{"id":"7960a4ea-a7e2-496a-80c1-dacaf831e2a6","arxiv_id":"2608.08937","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"In twisted MoTe2, optically created magnetic bubbles decay up to 100 times more slowly at integer and fractional Chern insulator fillings, a new nonequilibrium signature of topological order.","lead":"Researchers created a tiny magnetic bubble in a twisted semiconductor and watched it either grow or shrink. Near the switching field, the bubble's lifetime rose sharply at special electron fillings, exposing dynamics that standard measurements miss.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The two-order slowdown at commensurate fillings may be an artifact of comparing decay times at fixed B without accounting for the filling-dependent coercive field.","rationale":"The paper reports a striking raw observation: at fixed field, the spin-polarization decay time near the coercive field is sharply peaked at ν=-1 and ν=-2/3. The controls (no-bubble, seeded vs unseeded, spatial reproducibility) support the bubble interpretation and rule out some trivial preparation artefacts. However, the strongest quantitative claim, the two-order-of-magnitude prolongation and the resulting claim of extreme nonequilibrium sensitivity to filling, rests on comparing decay times at the same absolute magnetic field across fillings whose coercive fields differ. Since near-coercive dynamics are expected to be a steep function of the field relative to B_c, the observed contrast may be entirely explained by the filling dependence of B_c, an equilibrium property already visible in the hysteresis loops. This is more load-bearing than the reader's single-bubble concern: even if the bubble interpretation is correct, the filling-dependence claim would be undermined if the contrast disappears at matched B-B_c. The test I propose uses data already in the paper (Ext. Data Figs. 3 and 4), so it is immediately checkable. If the collapse test fails, the paper's qualitative observation survives but the central claim should be reframed as a measurement of B_c(ν) rather than a new nonequilibrium sensitivity. I therefore keep the conditional verdict but add a specific condition: the authors must demonstrate matched-B-B_c comparisons or a collapse analysis. The reader's weakest assumption about single-bubble versus multi-domain dynamics is a real secondary limitation, but it is explicitly acknowledged by the authors and does not by itself invalidate the raw timescale contrast; the coercive-field normalization issue is nowhere addressed and directly targets the headline claim.","tokens_in":20583,"tokens_out":6716,"duration_ms":64319,"concrete_test":"Using the already-reported hysteresis data in Ext. Data Fig. 3, assign B_c(ν) for ν=-1, -0.98, -1.04, -2/3, -0.71 and -0.62, and re-plot the decay times from Fig. 3a,b and Ext. Data Fig. 4 as functions of B-B_c(ν). If the t1/e curves for different fillings collapse onto a common curve, the two-order filling contrast is an artifact of the fixed-B comparison; if a residual peak at ν=-1 and ν=-2/3 remains at matched B-B_c, the central claim is supported. A complementary check is to repeat the Fig. 3a measurement at B = B_c(ν)+20 mT for each filling and compare the resulting decay times.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that near the coercive field the false-vacuum decay time is prolonged by up to two orders of magnitude at ν=-1 and ν=-2/3. The comparison in Fig. 3a,b is made at fixed field (B=100 mT for the ν=-1 series, B=40 mT for the ν=-2/3 series), while Methods Sec. 5 states that B_c exhibits a pronounced dependence on charge density and that proximity to B_c governs the energy barrier for domain wall formation and spontaneous bubble nucleation. Ext. Data Fig. 3 confirms that B_c varies between ν=-2/3, -0.98, -1.0 and -1.07. If the decay time is a steep function of B-B_c, then a filling with a larger B_c will automatically show a longer decay at the same absolute B, so the peaks at commensurate fillings could simply trace B_c(ν), a static equilibrium property, rather than a genuinely nonequilibrium filling sensitivity. The text specifies that B=100 mT is slightly above B_c for ν=-1 but does not state B-B_c for ν=-0.98 or ν=-1.04; Ext. Data Fig. 4 shows t1/e versus B for ν=-1 and ν=-0.98 but is not used to compare at matched B-B_c. The seeded/unseeded and no-bubble controls do not address this confound. Establishing the claim requires demonstrating that the two-order contrast survives when fields are set to B_c(ν)+δ for each filling, or when decay times are plotted against B-B_c.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports time-resolved optical pump-probe measurements in twisted bilayer MoTe2, using a Laguerre-Gauss pump beam to optically write a 'true vacuum' bubble inside a metastable ('false vacuum') spin-valley polarized state. The authors observe that the bubble collapses or expands depending on its initial size (controlled by LG pulse power) and on the applied magnetic field, and that near the coercive field the spin polarization decay time is prolonged by up to two orders of magnitude at commensurate fillings ν=-1 and ν=-2/3 compared with adjacent incommensurate fillings. Seeded versus unseeded preparation experiments are used to argue for a filling-dependent bulk nucleation contribution. The paper interprets the results in terms of false-vacuum decay, domain-wall pinning, and a filling-dependent free-energy landscape, and supports the interpretation with a pumped Ising model and Hartree-Fock calculations of the domain wall structure.","tokens_in":20812,"tokens_out":3718,"duration_ms":32528,"significance":"If the central claim holds, this work provides a new solid-state platform for studying metastable-state decay with spatial control of a single bubble, and demonstrates that nonequilibrium dynamics in moiré Chern ferromagnets are far more sensitive to filling than standard transport or steady-state optical measurements. The experimental effort is substantial: the optical preparation protocol is novel, and the paper includes important control experiments (no-bubble trace, seeded versus unseeded preparation) and spatial reproducibility across five positions. The theoretical modeling, though simplified and explicitly qualitative, is an appropriate complement to the experiment. The main significance rests on the correctness of the claims about the filling-dependent slowdown near the coercive field and on the single-bubble interpretation.","major_comments":[{"comment":"The central two-order-of-magnitude slowdown at commensurate fillings is established by comparing decay times at fixed absolute fields (B=100 mT for the ν=-1 series and B=40 mT for the ν=-2/3 series), while Methods Sec. 5 states that B_c exhibits a pronounced dependence on charge density and Ext. Data Fig. 3 shows that B_c varies among ν=-2/3, -0.98, -1.0, and -1.07. Since the decay time is expected to be a steep function of B-B_c, the peaks could simply trace the static B_c(ν) curve rather than a genuinely nonequilibrium filling sensitivity. To support the claimed interpretation, the authors should present measurements at matched B_c(ν)+δ for each filling, or plot t1/e as a function of B-B_c(ν) using the hysteresis data, and show that the commensurate-filling enhancement survives this rescaling.","section":"Methods Sec. 5 and Fig. 3a,b"},{"comment":"The authors explicitly note that 'multiple domains or spatially distributed conversion could produce a similar optical signal.' This caveat is load-bearing because the central narrative—critical-radius control in Fig. 2c-e and the seeded-versus-unseeded comparison in Fig. 3d—interprets the spatially averaged optical signal as the expansion or collapse of a single optically written true-vacuum bubble with a well-defined initial radius. The current data do not uniquely determine the single-bubble picture; for instance, the same traces could arise from many small domains or from a spatially distributed conversion. The authors should provide direct spatial imaging of the bubble (e.g., scanning the probe across the LG spot as a function of delay) or, failing that, substantially soften the bubble-specific claims and present the results as domain-conversion dynamics.","section":"Ext. Data Fig. 5"},{"comment":"The characteristic timescales t1/e and t1-1/e are reported without statistical uncertainties. The two-orders-of-magnitude contrast at ν=-1 compared with ν=-0.98 relies on single representative traces; although Ext. Data Fig. 5 shows spatial reproducibility of the peak position, it does not provide repeated-measurement statistics or error bars on the timescales. Please provide at least the point-to-point scatter or an explicit statement of the systematic uncertainty (e.g., from probe-power calibration) to justify the stated precision.","section":"Figs. 3a-c and Ext. Data Fig. 5"}],"minor_comments":[{"comment":"The abstract contains the typo 'vacuumbubble' and should read 'vacuum bubble'.","section":"Abstract"},{"comment":"The text refers to 'panel (d, bottom)' when describing the lower row of Ext. Data Fig. 4; this appears to be a typo for '(b, bottom)' because the figure has only panels (a) and (b).","section":"Ext. Data Fig. 4 caption and text"},{"comment":"The parameters p0 and α are obtained by fitting the data in Fig. 1g and then used to simulate the bubble shapes in Ext. Data Fig. 7. Because the simulation is explicitly qualitative and the parameters are described as rough estimates, this is not a concern for the experimental claims, but the self-cited unpublished Ref. [33] should be identified more clearly as the source of the model.","section":"Methods Sec. 8"},{"comment":"The coercive field is denoted inconsistently as B_c, B_C, and Bc in different places; please unify the notation.","section":"Notation throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is interesting and the experimental effort is substantial. My main concern is the B_c confounding, which I believe is fixable with additional analysis or measurements. I also think the single-bubble assumption needs to be addressed directly. The use of an unpublished self-cited theory reference for the model parameters is acceptable since the simulation plays only a qualitative role."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis paper is worth your time. It introduces a genuinely new capability: deterministic sub-wavelength injection of a true-vacuum bubble in a moiré Chern ferromagnet using a Laguerre-Gauss beam, and it reports a dramatic observation—near the coercive field, the spin-polarization decay time peaks sharply at ν=-1 and ν=-2/3, up to two orders of magnitude longer than neighboring fillings. The controls are solid: no-bubble traces show no decay, seeded vs unseeded dynamics differ meaningfully at intermediate fields, and the effect is spatially reproducible across five positions. The claim that nonequilibrium dynamics is more filling-sensitive than transport or steady-state optical measurements is plausible and, if it holds, important.\n\nThe main soft spot is real. The comparison in Fig. 3a,b is made at fixed external field, while Methods Sec. 5 states that B_c depends strongly on density, and Ext. Data Fig. 3 confirms it. At B=100 mT, ν=-1 is roughly 20 mT above its B_c≈80 mT; ν=-0.98 likely sits further above its lower B_c. Since the decay rate should be steep in B-B_c, the two-order contrast could substantially trace B_c(ν), a static equilibrium property, rather than a distinct dynamical filling sensitivity. The seeded/unseeded controls do not remove this confound. I don't think this kills the paper—the observation is still striking and the bubble-collapse/expansion data at low fields add real dynamical content—but the central claim needs a matched-field analysis: decay times plotted against B-B_c, or fields set to B_c(ν)+δ for each filling. That is the single most important revision.\n\nMinor issues: no error bars or repeated-measurement statistics on the timescales; the bubble-size interpretation relies on the self-cited pumped-Ising model (Ref. 33) with fitted p0 and α, which the authors themselves call rough estimates; the density-recalibration claim is unquantified; no data or code release. The authors' own admission in Ext. Data Fig. 5 that multi-domain conversion could mimic a single-bubble signal is honest but leaves the bubble interpretation less unique than the main text suggests.\n\nThe math and citation pattern look fine. The Hartree-Fock domain-wall calculation is standard and clearly presented. This is a paper for experimentalists in moiré quantum materials and theorists working on metastable dynamics in flat bands. It deserves a serious referee; I would recommend conditional acceptance pending the matched-field comparison and some error-bar/statistics work.\n\nBest.","headline":"Strong experimental technique and a striking filling-dependent decay near B_c, but the two-order contrast may partly trace the filling-dependent coercive field unless matched at B_c+δ.","tokens_in":21447,"tokens_out":3456,"would_cite":true,"duration_ms":32684,"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":"Near the coercive field, optically written bubbles show that metastable magnetic decay in a Chern ferromagnet slows by up to two orders of magnitude at the exact fillings of integer and fractional Chern insulator states.","keywords":["false vacuum decay","Chern ferromagnet","twisted MoTe2","moiré quantum materials","Laguerre-Gauss optical pumping","nonequilibrium dynamics","Chern insulator","optical bubble seeding"],"falsifier":"Spatially resolved imaging of the magnetization inside the probe spot during the decay would settle it: a single bubble boundary sweeping across the spot supports the critical-radius picture, whereas homogeneous dimming or breakup into many small domains would show that the single-bubble reading is not unique. A second decisive check is to sweep the hole density in fine steps around $\\nu=-1$ at $B\\simeq 100$ mT: the single-bubble picture predicts the two-order-of-magnitude slowdown to switch on within a few hundredths of a hole per moiré cell of commensurability, and a much broader crossover would point to a different mechanism.","tokens_in":20333,"feed_emoji":"🧲","tokens_out":17262,"duration_ms":155032,"temperature":0.7,"pith_summary":"This paper reports an experimental method for studying metastable 'false vacuum' decay directly in a solid-state magnet. Using a Laguerre-Gauss laser pulse in a twisted MoTe2 device, the authors write a sub-wavelength bubble of stable magnetization (the 'true vacuum') inside a metastable background, then follow whether the bubble grows or shrinks. The central result is that near the coercive field the spin-polarization decay time lengthens by up to two orders of magnitude at the commensurate fillings of the integer ($\\nu=-1$) and fractional ($\\nu=-2/3$) Chern insulator states compared with slightly doped fillings. The authors argue this makes nonequilibrium dynamics a far sharper probe of the electronic state than transport or steady-state optics, since the same doping change moves the Hall conductance by less than five percent while changing the decay time by two orders of magnitude.","feed_headline":"False-vacuum decay slows 100-fold at commensurate fillings","feed_subtitle":"In a moiré Chern ferromagnet, a laser-written bubble reveals decay dynamics 100× more sensitive than transport or optics.","key_machinery":"The central object is the optically injected true-vacuum bubble: a Laguerre-Gauss (LG) pump pulse, operated deep in the nonlinear saturation regime of optical spin pumping, flips the surrounding holes' spin-valley polarization while the beam's central vortex leaves an island of the original state whose radius is set by the pulse power, conceptually analogous to stimulated-emission-depletion microscopy. The bubble's fate is decided by the competition between domain-wall surface tension, which favors collapse, and the bulk free-energy gain from the magnetic field, which favors expansion, with a critical radius $R_c$ marking the crossover. Around that crossover, a pumped Ising model with Glauber dynamics and fitted intervalley-scattering rates supports the control of bubble shape and size, while self-consistent Hartree-Fock calculations show atomically sharp domain walls with two co-propagating chiral edge states, providing the theoretical account of why wall kinetics and surface tension depend on filling. The decisive experimental lever is the comparison of seeded decay at commensurate versus incommensurate fillings near the coercive field, together with seeded versus unseeded decays, which isolates the bulk-nucleation contribution.","core_discovery":"The discovery is that false-vacuum decay in a Chern ferromagnet can be deterministically seeded and tracked in time, and that its rate is governed by the electronic filling in a regime-dependent way. At fields comparable to the coercive field, the decay time at the integer Chern insulator filling $\\nu=-1$ is roughly one order of magnitude longer than at $\\nu=-0.98$ and two orders longer than at $\\nu=-1.04$, and a similar enhancement appears at the fractional filling $\\nu=-2/3$; the peaks in decay time are sharp enough to serve as a density calibration. At fields well below or well above the coercive field, by contrast, the dynamics are set by domain-wall motion and show only weak filling dependence, which the authors attribute to the domain wall hosting co-propagating chiral edge states that provide local gapless scattering channels. The proposed mechanism for the near-coercive slowdown is that exact commensurability suppresses both bulk nucleation and domain-wall depinning, while doping away from $\\nu=-1$ creates localized compressible regions that soften pinning and lower the surface tension.","pith_inferences":["If the single-bubble reading holds, the same protocol can measure the domain-wall surface tension of a Chern insulator directly: mapping the critical radius $R_c$ across field and filling would give a quantitative tension landscape that the Hartree-Fock calculation only estimates.","The sharp decay-time peak at commensurate filling suggests a general design principle for metastable magnets: any perturbation that adds compressibility or quasiparticles, such as doping, displacement field, temperature, or illumination, should disproportionately hasten metastable decay, a prediction testable in other moiré Chern magnets and in quantum-Hall ferromagnets.","Because the decay time responds so sharply to local filling, the same measurement could serve as a spatially resolved disorder probe: scanning the optically written bubble across the sample near $B_c$ would map nanoscale density variations that equilibrium measurements average over."],"forward_implications":["Near the coercive field, the false-vacuum decay time becomes a sharp quantitative probe of hole density, locating $\\nu=-1$ and $\\nu=-2/3$ with an accuracy that exceeds steady-state optical spectroscopy.","Nonequilibrium dynamics expose electronic-structure features hidden in equilibrium: a doping change of a few percent alters the Hall conductance by less than 5% yet changes the decay time by two orders of magnitude.","Doping away from commensurability accelerates metastable decay by enhancing bulk nucleation and softening domain-wall pinning, so the same disorder landscape pins domains far more effectively at exact integer or fractional filling.","Seeded-versus-unseeded comparisons isolate the bulk-nucleation bottleneck: at $B\\approx B_c$ the seed strongly accelerates decay at $\\nu=-1$ but only weakly at $\\nu=-0.98$, implying spontaneous nucleation is suppressed at commensurate filling.","The three-field regime picture ($B\\ll B_c$, $B\\approx B_c$, $B\\gg B_c$) maps false-vacuum decay onto distinct mechanisms, namely pinned domain-wall motion, nucleation and depinning, and field-driven wall kinetics, each with its own filling dependence."],"supporting_citations":[{"why":"Established optical spin-valley pumping of holes and the attractive-polaron reflectance readout in t-MoTe2 that this paper's pump-probe scheme is built on.","marker":"[18]"},{"why":"Demonstrated diffraction-limited optical switching of integer and fractional Chern insulator states, the capability the Laguerre-Gauss bubble-writing protocol extends.","marker":"[19]"},{"why":"Provided the driven-dissipative dark-state pumping picture and intervalley relaxation rates used to interpret the optical spin orientation.","marker":"[20]"},{"why":"Supplies the nonlinear-saturation principle (stimulated emission depletion) used to argue the LG vortex writes a sub-wavelength true-vacuum bubble.","marker":"[24]"},{"why":"Motivates filling-dependent bubble dynamics by predicting that chiral edge modes contribute to the surface tension of Chern insulator domain walls.","marker":"[25]"},{"why":"Supplies the pumped Glauber Ising model used to simulate and interpret the LG-written bubble's shape and size control.","marker":"[33]"},{"why":"Provides the nucleation-theory framework in which bubble growth, collapse, and the critical radius are interpreted as metastable-state decay.","marker":"[2]"},{"why":"Provides the transport figure (Hall conductance differing by less than 5% between $\\nu=-1$ and $\\nu=-1.04$) against which the two-order-of-magnitude dynamic sensitivity is contrasted.","marker":"[28]"}],"fun_headline_variants":["Chern ferromagnet false vacuum decay slows 100x at commensurate fillings","Laser bubbles show 100x slower decay at Chern insulator fillings","Decay slows 100x at integer and fractional Chern fillings","Bubble dynamics reveal 100x slower false vacuum decay at commensurate fillings"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The interpretation assumes that the measured optical signal is dominated by a single optically written true-vacuum bubble whose initial radius is set by the Laguerre-Gauss pulse; the authors themselves note that multiple domains or spatially distributed conversion could produce a similar optical signal, in which case the inferred critical-radius behavior and the false-vacuum-decay account would not be uniquely determined.","fun_headline_variants_meta":{"raw":{"variants":["Chern ferromagnet false vacuum decay slows 100x at commensurate fillings","Laser bubbles show 100x slower decay at Chern insulator fillings","Decay slows 100x at integer and fractional Chern fillings","Bubble dynamics reveal 100x slower false vacuum decay at commensurate fillings"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000964,"raw_usage":{"total_tokens":4132,"prompt_tokens":1000,"completion_tokens":3132,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":616,"completion_tokens_details":{"reasoning_tokens":3049}},"tokens_in":616,"tokens_out":3132,"duration_ms":22131,"temperature":1.0,"reasoning_tokens":3049,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:20:04.813364+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Spatially resolved imaging of the magnetization inside the probe spot during the decay would settle it: a single bubble boundary sweeping across the spot supports the critical-radius picture, whereas homogeneous dimming or breakup into many small domains would show that the single-bubble reading is not unique. A second decisive check is to sweep the hole density in fine steps around $\\nu=-1$ at $B\\simeq 100$ mT: the single-bubble picture predicts the two-order-of-magnitude slowdown to switch on within a few hundredths of a hole per moiré cell of commensurability, and a much broader crossover would point to a different mechanism.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrated diffraction-limited optical switching of integer and fractional Chern insulator states, the capability the Laguerre-Gauss bubble-writing protocol extends."},{"cited_title":"Humar, J.-Y","cited_arxiv_id":null,"evidence_quote":"Provided the driven-dissipative dark-state pumping picture and intervalley relaxation rates used to interpret the optical spin orientation."},{"cited_title":"The system was operated at a base temperature of approximately 17 mK, yielding effective electronic tem- peratures in the range of 100–200 mK [32]","cited_arxiv_id":null,"evidence_quote":"Provides the nucleation-theory framework in which bubble growth, collapse, and the critical radius are interpreted as metastable-state decay."},{"cited_title":"Holtzmann, W","cited_arxiv_id":null,"evidence_quote":"Provides the transport figure (Hall conductance differing by less than 5% between $\\nu=-1$ and $\\nu=-1.04$) against which the two-order-of-magnitude dynamic sensitivity is contrasted."}],"review_version":1}