuration and evolve the visons by nearest-neighbor hops with Glauber-type rates computed from the exact diagonalization of the Majorana fermion Hamiltonian for each instantaneous flux configuration, under the assumptions that the Majorana fermions equilibrate rapidly and that vison coherence is destroyed on a fast time scale. Three dynamical regimes are reported: at T/|J| = 0.1, a diffusion-limited annihilation with asymptotic density decay rho ~ t^{-1} ln t; at T/|J| = 0.01, an accelerated 'terminal-velocity limited' annihilation with rho ~ t^{-2} attributed to the long-range attractive vison interaction; and at T/|J| = 0.001, a dynamical freezing accompanied by the formation of metastable sqrt(3)xsqrt(3) vison crystals and slow coarsening of Z_3 super-clusters. The paper also proposes temperature scales T_pair ~ 0.26|J|, T_int ~ 0.04|J|, and T_freezing ~ 0.006|J| separating these regimes.","tokens_in":25505,"tokens_out":11468,"duration_ms":124962,"significance":"If correct, the paper would provide one of the first comprehensive pictures of the out-of-equilibrium annihilation dynamics of fractionalized excitations in an exactly solvable quantum spin liquid, connecting vison dynamics to classical reaction-diffusion universality classes and to a dynamical crystallization mechanism without additional couplings. The simulations are extensive, including exact diagonalization at every Monte Carlo step, multiple system sizes, finite-size extrapolations of interaction barriers, and checks of the parity constraint for physical states. These are genuine strengths. However, the central kMC model is built on an assumed time-scale separation and on transition rates that are not derived from the Kitaev thermal distribution, so the specific exponents and freezing scale are only as reliable as that model. The paper also claims a coarsening behavior that is not quantitatively demonstrated. The subject is timely and the qualitative taxonomy is plausible, but the present version does not fully establish its quantitative conclusions.","major_comments":[{"comment":"The kMC transition rates use the internal Majorana energy E(C,T) = -Sum_k (epsilon_k/2) tanh(epsilon_k/2T) defined in SM Eq. (S6), and SM Eq. (S13) weights each outcome by exp(-E_k/T), with SM Eq. (S11) stating that the Markov chain relaxes to rho_th(C) proportional to exp[-E(C,T)/T]. This stationary distribution is not the thermal distribution of the Kitaev model. In the exact solution, the reduced equilibrium weight of a flux sector is Z_M(C) = Tr_M exp[-H_M(C)/T] = exp[-F(C,T)/T], and the difference F - E = -T S_M(C) is configuration-dependent because visons modify the Majorana spectrum. At T up to 0.1|J| this entropy contribution is not negligible, and at T near 0.006|J| it can be comparable to the 0.035|J| energy barrier quoted for freezing. Consequently the barriers, the scales T_int and T_freezing, and the frozen-crystal scenario are outputs of a classical model with internal-energy rates, not derivations from the Kitaev thermal distribution. Please either replace Delta E by Delta F in the rates with a detailed-balance justification, or derive the rates from a concrete microscopic bath coupling, and then rerun the simulations to test whether the three regimes survive.","section":null},{"comment":"The central quantitative claims are the asymptotic laws rho ~ t^{-1} ln t and rho ~ t^{-2}. These are asserted from visual agreement with the curves in Fig. 1; no fitted exponents, confidence intervals, residuals, or fitting windows are provided. Since d = 2 is the critical dimension, the distinction between t^{-1} ln t and a pure power law is delicate, and the claimed t^{-2} should be tested against, e.g., t^{-1.8} or t^{-2} ln t. Please provide quantitative fits with error bars from the ensemble of runs, state the time window used, and show how the inferred exponents depend on system size.","section":null},{"comment":"The abstract lists a 'hidden coarsening of super-clusters associated with a broken Z3 symmetry' as a main result, but the only evidence is a snapshot in Fig. 4 together with a qualitative statement that the growth is 'rather slow'. No time-dependent domain-size data, correlation-function analysis, or growth-law exponent are given, and the Discussion itself describes this part as preliminary. Please either provide quantitative coarsening data (e.g., mean Z_3 domain size versus time, with system-size dependence) or remove this claim from the abstract and present it as an outlook.","section":null},{"comment":"The reduction to a classical Markov chain rests on the inequalities tau_M << tau_1 and tau_2 << tau_1. The authors state these assumptions but do not estimate them from any microscopic coupling; the only justification offered is the qualitative statement that the gapless Majorana continuum couples efficiently to the bath. If either time-scale separation fails, the diagonal Markov-chain description and the resulting freezing may not describe the actual quantum dynamics. Please provide estimates or a model calculation supporting these inequalities, and clearly state in the abstract that the results are conditional on this adiabatic Markovian regime.","section":null}],"minor_comments":[{"comment":"The manuscript contains numerous OCR-type artifacts and typos, for example 'v isions' in the abstract, 'v ision' in the introduction, 'Ã flux' in the introduction, 'asscioated' in the Fig. 6 caption, and 'Diûusion' in the Fig. 1 caption. A careful proofreading pass is needed before publication.","section":null},{"comment":"The main text quotes the NN/NNN barrier as Delta V ~ 0.035|J|, while SM Sec. S2 defines E_barrier = Delta V(r=2) - Delta V(r=3) and extrapolates it to 0.0403(5)|J|. Please reconcile these values and specify exactly which pair separation defines the barrier used for T_freezing.","section":null},{"comment":"It would greatly improve readability if the expected asymptotic forms t^{-1} ln t and t^{-2} were drawn as dashed lines over the fitting windows, and if the run-to-run spread were shown as shaded bands rather than only averaged curves.","section":null},{"comment":"The finite-size discussion of Dirac-point folding for system sizes not divisible by 3 is careful, but the main text should state explicitly whether the N = 60 x 60 results (which are divisible by 3) are representative of generic system sizes, since the SM shows the same behavior for 3L, 3L+1, and 3L+2.","section":null}],"recommendation":"major_revision","confidential_remarks":"The free-energy issue is the most serious concern: the kMC model's stationary distribution is not the Kitaev distribution, so the simulated regimes may be an artifact of the internal-energy rates. I do not recommend rejection because the simulations are extensive and the qualitative regimes could survive a free-energy correction, but the revision must be substantive. If the authors can show that the entropy correction is negligible or that the internal-energy rates are justified, the paper could be a strong contribution; otherwise the conclusions need to be reframed as properties of a phenomenological model. I would also ask the editor to ensure the final version is free of the many garbled symbols, which currently make parts of the manuscript hard to read."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First things first: this is a real paper, not a toy. Yang and Chern have done extensive kMC simulations of vison relaxation in the gapless Kitaev model and found three distinct post-quench regimes: a high-T diffusion-limited annihilation with the expected t^-1 ln t, an intermediate-T terminal-velocity-limited t^-2 decay, and a low-T freezing accompanied by metastable sqrt(3)x sqrt(3) vison crystals. The three-regime classification and the dynamical formation of these crystals are new; previous quench studies focused on Majorana dynamics or defect production, and vison crystals were only discussed in equilibrium with extra couplings. The structure factor showing K-point peaks and the Z3 coarsening of super-clusters are nice, concrete observations.\n\nThe soft spots are also concrete. The kMC transition rates (SM Eq. S13) use the internal energy E(C) = <-sum_k eps_k/2 tanh(eps_k/2T)> as the weight for vison configurations. But if the Majorana fermions are thermalized at each instantaneous C, the correct reduced weight for slow flux configurations is the Majorana partition function, i.e., exp[-F(C)/T] with F(C) = -T ln Z_M(C). The difference is the configuration-dependent Majorana entropy, which is not small at T = 0.1|J|. As a result, the stationary distribution of the Markov chain is not the Kitaev thermal distribution, and the barriers, Tint, Tfreezing, and the freezing scenario are not yet derived from the actual model. The qualitative regimes might survive the correction, but the paper doesn't show that.\n\nSecond, the power-law identifications are visual. The plots clearly show different behavior at the three temperatures, but there are no fitted exponents with error bars, and the t^-2 claim rests on the terminal-velocity argument plus an eyeball match. That's fixable with a log-log fit over the scaling window. Third, the work would be much easier to trust with released code and data. Finally, the time-scale separations (tau_M, tau_2 << tau_1) are assumed; that's standard for this kind of open-system treatment, but it's worth a sentence on when it would break.\n\nWho's for: people working on nonequilibrium dynamics of spin liquids, fractionalized excitations, and phase ordering. It's a serious candidate for a peer-reviewed journal; the referee should push on the free-energy point above. My recommendation: send it to review, but with the expectation of major revision.","headline":"Solid kMC study of Kitaev vison dynamics with a three-regime picture, but the rates use internal energy rather than free energy, so the quantitative scales need re-examination.","tokens_in":26040,"tokens_out":3986,"would_cite":false,"duration_ms":39240,"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":"The paper claims that after a thermal quench, the gapless Kitaev honeycomb spin liquid relaxes through the diffusion and pair annihilation of visons, with three temperature regimes that end in frozen metastable √3×√3 vison crystals.","keywords":["Kitaev honeycomb model","vison dynamics","thermal quench","kinetic Monte Carlo","diffusion-limited annihilation","Majorana fermions","metastable vison crystals","Z3 sublattice coarsening"],"falsifier":"A concrete test would be to repeat the same quench with a heat-bath coupling that makes the vison decoherence time comparable to the vison hop time; if the intermediate $t^{-2}$ decay and the low-temperature freeze plateau survive that change, the Markov-chain reduction is not the controlling mechanism, while if they disappear the paper's time-scale separation is confirmed.","tokens_in":25006,"feed_emoji":"🌀","tokens_out":12968,"duration_ms":105438,"temperature":0.7,"pith_summary":"This paper claims that the relaxation of the gapless Kitaev honeycomb model after a thermal quench is controlled by the quasi-stochastic diffusion and pair annihilation of visons, the gapped $\\mathbb{Z}_2$ flux excitations of the emergent gauge field. The claim matters because it reduces a strongly fractionalized quantum system to classical annihilation universality classes, and because it predicts a low-temperature freeze that leaves the spin liquid trapped in metastable $\\sqrt{3}\\times\\sqrt{3}$ vison crystals. According to the paper, all of this follows from effective vison dynamics in which the Majorana fermions are assumed to remain in thermal equilibrium with the instantaneous vison configuration.","feed_headline":"Quenched Kitaev visons slow to t^-2 decay, then freeze","feed_subtitle":"Kinetic Monte Carlo maps three relaxation regimes for spin-liquid fluxes, ending in frozen √3 crystals.","key_machinery":"The kinetic Monte Carlo algorithm that simulates a Markov chain over vison configurations, with single-vison nearest-neighbor hops accepted by heat-bath transition probabilities $P(E_k) = e^{-E_k/T}/\\sum_j e^{-E_j/T}$, and the energy $E(C)$ of each configuration computed by exact diagonalization of the quadratic Majorana Hamiltonian for that flux configuration. The argument also uses a parity-removal trick to keep only physical states, and rests on two time-scale separations: Majorana relaxation is fast enough that fermions stay equilibrated, and vison decoherence is fast enough that only diagonal vison probabilities survive.","core_discovery":"The central discovery is a temperature-stratified relaxation law for vison density $\\rho(t)$ after a quench from infinite temperature. For $T \\gtrsim T_{\\rm int} \\approx 0.04|J|$, visons perform nearly unbiased random walks and annihilate with the $A+A\\to\\emptyset$ asymptotic behavior $\\rho \\sim t^{-1}\\ln t$. For $T_{\\rm int} \\gtrsim T \\gtrsim T_{\\rm freezing} \\approx 0.006|J|$, the long-range $1/r$ attraction turns the motion into a terminal-velocity-limited drift and the density falls as $\\rho \\sim t^{-2}$. For $T \\lesssim T_{\\rm freezing}$, the oscillatory short-range interaction traps visons in local minima; the frozen state is a fragmented $\\sqrt{3}\\times\\sqrt{3}$ vison crystal with a broken $\\mathbb{Z}_3$ sublattice symmetry whose large-scale domains coarsen only slowly.","pith_inferences":["Beyond this paper, the same three-regime phenomenology should appear in any spin liquid whose defect interactions combine a $1/r$ attractive tail with short-period oscillations, not just the Kitaev honeycomb model.","If the trapping mechanism is right, the freezing temperature should scale with the nearest-neighbor to next-nearest-neighbor barrier height ($\\Delta V \\sim 0.035|J|$); future simulations could test this scaling directly.","A quantum master-equation calculation that lets the vison decoherence time approach the hop time is the natural next check; if the freeze and the $t^{-2}$ regime persist there, the classical Markov description is not the only cause, and if they vanish the time-scale separation is load-bearing.","The slow $\\mathbb{Z}_3$ coarsening suggests that late-time snapshots have two length scales, frozen crystallite size and slowly growing sublattice domains, which could be separated by the structure factor in future simulations or experiments."],"forward_implications":["Above the interaction scale, vison relaxation belongs to the same universality class as two-dimensional single-species annihilation, so the late-time density decay carries a logarithmic correction to $t^{-1}$.","Between the interaction scale and the freezing scale, the attractive $1/r$ vison interaction produces a terminal-velocity-limited annihilation with an unusually fast $t^{-2}$ decay.","Below the freezing scale, the quenched system does not reach its equilibrium state but arrests at a nonzero vison density through self-generated trapping barriers.","The arrested state is a metastable $\\sqrt{3}\\times\\sqrt{3}$ vison crystal, and the slow coarsening of its $\\mathbb{Z}_3$ sublattice domains is a separate nonequilibrium ordering process.","The three crossover temperatures $T_{\\rm pair}\\sim0.26|J|$, $T_{\\rm int}\\sim0.04|J|$, and $T_{\\rm freezing}\\sim0.006|J|$ organize the full quench phase diagram."],"supporting_citations":[{"why":"Supplies the exactly solved model, defines visons as $\\mathbb{Z}_2$ flux excitations, and gives the pair-annihilation energy and interaction potential used throughout.","marker":"[1]"},{"why":"Establishes the thermodynamic stages of the Kitaev spin liquid and the Kitaev paramagnet, setting the $T_{\\rm pair}$ scale.","marker":"[2]"},{"why":"Provides the spin-ice thermal-quench monopole annihilation analogy that motivates defect-annihilation-controlled relaxation.","marker":"[22]"},{"why":"Shows how effective vison hopping emerges in perturbed Kitaev models, justifying the Markov-chain treatment.","marker":"[23]"},{"why":"Supplies the diffusive particle-antiparticle annihilation framework whose 2D result is the $t^{-1}\\ln t$ asymptotic behavior.","marker":"[33]"},{"why":"Gives the reaction-diffusion equation used for the diffusion-limited analysis and its homogeneous solution.","marker":"[38]"},{"why":"Provides the terminal-velocity-limited annihilation model with long-range interactions that yields $t^{-d}$ decay.","marker":"[42]"},{"why":"Introduces equilibrium vison crystals in extended Kitaev models, the contrast case for the metastable crystals found here.","marker":"[43]"},{"why":"Demonstrates super-cluster coarsening in a correlated-electron model, the analogue used for the hidden $\\mathbb{Z}_3$ domain growth.","marker":"[44]"}],"fun_headline_variants":["Vison quench: diffusion, drift, and freezing regimes","Kitaev visons slow to t^-2 decay, then freeze into crystal","Thermal quench drives visons into frozen √3×√3 crystal","From random walk to t^-2 drift: vison relaxation after quench","Kinetic Monte Carlo reveals three vison relaxation regimes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that in the open quantum system the Majorana fermions equilibrate almost instantly and the visons lose quantum coherence so quickly that the vison state can be represented by a classical probability distribution over flux configurations; if either time scale becomes comparable to the vison hop time, the predicted decay laws and the freeze may not occur.","fun_headline_variants_meta":{"raw":{"variants":["Vison quench: diffusion, drift, and freezing regimes","Kitaev visons slow to t^-2 decay, then freeze into crystal","Thermal quench drives visons into frozen √3×√3 crystal","From random walk to t^-2 drift: vison relaxation after quench","Kinetic Monte Carlo reveals three vison relaxation regimes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1584,"prompt_tokens":911,"completion_tokens":673,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":580}},"tokens_in":527,"tokens_out":673,"duration_ms":6733,"temperature":1.0,"reasoning_tokens":580,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T05:09:04.697200+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete test would be to repeat the same quench with a heat-bath coupling that makes the vison decoherence time comparable to the vison hop time; if the intermediate $t^{-2}$ decay and the low-temperature freeze plateau survive that change, the Markov-chain reduction is not the controlling mechanism, while if they disappear the paper's time-scale separation is confirmed.","supporting_citations":[{"cited_title":"terminal velocity","cited_arxiv_id":null,"evidence_quote":"Supplies the exactly solved model, defines visons as $\\mathbb{Z}_2$ flux excitations, and gives the pair-annihilation energy and interaction potential used throughout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the diffusive particle-antiparticle annihilation framework whose 2D result is the $t^{-1}\\ln t$ asymptotic behavior."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the reaction-diffusion equation used for the diffusion-limited analysis and its homogeneous solution."}],"review_version":1}