{"id":"fcb4185f-b53f-4b29-85e1-592a286c6480","arxiv_id":"2412.09210","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Observation of a quantized 1/3 fractional quantum Hall plateau in balanced large-angle twisted bilayer graphene, identified as the interlayer coherent (333) state.","lead":"Large angle twisted bilayer graphene shows a quantized Hall plateau at one third filling when both layers have equal density, a new fractional quantum Hall state. The authors interpret it as an excitonic superfluid made of fractional charges, a state long predicted but never cleanly observed.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theoretical identification of the 1/3 state as the interlayer coherent (333) state rests on an undisclosed short-range interaction parameter; without its value and energy error bars, the Monte Carlo support is not independently verifiable from the arXiv submission.","rationale":"The reader's weakest assumption correctly identifies the opaque Monte Carlo Hamiltonian parameter as the most load-bearing concern. The experimental evidence for a 1/3 incompressible state is reasonably strong (conductivity minima on multiple devices, quantized Hall plateau on D2), so the main risk lies in the theoretical assignment of the state. The paper's abstract and discussion emphasize the interlayer coherent (333) state as the key result; this assignment rests on trial-wavefunction energy comparisons whose Hamiltonian includes a phenomenological short-range term with unspecified strength. Without knowing this parameter, one cannot rule out that the phase diagram was tuned to match the data, and without error bars one cannot assess whether the competing states are actually distinguished. A single parameter-variation and error-bar check would settle whether the theoretical claim is robust. Since this is exactly the reader's concern, the verdict remains CONDITIONAL (i.e., unchanged).","tokens_in":10480,"tokens_out":7338,"duration_ms":72808,"concrete_test":"Obtain the short-range interaction strength from the SI (or the authors) and re-run the Monte Carlo energy comparison for νtot=1/3 at zero displacement field with this parameter set to zero and to ±50% of its nominal value, computing the energy difference between the (333) state and the next-lowest competing trial state with error bars. If the (333) state is not the lowest-energy state across this range, or if the energy gap is within one standard deviation, the identification is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is not merely that a 1/3 FQH state exists, but that the state is the interlayer coherent (333) state, an excitonic superfluid of fractional charges. This identification is supported exclusively by Monte Carlo energy simulations (Fig. 2) whose Hamiltonian includes a 'phenomenological short-range interaction term' (Methods, Monte Carlo simulation). The strength of this term is not stated in the main text or in the arXiv version; the details are deferred to Supplementary Notes 4–7 of the SI, which is not part of the posted preprint. If this parameter is adjusted to favor the observed phases, or if the energy difference between the (333) state and a competing trial wave function is within the Monte Carlo statistical error, the theoretical interpretation is not an independent prediction. No error bars are reported for the energy comparisons in the main text, and the trial wave function set may not include the true ground state. Since the unique claim of the paper—that the 1/3 state is an interlayer coherent state of fractional charges—depends entirely on this simulation, the missing parameter and error estimates constitute a load-bearing gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports magnetotransport measurements on four large-angle twisted bilayer graphene devices and observes a conductivity minimum at total filling νtot = 1/3 under balanced layer population, together with a quantized Hall plateau at σxy = -1/3 e²/h in device D2. The authors interpret this state as an interlayer-coherent two-component (333) state, the fractional analogue of the (111) excitonic superfluid, based on Monte Carlo energy comparisons among trial wave functions. They also report displacement-field-driven transitions at several fractional fillings and compare the observed features with a calculated phase diagram.","tokens_in":10688,"tokens_out":5575,"duration_ms":55804,"significance":"If correct, the result would demonstrate an excitonic superfluid made of fractional charges, with the same topological properties as the single-layer 1/3 Laughlin state but with the wave function spread coherently across both layers. The experimental evidence is a clear strength: the 1/3 feature appears in multiple devices, develops with magnetic field, and a Hall plateau is observed at -1/3. The theoretical comparison covers several fillings and explicitly identifies states with different symmetry and topology. The main weakness is that the central identification rests on Monte Carlo simulations whose Hamiltonian includes a phenomenological short-range interaction term whose value, and the energy differences with statistical errors, are not reported in the posted manuscript. This gap is load-bearing because the unique claim of the paper is not just the existence of a 1/3 incompressible state, but its assignment to the interlayer-coherent (333) state.","major_comments":[{"comment":"The Hamiltonian used for the phase diagram contains a phenomenological short-range interaction term, but neither its strength nor the full model parameters (dielectric constant, screening, capacitive energy) are stated in the main text; the details are deferred to Supplementary Notes 4–7. Since the identification of the νtot = 1/3 ground state as the (333) state, and the phase boundaries in Fig. 2(a), are decided by energy differences among trial wave functions, this missing parameter makes the central theoretical claim unreproducible from the posted manuscript. Please report the interaction strength, the Hamiltonian, and the Monte Carlo energy differences with statistical error bars for the competing states at νtot = 1/3.","section":"Methods, Monte Carlo simulation"},{"comment":"The Monte Carlo comparison is performed only over a finite set of trial wave functions; the paper does not state that the true ground state is guaranteed to be in this set or provide any unbiased check, such as exact diagonalization of small systems with the same Hamiltonian. The assertion that the balanced 1/3 state is the interlayer-coherent (333) state is therefore conditional on this assumption. Please state this assumption explicitly or add a small-system exact-diagonalization confirmation.","section":"Theoretical phase diagram, Fig. 2(a)"},{"comment":"The text states that the simulated phase diagram in Fig. 2(a) shows 'reasonable agreement' with the transport data, but no quantitative criterion is given for matching conductivity minima and transition features to the calculated state changes. Because the simulations are the sole basis for identifying the nature of the states, at least one quantitative comparison (for example, the displacement field of a transition at a fixed filling, or an estimated energy gap at B = 19 T) would substantially strengthen the claim.","section":"Results, Magnetotransport data; Fig. 1(a)"}],"minor_comments":[{"comment":"The label 'Dv/εint' in panel (e) appears to be a typo for 'D/εint'; please correct it.","section":"Fig. 1 caption, panel (e)"},{"comment":"The sentence 'The algorithm is ahead optimized by acceptance rates and integrated autocorrelation times computed from the pre-run data' is unclear; please rephrase, for example as 'The algorithm is optimized using acceptance rates and integrated autocorrelation times computed from pre-run data.'","section":"Methods, Monte Carlo simulation"},{"comment":"The text should state explicitly that the lock-in saturation artefact is confined to the region near charge neutrality and does not affect the plateau at νtot = -1/3, since the plateau is a central quantitative result.","section":"Fig. 3(b)"},{"comment":"The main text repeatedly refers to Supplementary Notes 4–10 for the Hamiltonian, trial wave functions, and topological properties; the arXiv version lacks these notes, so the preprint is not self-contained. Please ensure the Supplementary Information is included with the submission or summarize the essential equations in the main text.","section":"General presentation"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern is valid and is the main reason for my recommendation. The experimental observation of the 1/3 incompressible state appears solid, but the paper's unique theoretical claim depends on parameters and error estimates that are not disclosed in the posted manuscript. This gap is repairable: the authors should provide the Hamiltonian, the short-range interaction strength, energy differences with Monte Carlo error bars, and either an explicit statement of the trial-wave-function limitation or an unbiased check. I would not reject on the present evidence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe headline: they have a genuine 1/3 fractional quantum Hall plateau at balanced layer population in large-angle twisted bilayer graphene, with Hall quantization, and that is new. The earlier double-layer graphene observation used Corbino geometry and had no Hall measurement; here they show σxy quantized at -1/3 in D2, and the state appears in four devices. On the experimental side, the paper is careful. They flag the lock-in saturation artifact near charge neutrality, show field and temperature dependence, and the data look plausible.\n\nThe theory is another matter, in a mixed sense. The claim that the 1/3 state is the interlayer coherent (333) fractional excitonic superfluid is supported only by Monte Carlo energy comparisons with trial wave functions. The Hamiltonian includes a phenomenological short-range interaction term whose strength is not stated in the arXiv main text, and the energy differences have no reported error bars. That is a genuine gap. If the SI does not give the parameter and a robustness check over its value, the theoretical assignment is not independently verifiable from the preprint. The stress-test has it basically right.\n\nBut I want to push back on the 'load-bearing' label. The experimental discovery does not depend on the Monte Carlo. Even if the true ground state were a different two-component state, the observation of an incompressible state at νtot = 1/3 at zero displacement field, with interlayer interactions as the only plausible cause, is a strong result. The (333) assignment is an interpretation, and a plausible one, but it is secondary. So the missing SI detail affects the paper's interpretive claim, not its primary observation.\n\nOther soft spots: the quantized Hall plateau is shown for one device; the other samples show only longitudinal response. That is a minor complaint for a transport paper, but I would rather see a second Hall trace. The phase diagram is complex, and how the trial states are constructed is not easy to verify without the SI. A referee should push for parameter values, error bars, and a statement of whether the energy ordering is stable across reasonable variations of the short-range term.\n\nOverall: this deserves serious peer review. The experiment is a clean advance in a crowded field, and the theory, while incomplete in the preprint, can be fixed with transparency. I would bring it to reading group and would cite it if I worked on interlayer FQH.\n\nRecommendation: send it to referee, with a request for the SI and explicit parameter values.","headline":"Solid experimental observation of a quantized 1/3 FQH plateau in balanced large-angle twisted bilayer graphene; the (333) identification needs the SI before it can be evaluated, but that gap does not undermine the core experimental result.","tokens_in":11234,"tokens_out":2283,"would_cite":true,"duration_ms":23927,"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":"The paper reports that in balanced large-angle twisted bilayer graphene, the ground state at total filling 1/3 is an interlayer coherent excitonic superfluid of fractional charges, with the same topological order as the 1/3 Laughlin state…","keywords":["fractional quantum Hall effect","twisted bilayer graphene","interlayer exciton condensate","(333) state","Laughlin state","Monte Carlo simulation","displacement field","quantized Hall plateau"],"falsifier":"A decisive check is to measure the Hall conductivity at $\\nu_{\\text{tot}}=-1/3$ and zero displacement field in a device free of contact saturation; the claimed state requires a true plateau at $\\sigma_{xy}=-(1/3)e^2/h$ that persists over a range of magnetic field and temperature. A second check is to rerun the Monte Carlo comparison with varied strengths of the short-range interaction and with trial states outside the selected set, such as two decoupled $1/3$ Laughlin layers; if any of these beats the $(333)$ state, the identification fails.","tokens_in":10276,"feed_emoji":"🧲","tokens_out":9801,"duration_ms":90732,"temperature":0.7,"pith_summary":"Large-angle twisted bilayer graphene brings two electron layers within $0.33$ nm while the twist suppresses single-particle tunneling, so interlayer Coulomb interactions are far stronger than in semiconductor bilayers. The paper reports a fractional quantum Hall state at total filling $\\nu_{\\text{tot}}=1/3$ for balanced layer population, and Monte Carlo energetics identify it as the $(333)$ state: an excitonic superfluid whose excitons are made of quasiparticles of charge $+1/3$ and $-1/3$. This state has the same topological order as a single-layer $1/3$ Laughlin state, but its wave function is spread coherently across both layers. The same simulations account for displacement-field-driven transitions to layer-polarized states at $1/3$, $2/3$, $4/3$, $8/5$, and $5/3$ filling.","feed_headline":"1/3 state in twisted bilayer graphene is an excitonic superfluid","feed_subtitle":"A 1/3 quantum Hall plateau built from paired ±1/3 charges across two layers, with Laughlin-type topology.","key_machinery":"The central object is the $(333)$ trial wave function, a two-component state in which same-layer and opposite-layer electron pairs both get the same Jastrow factor, $(z_i-z_j)^3$, giving total filling $1/3$ with equal occupation of both layers. It is the workhorse of the argument because Monte Carlo energy evaluation selects it as the ground state at zero displacement field and shows it loses to a layer-polarized $1/3$ Laughlin state as the displacement field grows. The energy calculations use a Hamiltonian for large-angle twisted bilayer graphene that combines Coulomb interaction, displacement-field potential, capacitive energy, and a phenomenological short-range term, with candidate states including composite-fermion, pseudospin-singlet, and interlayer-coherent wave functions. The physical mechanism that makes the state possible is the atomic layer spacing of $0.33$ nm: with magnetic length $l_B$, the interlayer-to-intralayer Coulomb ratio $l_B/d$ can reach about $20$, far beyond what semiconductor bilayers achieve.","core_discovery":"The central claim is that, for balanced population and zero displacement field, the ground state at $\\nu_{\\text{tot}}=1/3$ is the interlayer coherent two-component $(333)$ state, the fractional analogue of the $(111)$ state seen at integer filling. Excitons form out of quasiparticles with fractional charge $+1/3$ and $-1/3$; their Bose-Einstein condensation makes the state incompressible. Topologically the $(333)$ state is identical to the $1/3$ Laughlin state, with the same ground-state degeneracy on a torus and the same fractional excitations, but the wave function is shared equally and coherently by the two layers. When the displacement field is increased, the $(333)$ state gives way to a fully layer-polarized single-component $1/3$ Laughlin state. The paper supports this assignment with a quantized Hall plateau at $\\nu_{\\text{tot}}=-1/3$ in one device, minima at $1/3$ in four devices, and Monte Carlo energy comparisons over a set of candidate trial states.","pith_inferences":["Not reported here: interlayer counterflow or tunnelling spectroscopy. An excitonic superfluid should show a sharp interlayer transport anomaly even though single-particle tunnelling is suppressed by the twist-induced momentum mismatch.","Shot-noise or local charge sensing could test the fractional charge $e/3$ of the quasiparticles inside the coherent state, distinguishing a true fractional exciton condensate from two independent Laughlin layers.","If the short-range interaction parameter is fixed independently, the Monte Carlo phase diagram becomes a quantitative prediction; measuring the critical displacement field at which the $1/3$ state loses coherence would test it directly.","The same physics may appear at other fillings where trial states with interlayer coherence are energetically competitive, suggesting a broader family of fractional excitonic states in this platform."],"forward_implications":["If the assignment is right, large-angle twisted bilayer graphene is a platform for fractional exciton condensates, extending the integer $(111)$ exciton superfluid to a state built from fractional charges.","The displacement field becomes a tuning knob that switches between an interlayer-coherent fractional state and a layer-polarized $1/3$ Laughlin state without changing the topological order.","The measured quantized transport at $\\nu_{\\text{tot}}=-1/3$ should carry Hall conductance exactly $-(1/3)e^2/h$ in the interlayer-coherent regime.","The same Monte Carlo procedure predicts the ground states and transition fillings at $2/3$, $4/3$, $8/5$, and $5/3$, giving a testable phase diagram for future devices."],"supporting_citations":[{"why":"reports a 1/3 state in double-layer graphene in Corbino geometry without theoretical identification; the present work provides the missing state assignment.","marker":"[9]"},{"why":"establishes the integer interlayer-coherent quantum Hall states in large-angle twisted bilayer graphene that the fractional 1/3 state generalises.","marker":"[11]"},{"why":"supplies the model Hamiltonian used for the Monte Carlo simulation and reports the interlayer-coherent states at integer filling in this material.","marker":"[12]"},{"why":"provides the Monte Carlo method for evaluating trial wave-function energies in fractional quantum Hall systems.","marker":"[21]"},{"why":"supplies composite-fermion trial states and the general framework used to construct candidate ground states in the energy comparison.","marker":"[22]"},{"why":"together with [12] it is cited as the basis of the model Hamiltonian, including the displacement-field and capacitive terms.","marker":"[23]"}],"fun_headline_variants":["Excitons of 1/3 charges condense in twisted bilayer graphene","1/3 fractional Hall state in twisted bilayer is a Laughlin twin","Balanced twisted bilayer graphene reveals fractional excitonic superfluid","Fractional quantum Hall effect at 1/3 filling in twisted bilayer","Interlayer coherent 1/3 state forms excitonic superfluid"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The theoretical assignment of the $1/3$ state rests on Monte Carlo energies computed with a Hamiltonian that includes a phenomenological short-range interaction term whose strength is not stated in the main text; the identification is only as secure as that model choice.","fun_headline_variants_meta":{"raw":{"variants":["Excitons of 1/3 charges condense in twisted bilayer graphene","1/3 fractional Hall state in twisted bilayer is a Laughlin twin","Balanced twisted bilayer graphene reveals fractional excitonic superfluid","Fractional quantum Hall effect at 1/3 filling in twisted bilayer","Interlayer coherent 1/3 state forms excitonic superfluid"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000899,"raw_usage":{"total_tokens":3861,"prompt_tokens":927,"completion_tokens":2934,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":543,"completion_tokens_details":{"reasoning_tokens":2841}},"tokens_in":543,"tokens_out":2934,"duration_ms":20968,"temperature":1.0,"reasoning_tokens":2841,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T17:12:08.282891+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check is to measure the Hall conductivity at $\\nu_{\\text{tot}}=-1/3$ and zero displacement field in a device free of contact saturation; the claimed state requires a true plateau at $\\sigma_{xy}=-(1/3)e^2/h$ that persists over a range of magnetic field and temperature. A second check is to rerun the Monte Carlo comparison with varied strengths of the short-range interaction and with trial states outside the selected set, such as two decoupled $1/3$ Laughlin layers; if any of these beats the $(333)$ state, the identification fails.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"reports a 1/3 state in double-layer graphene in Corbino geometry without theoretical identification; the present work provides the missing state assignment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"establishes the integer interlayer-coherent quantum Hall states in large-angle twisted bilayer graphene that the fractional 1/3 state generalises."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the model Hamiltonian used for the Monte Carlo simulation and reports the interlayer-coherent states at integer filling in this material."},{"cited_title":"Morf and B","cited_arxiv_id":null,"evidence_quote":"provides the Monte Carlo method for evaluating trial wave-function energies in fractional quantum Hall systems."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"together with [12] it is cited as the basis of the model Hamiltonian, including the displacement-field and capacitive terms."}],"review_version":1}