{"id":"a0d14ec6-c251-4605-92b9-72ef381e1a7c","arxiv_id":"2411.18910","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A displacement field in ABA trilayer graphene controllably breaks particle-hole symmetry of fractional quantum Hall states by mixing monolayer-like and bilayer-like Landau levels.","lead":"In trilayer graphene, fractional quantum Hall states and their particle-hole partners have identical energy gaps when the displacement field is small. A larger field couples two Landau levels and selectively destroys one member of each pair, giving a controlled way to break this quantum symmetry.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The inferred enhancement of η is built on treating the measured integer-QH activation gap as the single-particle LL spacing, but at the crossing this gap is interaction-lifted; the causal mechanism may therefore be overstated.","rationale":"I agree with the reader that the weakest assumption is the use of the measured integer-QH activation gap as a proxy for the single-particle LL spacing. The rest of the paper's evidence is strong: low-D PH symmetry is quantified by symmetric gaps (Fig. 2), the D-window of PHS violation matches computed LL crossings across B (Fig. S6), and the effect is reproduced in multiple devices. The fitting ambiguity for 1/3 and 2/3 states is a secondary caveat because both fitting forms still show PH-symmetric parameters. The proposed mechanism is, however, inferred rather than directly measured: the paper borrows the three-body interaction argument from theory and ties it to an η enhancement that is itself derived from the measured gap dip. Because the authors explicitly concede that interactions lift the accidental degeneracy, the measured gap cannot be unambiguously assigned to single-particle LL spacing. If the gap dip is interaction-dominated, the η peak is overestimated and the causal attribution to enhanced LL mixing is weakened, though the qualitative observation of controlled PHS breaking survives. A two-LL interaction calculation at the crossing would settle whether the residual gap is single-particle or many-body in origin. The verdict should remain conditional, exactly as the reader concluded.","tokens_in":17463,"tokens_out":6020,"duration_ms":58403,"concrete_test":"Perform a Hartree-Fock or exact-diagonalization calculation of the two crossing LLs (LL0+_M ↑ and LL2+_B ↑) at B = 10 T for D = 0.82–0.86 V/nm, using the same tight-binding parameters, and compute the interaction-corrected charge gap at the integer filling corresponding to ν = 3. Compare this gap with the measured activation gap in Fig. 4e and with the non-interacting single-particle gap. If the interaction-corrected gap closely tracks the measured ΔE3 while the non-interacting gap is much smaller, then the measured gap is interaction-dominated and cannot be used as E_cyc to infer η; the η peak in Fig. 4g would then be an artifact of the proxy.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central causal chain is: D compresses the spacing between LL0+_M and LL2+_B, η = E_C/ΔE3,5 rises, and LL mixing plus three-body terms break PHS. The quantitative link is made by measuring the activation gap of the ν = 3 and ν = 5 IQH states (Fig. 4e–f) and identifying it with the single-particle cyclotron gap. But in the same D window the non-interacting calculation predicts a crossing; the authors state in the text that 'in reality, any interaction will lift this accidental degeneracy, leading to a reduced but finite activation gap.' Thus the measured ΔE3,5 is a many-body gap, not the single-particle spacing. If the residual gap is set by interaction-induced level repulsion rather than by the underlying single-particle separation, then η = E_C/ΔE3,5 is not a clean measure of LL mixing; the factor-of-2.5 dip could partly reflect the interaction that opens the avoided crossing. The selective collapse of 8/3 versus 7/3 is direct evidence of PHS breaking, but the specific mechanism—enhanced η activating three-body interactions—is not uniquely established by this proxy. This is the load-bearing assumption because the paper's headline conclusion, 'extrinsic PHS violation via inter-band LL mixing,' depends on the measured gap dip being a faithful measure of single-particle LL proximity.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports transport measurements on ABA-stacked trilayer graphene (TLG) showing that fractional quantum Hall (FQH) states around half-filling of the monolayer-like zeroth Landau level (LL0+_M) are particle-hole symmetric at small displacement field D, and that this symmetry is broken in a narrow window of D where LL0+_M crosses the bilayer-like LL2+_B level. The authors attribute the symmetry breaking to enhanced Landau level mixing (increased eta = E_C/Delta E_{3,5}) and the activation of three-body interactions, and argue that this constitutes an extrinsic, controlled violation of particle-hole symmetry distinct from intrinsic interaction-driven breaking in single-layer and bilayer graphene.","tokens_in":17715,"tokens_out":6860,"duration_ms":58728,"significance":"If the result holds, this is the first demonstration of controlled, extrinsic particle-hole symmetry violation in the fractional quantum Hall regime, which would be a notable advance. The paper brings multiple devices (three), temperature-dependent activation measurements, and an external tight-binding calculation with literature parameters; the agreement between the theoretically predicted and experimentally observed crossing fields (Fig. S6) is a strong independent check. The direct observation that one conjugate FQH state (e.g., 8/3) collapses while the other (7/3) survives in the same D window is compelling evidence of PHS breaking, independent of the specific mechanism. The central causal attribution, however, relies on a proxy whose validity is not fully established.","major_comments":[{"comment":"The estimate eta = E_C/Delta E_{3,5} is obtained by identifying the measured activation gap of the nu = 3 and 5 integer quantum Hall states with the single-particle inter-LL spacing. In the same D window, the non-interacting calculation predicts a crossing, and the authors state that 'in reality, any interaction will lift this accidental degeneracy, leading to a reduced but finite activation gap.' The measured Delta E_{3,5} is therefore a many-body gap, and the peak in eta(D) shown in Fig. 4(g-h) may overstate the enhancement of Landau-level mixing. Because the headline conclusion attributes the PHS breaking to enhanced eta and three-body interactions, the paper should either provide a quantitative estimate of the interaction-induced contribution to Delta E_{3,5} (for example, by comparing the data across multiple B fields where the crossing is avoided) or explicitly present the eta enhancement as a qualitative indicator rather than a quantitative measure, with the causal mechanism framed as a plausible interpretation. This issue is load-bearing for the central causal claim.","section":"Main text, Fig. 4(e-h)"},{"comment":"The paper asserts that the observed PHS violation arises from enhanced Landau-level mixing and the activation of three-body interactions, which 'explicitly break the PHS of FQHs.' While the data establish a correlation between the D window of PHS breaking and the LL-crossing window, the specific role of three-body interactions is not directly evidenced. The selective collapse of 8/3 (while 7/3 survives) is direct evidence of PHS breaking, but it does not by itself discriminate between the two proposed mechanisms (enhanced eta and three-body terms). To support the mechanism, the authors should compare the measured D-dependence of the gaps with a theoretical model that includes three-body interactions, or at least provide a calculation of the expected PHS asymmetry from eta alone. Without this, the statement that both factors are responsible goes beyond what the data demonstrate.","section":"Abstract and Discussion"}],"minor_comments":[{"comment":"The claim that 'conventional FQHs are completely destabilized' is an overgeneralization; the data show that only one member of each conjugate pair (e.g., 8/3 but not 7/3) collapses in the crossing window.","section":"Abstract"},{"comment":"The mobility of device 2 is written as '11,00,000 cm2V-1s-1'; this should be 1,100,000 cm2V-1s-1.","section":"Supplementary Information, Device characterization"},{"comment":"The row for nu between -4 and -5 lists the most affected state as -11/3, but the text and Fig. S8(d) identify -14/3 as the affected hole-conjugate state; this appears to be a typo.","section":"Supplementary Table 1"},{"comment":"The statement that particle-hole symmetry requires 'the cyclotron energy is significantly greater than the interaction strength' is imprecise: PHS in a single Landau level holds for arbitrary interaction strength as long as LL mixing is negligible. The condition should instead be phrased as the validity of single-LL projection.","section":"Conclusion"},{"comment":"The definition of eta changes from eta = E_C/E_cyc in the Introduction to eta = E_C/Delta E_{3,5} in the results; the relation between the two should be clarified, noting that Delta E_{3,5} serves as a proxy for the cyclotron gap.","section":"Introduction and Main text, Fig. 4(e-h)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript has strong experimental evidence for the central observation: PHS of FQHs in ABA TLG is broken in the D window where LL0+_M and LL2+_B cross, with the crossing field agreeing with a tight-binding calculation using literature parameters. The main weakness is the interpretation of the measured integer-gap dip as a quantitative measure of LL mixing, which is undermined by the interaction-lifted degeneracy. I recommend major revision, asking the authors to either strengthen the link between Delta E_{3,5} and the single-particle spacing, or reframe the mechanism statement as a plausible interpretation rather than a demonstrated fact. The paper is otherwise well-structured and the multi-device reproducibility is a strength."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth your time. The key new thing: in the LL0+_M level of ABA TLG, the 7/3 and 8/3 FQH states have essentially identical activation gaps at low D, and when the displacement field pushes LL0+_M and LL2+_B through a crossing window (0.82–0.86 V/nm), the hole-conjugate 8/3 collapses while 7/3 survives. That is a clean, controlled violation of FQH particle-hole symmetry, and it is not in the earlier SLG/BLG or TLG work. The authors also show the same behavior for 13/3 vs 14/3, and the B-dependence of the crossing field tracks the SWMC tight-binding calculation with literature parameters (Fig. S6). Multiple devices, activated transport, and the Supplementary Table of which states die at which crossings all support the observation. Good paper in that sense.\n\nThe soft spots are about mechanism, not observation. Their headline mechanism is that D shrinks the inter-LL spacing, raises eta = Ec/ΔE3,5, and thereby turns on three-body terms that break PHS. But eta is estimated from the measured activation gap of the ν = 3 and 5 integer states, and the authors themselves note that interactions will lift the accidental crossing degeneracy, leaving a reduced but finite gap. So the measured gap dip is partly a many-body effect, and the eta enhancement is to some degree a restatement of that dip. The selective collapse of 8/3 is direct evidence of PHS breaking; the attribution to enhanced LL mixing plus three-body interactions is the most plausible option but is not uniquely pinned down by this proxy. Also, the “completely destabilized” phrasing oversells it: the breakdown is in a narrow D window and one conjugate survives.\n\nTwo minor things: they admit the 1/3 and 2/3 gaps fit linear-B and sqrt-B equally well, so the g-factor and effective mass extraction for those states is not overinterpreted; and the table contains one apparent typo (–11/3 listed for two different crossings), easy to fix.\n\nBottom line: the empirical result deserves a serious referee and a cite. I would ask the authors to soften the causal claim, quantify how much of the gap dip could be interaction-induced, and check the typo. The paper is not fatally flawed. For the record, the LL-crossing benchmark is not fitted to the symmetry-breaking data, so the core correlation stands.","headline":"A careful transport study showing controlled, displacement-field-induced particle-hole symmetry breaking in ABA trilayer graphene FQH states near a Landau-level crossing; the empirical correlation is solid, the causal mechanism attribution is the soft spot.","tokens_in":18309,"tokens_out":2295,"would_cite":true,"duration_ms":22175,"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":"In ABA trilayer graphene, a displacement field near a Landau-level crossing breaks particle-hole symmetry of fractional quantum Hall states by enhancing Landau-level mixing and three-body interactions.","keywords":["fractional quantum Hall effect","particle-hole symmetry","ABA trilayer graphene","Landau level mixing","displacement field","three-body interactions","composite fermions","activation energy gap"],"falsifier":"Measure the single-particle spacing between $\\mathrm{LL}^{0+}_M$ and $\\mathrm{LL}^{2+}_B$ by a method that does not rely on the transport activation gap, such as magneto-capacitance or inter-Landau-level tunneling, across the displacement-field window $|D| = 0.82$–$0.86$ V/nm. If the true spacing does not dip by the factor of about 2.5 seen in $\\Delta E_{3,5}$, or if the dip is an interaction-induced avoided crossing, then the enhanced-$\\eta$ mechanism would not be the cause of the observed collapse of the hole-conjugate fractional state.","tokens_in":17249,"feed_emoji":"🧲","tokens_out":7570,"duration_ms":59489,"temperature":0.7,"pith_summary":"This paper reports transport measurements on Bernal-stacked trilayer graphene showing that particle-hole symmetry of fractional quantum Hall states around half filling is exact when the lowest monolayer-like Landau level is protected from mixing by the lattice mirror symmetry, and that this symmetry can be switched off by an applied displacement field. At small displacement field, the odd-denominator fractional states and their hole conjugates have equal activation gaps, effective masses, Landé $g$-factors, and disorder broadening. In the narrow displacement-field window where the monolayer-like and bilayer-like Landau levels cross, the measured integer quantum Hall gap falls by a factor of 2.5, the Landau-level mixing parameter $\\eta$ rises to a peak, and one conjugate fractional state collapses while the other survives. The authors conclude that inter-band Landau-level mixing enhances $\\eta$ and activates three-body interactions, which explicitly break particle-hole symmetry; if correct, this makes trilayer graphene the first platform where particle-hole symmetry breaking in the fractional quantum Hall regime can be controlled externally.","feed_headline":"Displacement field flips particle-hole symmetry of FQH states","feed_subtitle":"Odd-denominator FQH gaps stay symmetric until Landau-level mixing near the crossing kills one conjugate state.","key_machinery":"The load-bearing object is the field-tunable crossing between the monolayer-like Landau level $\\mathrm{LL}^{0+}_M$ and the bilayer-like Landau level $\\mathrm{LL}^{2+}_B$ in ABA trilayer graphene. In the pristine mirror-symmetric lattice the two bands cannot mix, so the FQH states in $\\mathrm{LL}^{0+}_M$ are effectively single-Landau-level and particle-hole symmetric. Applying a displacement field breaks the mirror symmetry and, in the window $0.82 < |D| < 0.86$ V/nm, brings the two levels close enough to hybridize. The paper quantifies the resulting mixing through the Landau-level mixing parameter $\\eta = E_c/\\Delta E_{3,5}$, estimated from the measured integer activation gap; the growth of $\\eta$ at the crossing is the proposed trigger that activates three-body interactions and breaks particle-hole symmetry.","core_discovery":"The central claim is that in ABA trilayer graphene the particle-hole symmetry of fractional quantum Hall states about half filling is exact at low displacement fields and can be broken controllably by a displacement field that brings the monolayer-like $\\mathrm{LL}^{0+}_M$ and bilayer-like $\\mathrm{LL}^{2+}_B$ Landau levels together. Pristine TLG hosts FQH states in $\\mathrm{LL}^{0+}_M$ whose activation gaps, effective CF mass parameter, effective $g$-factor, and disorder broadening match those of their hole conjugates; this symmetry is protected by the lattice mirror symmetry that forbids Landau-level mixing. For $|D|$ between 0.82 and 0.86 V/nm, the Landau levels cross, the integer gap $\\Delta E_{3,5}$ drops by a factor of 2.5, $\\eta = E_c/\\Delta E_{3,5}$ peaks, and the hole-conjugate state $8/3$ disappears while its partner $7/3$ remains. The paper argues this is extrinsic particle-hole symmetry breaking: virtual scattering between $\\mathrm{LL}^{0+}_M$ and $\\mathrm{LL}^{2+}_B$ enhances Landau-level mixing and activates three-body interactions, which are known to destabilize conventional FQH states. This is presented as fundamentally different from the intrinsic, interaction-driven symmetry breaking seen in the lowest Landau levels of single-layer and bilayer graphene.","pith_inferences":["If three-body interactions are indeed the active symmetry-breaking channel, states whose stability depends on three-body physics (for example candidate non-Abelian phases) should be most affected near the crossing; searching for their appearance or destruction in the same $D$ window would test this mechanism.","The same field-tunable inter-Landau-level spacing is available in other multiband graphene systems with tunable band structure, so the extrinsic route to particle-hole symmetry breaking may generalize beyond ABA trilayer graphene.","A direct, model-independent measure of $\\eta$ (for example through Landau-level spectroscopy) would convert the inferred peak in $\\eta$ from a transport proxy into a quantitative input for theories of three-body interaction effects.","Because the collapse is selective (the particle state survives while the hole state vanishes), the asymmetry could be used as a sensitive probe of the sign and magnitude of three-body interaction terms, which are usually hard to isolate."],"forward_implications":["At low displacement field, odd-denominator FQH states in $\\mathrm{LL}^{0+}_M$ and their hole conjugates have equal activation gaps, effective masses, $g$-factors, and disorder broadening, directly confirming particle-hole symmetry.","In the crossing window $|D| = 0.82$–$0.86$ V/nm, the integer gap $\\Delta E_{3,5}$ drops by a factor of 2.5, $\\eta$ peaks, and one conjugate FQH state (e.g., $\\nu=8/3$) collapses while its partner ($\\nu=7/3$) survives.","The displacement-field range of particle-hole asymmetry tracks the theoretically predicted Landau-level crossing as the magnetic field is varied, indicating a causal link between inter-band mixing and symmetry breaking.","The symmetry breaking is extrinsic, driven by enhanced Landau-level mixing and three-body interactions, and is distinct from the intrinsic, interaction-driven breaking reported in single-layer and bilayer graphene.","Displacement field $D$ therefore serves as a continuous external control knob to switch particle-hole symmetry on and off in the fractional quantum Hall regime."],"supporting_citations":[{"why":"Establishes that without Landau-level mixing, FQH states and their hole conjugates have identical energy scales and that three-body interactions break this symmetry.","marker":"[13]"},{"why":"Theory showing that Landau-level mixing breaks particle-hole symmetry of FQH states, used as the basis for attributing the observed breaking to $\\eta$.","marker":"[14]"},{"why":"Predicts that enhanced Landau-level mixing destroys conventional FQH states, supporting the collapse of the hole-conjugate state.","marker":"[19]"},{"why":"Tight-binding model of ABA trilayer graphene Landau levels used to identify the $\\mathrm{LL}^{0+}_M$–$\\mathrm{LL}^{2+}_B$ crossing.","marker":"[21]"},{"why":"Provides the Slonczewski-Weiss-McClure parameters and the relation $\\Delta_1 = 82D$ meV/(V/nm) used in the Landau-level spectrum simulation.","marker":"[23]"},{"why":"Supplies the single-particle wavefunction of $\\mathrm{LL}^{0+}_M$, used to argue the level is single-component and valley-isospin transitions are irrelevant.","marker":"[33]"},{"why":"Provides the composite-fermion $\\sqrt{B}$ gap formula and experimental context for activation-gap extraction in graphene.","marker":"[34]"},{"why":"Supports the $\\sqrt{B}$ dependence of composite-fermion gaps used to fit the measured activation energies.","marker":"[36]"},{"why":"The comparison point for intrinsic, interaction-driven particle-hole symmetry breaking in bilayer graphene, which this paper distinguishes from the extrinsic mechanism.","marker":"[43]"}],"fun_headline_variants":["Displacement field toggles FQH particle-hole symmetry","Controllable symmetry breaking in trilayer FQH states","Landau mixing breaks FQH symmetry in trilayer","Field-induced symmetry breaking of FQHs","Extrinsic control of FQH particle-hole symmetry"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The causal attribution rests on assuming that the measured activation gap of the integer quantum Hall state is a faithful measure of the single-particle spacing between the two Landau levels; if interactions substantially lift the degeneracy at the crossing, the inferred peak in $\\eta$ would overstate the actual Landau-level mixing.","fun_headline_variants_meta":{"raw":{"variants":["Displacement field toggles FQH particle-hole symmetry","Controllable symmetry breaking in trilayer FQH states","Landau mixing breaks FQH symmetry in trilayer","Field-induced symmetry breaking of FQHs","Extrinsic control of FQH particle-hole symmetry"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000355,"raw_usage":{"total_tokens":1991,"prompt_tokens":1070,"completion_tokens":921,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":686,"completion_tokens_details":{"reasoning_tokens":842}},"tokens_in":686,"tokens_out":921,"duration_ms":7606,"temperature":1.0,"reasoning_tokens":842,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:46:04.374452+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the single-particle spacing between $\\mathrm{LL}^{0+}_M$ and $\\mathrm{LL}^{2+}_B$ by a method that does not rely on the transport activation gap, such as magneto-capacitance or inter-Landau-level tunneling, across the displacement-field window $|D| = 0.82$–$0.86$ V/nm. If the true spacing does not dip by the factor of about 2.5 seen in $\\Delta E_{3,5}$, or if the dip is an interaction-induced avoided crossing, then the enhanced-$\\eta$ mechanism would not be the cause of the observed collapse of the hole-conjugate fractional state.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that without Landau-level mixing, FQH states and their hole conjugates have identical energy scales and that three-body interactions break this symmetry."},{"cited_title":"Bishara and C","cited_arxiv_id":null,"evidence_quote":"Theory showing that Landau-level mixing breaks particle-hole symmetry of FQH states, used as the basis for attributing the observed breaking to $\\eta$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Predicts that enhanced Landau-level mixing destroys conventional FQH states, supporting the collapse of the hole-conjugate state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Slonczewski-Weiss-McClure parameters and the relation $\\Delta_1 = 82D$ meV/(V/nm) used in the Landau-level spectrum simulation."},{"cited_title":"Landau level evolution driven by band hybridization in mirror symmetry broken ABA-stacked trilayer graphene","cited_arxiv_id":"1611.02395","evidence_quote":"Supplies the single-particle wavefunction of $\\mathrm{LL}^{0+}_M$, used to argue the level is single-component and valley-isospin transitions are irrelevant."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the composite-fermion $\\sqrt{B}$ gap formula and experimental context for activation-gap extraction in graphene."},{"cited_title":"Schulze-Wischeler, E","cited_arxiv_id":null,"evidence_quote":"Supports the $\\sqrt{B}$ dependence of composite-fermion gaps used to fit the measured activation energies."}],"review_version":1}