{"id":"540c554b-075e-49c4-8d43-11a4fe6a55fa","arxiv_id":"2509.03921","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"For attractive potentials ∝1/r^γ with γ>1 in graphene, atomic collapse occurs for arbitrarily small strength, and quasi-bound state energies obey |E_n|^((γ-1)/γ) ∝ n, with some states above the Dirac point.","lead":"This paper shows that in graphene, potentials falling off faster than 1/r, such as 1/r^2, can trap electrons into atomic-collapse-like states even for arbitrarily weak coupling. It predicts the trapped-state energies follow a power-law pattern, and that some states appear above the Dirac point, impossible for Coulomb potentials.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Power-sequence law for γ>1 may be an artifact of dropping the m^2/r^2 term in the WKB integral; low-lying ACSs could violate Eq. (6).","rationale":"The reader's weakest_assumption correctly identifies the WKB window (dropping m^2/r^2 and the r1 approximation) as a load-bearing premise for Eq. (6). My stress-test focuses on that premise and proposes a concrete numerical check that would settle it. I do not find a fatal flaw: the authors' numerics suggest the power sequence may hold, but they lack a quantitative comparison of the fitted exponent/slope to the analytic law, and they do not test the m-dependence or the approach to the boundary of the window. Therefore the appropriate verdict remains CONDITIONAL, consistent with the reader's assessment.","tokens_in":18413,"tokens_out":20292,"duration_ms":189756,"concrete_test":"For fixed γ (e.g., γ=2 and γ=1.5), compute ACS energies two independent ways: (a) numerically from the full radial Dirac equation with the same regularized potential V(r)=−β/(r+r0)^γ, r0=0.02; (b) from the full WKB integral Eq. (4) including the m^2/r^2 term. Fit the resulting ε_n to |ε_n|^p = a − b n, and also perform a free power-law fit. Compare the fitted exponent p and slope b to the Eq. (6) predictions p=(γ−1)/γ and b=(γ−1)π/γ (or the β-scaled version). Repeat for m=3/2 and 5/2, and for smaller r0 (0.01, 0.005) to test the validity window. If p deviates by more than 10% for any state with |ε|/(m^2/β) < 10, or if the slope disagrees with the predicted b after accounting for the next-order m-correction, the universal power-sequence claim is not supported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central quantitative claim is Eq. (6): ACS energies satisfy |ε_n|^{(γ−1)/γ} ≈ const − ((γ−1)π/γ) n. This is derived from the WKB/EBK integral Eq. (4) by two related approximations: (i) replacing the inner turning point r1 by (β/|ε|)^{1/γ}, requiring |ε| ≫ m^2/β, and (ii) dropping the m^2/r^2 term inside the square root, replacing pr by (β/r^γ+ε). This is only valid in the window m^2/β ≪ |ε| ≪ β/r0^γ (Sec. SIII). The numerical fits in Fig. 3 use the first six computed ACS energies from the full Eq. (4), but no quantitative comparison is made to the predicted slopes or intercepts, and the fits are not tested against the m-dropping approximation. For γ=2, the m^2 term contributes at relative order m^2/(β|ε|)^{1/2}; near the top of the sequence |ε_n| falls toward m^2/β, and for higher angular momenta the condition |ε| ≫ m^2/β is even harder to satisfy. If the true ACS energies from direct diagonalization deviate from the power exponent (γ−1)/γ for low-lying states or for m>1/2, then the headline 'power sequence' is not established. This is load-bearing because the infinitesimal-β claim and the SACS discussion both build on Eq. (6).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript studies massless Dirac fermions in graphene in the presence of an attractive singular potential V(r) = -β/r^γ. Using a WKB/EBK quantization condition (Eq. (4)), the authors derive an approximate quantization law, Eq. (6), which predicts that for γ > 1 the atomic-collapse-state (ACS) energies form a power sequence, |ε_n|^{(γ−1)/γ} ∝ n, in contrast to the geometric sequence known for Coulomb impurities. They further argue that the supercritical regime extends to arbitrarily small β in the ideal singular limit, and they identify a class of positive-energy ACSs (SACSs) above the bulk Dirac point. The analytic results are compared with finite-difference solutions of the radial Dirac equation and with tight-binding pybinding LDOS simulations, which show approximately equally spaced resonances on the -|ε|^{(γ−1)/γ} axis.","tokens_in":18938,"tokens_out":24738,"duration_ms":220071,"significance":"If correct, the paper identifies a simple, falsifiable spectral fingerprint for high-order singular potentials in graphene and predicts positive-energy resonances with no Coulomb counterpart. Its strengths include a transparent analytic derivation, two independent numerical verifications (finite-difference Dirac LDOS and tight-binding kernel-polynomial LDOS), and an explicit discussion of regularization dependence in the Supplemental Material. These strengths are partly offset by an algebraic error in the SACS critical-energy expression and by insufficient quantitative support for the m-dropping approximation that underlies the central power-law formula.","major_comments":[{"comment":"Eq. (7) as printed simplifies to ε_c = |m|(β/γ)^{1/(γ−1)}, which decreases as β decreases. This contradicts the sentence immediately below it ('ε_c will increase with the decline of β') and the behavior shown in Fig. 3(c). The correct tangency condition for F1(x)=βx^γ+ε and F2(x)=|m|x gives ε_c = |m|(|m|/β)^{1/(γ−1)}(γ^{−1/(γ−1)} − γ^{−γ/(γ−1)}), which is the expression used in Sec. SVII. Please correct Eq. (7) and re-examine any downstream conclusions that use this critical energy.","section":"The special ACSs at γ > 1, Eq. (7)"},{"comment":"The headline claim of atomic collapse at infinitesimal β is based on the V_eff(r→0)→−∞ argument, but all numerical results use a finite regularization cutoff r0. In fact, Eq. (6) itself implies that the first ACS exists only for β > β_c ≈ (γ−1)π r0^{γ−1} in the hard-cutoff model; Sec. SVI acknowledges that β_c is finite. The manuscript should state this relation explicitly and qualify the abstract and phase diagram so that 'infinitesimal charge' is understood as the r0→0 limit, not as a statement about any regularized physical potential.","section":"Abstract, Fig. 1, and Sec. SVI"},{"comment":"Equation (6) is derived by dropping the m^2/r^2 term and setting r1≈(β/|ε|)^{1/γ}. The validity condition γ(|ε|/β)^{(γ−1)/γ} ≫ |m|/β is stated in Sec. SIII but is never verified for the states used in the fits. In particular, Fig. 3(a) does not state the β values, and the number of available ACSs at γ=1.5 is very sensitive to β. Please add a table comparing the full WKB energies from Eq. (4) with Eq. (6) for each fitted state, including the value of γ(|ε_n|/β)^{(γ−1)/γ}/(|m|/β), so the reader can judge whether the power law is established in the claimed regime.","section":"Sec. SIII and Figs. 2/3"}],"minor_comments":[{"comment":"The β values used for the 'first six ACSs' fits should be stated explicitly. The text mentions β=0.45 in several places, but for γ=1.5 and r0=0.02, Eq. (6) allows only about two states at that β; the fits in Fig. 3(a) must therefore use different (likely larger) β values.","section":"Figs. 2 and 3 captions"},{"comment":"The expression for ε_c appears in two different forms in the main text and in Sec. SVII. The Supplemental version is consistent with the tangency derivation; please unify the notation and ensure Eq. (7) matches.","section":"Notation in Eq. (7) vs. Sec. SVII"},{"comment":"The critical values β_c are reported in physical units (eV^{-...}) for the tight-binding model. It would be helpful to also give the corresponding dimensionless values and a direct comparison with β_c ≈ (γ−1)π r0^{γ−1} obtained from Eq. (6).","section":"Sec. SVI"},{"comment":"There are several minor typographical errors (e.g., 'whithin' in Sec. SI, 'monmentum' in the same section). A careful proofreading pass is recommended.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is potentially publishable after revision. The central power-sequence claim is well supported by two independent numerical methods, but the Eq. (7) error and the lack of quantitative WKB-window validation are load-bearing and must be fixed. I would also ask the authors to moderate the abstract's 'infinitesimal charge' phrasing so that it is explicitly tied to the r0→0 limit, consistent with their own discussion in Sec. SVI."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this paper makes a credible case that in graphene, potentials going as −β/r^γ with γ>1 collapse at arbitrarily small β, and the collapse-state energies follow a power sequence with exponent (γ−1)/γ. The positive-energy collapse states (SACSs) are the genuinely new piece; I haven't seen that before. The core argument holds up; my main objection is thinner than the stress-test suggests.\n\nThe phase diagram in Fig. 1 is correct, and the WKB/EBK derivation is honest about its approximations. The numerical support is double: a finite-difference Dirac LDOS calculation and a tight-binding pybinding simulation, with different regularization schemes. That's real evidence. The supplement also does something I like: it explicitly says βc is finite for any finite cutoff, so the 'infinitesimal charge' line is an idealization, not a claim about any physical sample. The caveat belongs in the main text though; the abstract overstates it.\n\nWhere the paper wobbles: the fits in Figs. 2 and 3 are visual, not quantitative. There are no error bars, no comparison of the fitted slope to the predicted slope from Eq. (5)/(6). That's a real gap because the power exponent is the headline. It wouldn't take much to fix: report the slopes and intercepts and show they match β/r0 and the predicted n-coefficient within expected corrections. The stress-test's worry about dropping the m^2/r^2 term in the WKB integral doesn't land for the parameters used; with r0=0.02 and β=0.45, the m-term is smaller by orders of magnitude at the inner cutoff, and the fits include the first six states, which are in the window where the approximation is valid. So I don't think Eq. (6) is an artifact.\n\nThe SACS above the Dirac point is supported by both a WKB turning-point argument and the tight-binding LDOS, so I'd trust it.\n\nThis is a paper for people working on graphene artificial atoms and singular potentials. It deserves a serious referee. I'd send it out, with a request for quantitative fits and a main-text qualification of the infinitesimal-β claim.","headline":"A credible and genuinely useful paper; the power-sequence law for γ>1 is plausible, the positive-energy collapse states are new, and the main fixable gap is the lack of quantitative slope comparison.","tokens_in":19301,"tokens_out":5674,"would_cite":true,"duration_ms":53928,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["73.22.Pr","72.80.Vp"],"model":"deepseek-v4-flash","headline":"This paper argues that any attractive potential steeper than 1/r — V(r)=-β/r^γ with γ>1 — collapses atomic orbits in graphene at arbitrarily small β, arranging the collapse-state energies as a power sequence and even placing some states abo","keywords":["atomic collapse","graphene","singular potential","power sequence","Dirac fermions","quasi-bound states","Klein tunneling","local density of states"],"falsifier":"Compute exact finite-difference or tight-binding LDOS for a bare singular potential while decreasing the cutoff radius by an order of magnitude; if the first collapse resonance does not appear at arbitrarily small β, or if the |ε_n|^{(γ−1)/γ} versus n fit degrades for the lowest-energy states (where the m^2/r^2 term is not negligible), the central claim is falsified. Alternatively, an STM experiment measuring LDOS peak positions on the |E|^{(γ−1)/γ} axis should find equally spaced resonances with a slope that tracks 1/(γ r0^{γ−1}); any clear deviation would invalidate Eq. (6) as stated.","tokens_in":18371,"feed_emoji":"⚛️","tokens_out":5129,"duration_ms":56522,"temperature":0.7,"pith_summary":"The paper asks what happens to massless Dirac fermions in graphene when an attractive potential is more singular than the Coulomb 1/r law, i.e. V(r)=-β/r^γ with γ>1. It argues that, unlike Coulomb potentials which require a supercritical charge to trigger atomic collapse, a high-order singular potential collapses atomic orbits at any strength β>0. It further claims that the resulting atomic collapse states follow a universal power sequence in energy, not the geometric sequence known for Coulomb potentials, and that some of these states can sit above the bulk Dirac point. If correct, weak screened-Coulomb impurities or other fast-decaying potentials should produce measurable collapse resonances with a clean spectral fingerprint in graphene.","feed_headline":"Atomic collapse needs no critical charge in 1/r^γ wells","feed_subtitle":"For potentials steeper than 1/r, graphene's collapse resonances should space as a power sequence — a clear experimental fingerprint.","key_machinery":"The central object is the semiclassical radial-momentum expression p_r^2=(β/r^γ+ε)^2−m^2/r^2, which determines where electrons can move classically and where they are trapped. Combined with the Einstein–Brillouin–Keller quantization integral Eq. (4), it yields the power-sequence formula Eq. (6): the quantization integral is dominated by β/r^γ+ε, so the resulting energies satisfy |ε_n|^{(γ−1)/γ}∝n. This same machinery produces the phase diagram: for γ>1 the effective potential at r→0 is always attractive, so no critical β exists, and the tangency condition between βx^γ+ε and |m|x produces special positive-energy collapse states.","core_discovery":"For a massless Dirac fermion in graphene under a singular attractive potential β/r^γ, the paper finds that the supercritical regime — where classical electron orbits spiral into the center — appears for every β>0 once γ>1, whereas at γ=1 it requires β>|m|. Using a WKB/EBK quantization integral with the inner turning point r1≈(β/|ε|)^{1/γ}, the authors derive a universal power law: I≈β/(γ−1)(r0^{-(γ−1)} − γ(|ε_n|/β)^{(γ−1)/γ})=nπ, so |ε_n|^{(γ−1)/γ} is approximately linear in the level index n. This replaces the geometric energy sequence of Coulomb atomic collapse. The same WKB picture also shows that for γ>1 two turning points can exist even for positive energies, producing special atomic co","pith_inferences":["The paper leaves implicit that the measured power exponent (γ−1)/γ could be used as an inverse-spectroscopy tool: extracting the exponent from LDOS peak spacings would infer the effective decay exponent of the potential in a given device.","A testable extension would be to vary the cutoff radius r0 systematically while monitoring the first collapse resonance: the claim that βc→0 requires r0→0 implies a quantitative scaling of βc with r0 that could be checked directly in simulations or engineered tip potentials.","Positive-energy SACSs, if realized, might alter transport at energies above the Dirac point in p-n junctions, since they provide resonances where Klein tunneling would otherwise predict full transmission; conductance dips or phase shifts could accompany these states.","The geometric-to-power-sequence crossover as γ passes through 1 suggests a sharp experimental marker: near γ=1 the ratio between adjacent collapse energies is nearly constant, while for any γ>1 it slowly changes; measuring this ratio could distinguish screened from bare Coulomb potentials."],"forward_implications":["If the central claim holds, STM measurements of local density of states near a charged impurity or quantum dot in graphene should show resonances equally spaced on the |E|^{(γ−1)/γ} axis whenever the effective potential decays faster than 1/r.","Screened Coulomb impurities in doped graphene, which develop an asymptotic 1/r^3 tail, would be expected to display atomic collapse even at subcritical bare charges, broadening the parameter regime for collapse experiments.","The appearance of positive-energy collapse states above the bulk Dirac point provides a distinctive experimental signature that a potential is genuinely non-Coulomb, since such states cannot occur for a pure 1/r well.","As γ increases toward very large values, the potential approaches a square-well limit; the predicted power (γ−1)/γ tends to 1, so the collapse-state spacing should continuously morph into an equally spaced sequence — a trend the paper explicitly notes.","Two-terminal conductance measurements through a graphene flake with a central charge impurity should show resonant conductance peaks at the predicted power-sequence energies, giving a transport-based test of the spectrum."],"supporting_citations":[{"why":"Supplies the Coulomb-potential atomic collapse baseline and the geometric ACS energy series that the γ>1 power sequence is contrasted against.","marker":"[5]"},{"why":"Provides the discrete scale invariance / scale anomaly context used to interpret why the geometric sequence at γ=1 gives way to a power sequence at γ>1.","marker":"[19]"},{"why":"Motivates high-order singular potentials by showing nonlinear screening of charged impurities in doped graphene produces an asymptotic 1/r^3 tail.","marker":"[38]"},{"why":"Supplies the Dirac radial-equation decomposition and the finite-difference method used to compute the LDOS maps that support the analytic results.","marker":"[42]"},{"why":"Gives the Einstein–Brillouin–Keller quantization rule on which the central quantization integral Eq. (4) rests.","marker":"[45]"},{"why":"Provides the finite-nuclear-radius analogy used to explain that a finite cutoff r0 makes the numerically observed critical β finite even though the singular limit permits β→0.","marker":"[56]"}],"fun_headline_variants":["Collapse without a critical charge in steep graphene wells","Power-law collapse states: no critical charge needed","Infinitesimal charge triggers collapse in steep potentials","Steeper than Coulomb: no critical charge for collapse","Graphene's steeper wells collapse at any strength"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The derivation assumes the angular-momentum barrier can be ignored in the quantization integral, which only holds for states with intermediate energies, and it assumes the cutoff radius can be made arbitrarily small, while every numerical demonstration uses a finite cutoff that restores a finite critical strength.","fun_headline_variants_meta":{"raw":{"variants":["Collapse without a critical charge in steep graphene wells","Power-law collapse states: no critical charge needed","Infinitesimal charge triggers collapse in steep potentials","Steeper than Coulomb: no critical charge for collapse","Graphene's steeper wells collapse at any strength"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000702,"raw_usage":{"total_tokens":2995,"prompt_tokens":726,"completion_tokens":2269,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":470,"completion_tokens_details":{"reasoning_tokens":2193}},"tokens_in":470,"tokens_out":2269,"duration_ms":17730,"temperature":1.0,"reasoning_tokens":2193,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T10:33:00.629144+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute exact finite-difference or tight-binding LDOS for a bare singular potential while decreasing the cutoff radius by an order of magnitude; if the first collapse resonance does not appear at arbitrarily small β, or if the |ε_n|^{(γ−1)/γ} versus n fit degrades for the lowest-energy states (where the m^2/r^2 term is not negligible), the central claim is falsified. Alternatively, an STM experiment measuring LDOS peak positions on the |E|^{(γ−1)/γ} axis should find equally spaced resonances with a slope that tracks 1/(γ r0^{γ−1}); any clear deviation would invalidate Eq. (6) as stated.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Coulomb-potential atomic collapse baseline and the geometric ACS energy series that the γ>1 power sequence is contrasted against."},{"cited_title":"Ovdat, J","cited_arxiv_id":null,"evidence_quote":"Provides the discrete scale invariance / scale anomaly context used to interpret why the geometric sequence at γ=1 gives way to a power sequence at γ>1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Motivates high-order singular potentials by showing nonlinear screening of charged impurities in doped graphene produces an asymptotic 1/r^3 tail."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Dirac radial-equation decomposition and the finite-difference method used to compute the LDOS maps that support the analytic results."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the Einstein–Brillouin–Keller quantization rule on which the central quantization integral Eq. (4) rests."},{"cited_title":"14,673(1972)","cited_arxiv_id":null,"evidence_quote":"Provides the finite-nuclear-radius analogy used to explain that a finite cutoff r0 makes the numerically observed critical β finite even though the singular limit permits β→0."}],"review_version":1}