{"id":"3b4a0ab8-2374-4cc7-8daf-cdf456989d5f","arxiv_id":"2502.01169","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"The renormalization group with the zero-sound diagram predicts the Kohn-Luttinger pairing gap exponent scales as -ℓ instead of the standard -ℓ^4.","lead":"A theoretical paper revisits the Kohn-Luttinger mechanism, in which repulsive interactions still produce superconductivity, and argues that the pairing gap's exponential size is controlled by angular momentum ℓ rather than ℓ^4. If correct, the effect is dramatically stronger than previously thought, which matters for neutron star cooling and unconventional superconductors.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The −ℓ gap exponent in Eq. (71) is not yet derived: the partial-wave projection (55) is dominated by q→0, the one regime where the hard-cutoff βZS expression (50)–(52) is invalid.","rationale":"The paper's headline claim is Eq. (71), and its only derivation path goes through βZS,ℓ in Eq. (54). That coefficient is obtained in three steps: (i) the hard-cutoff integral is evaluated in the limit q̃ ≡ 2kF − q ≪ Λ, giving Λ dI/dΛ ∝ 1/q; (ii) the result is treated as valid for all q; (iii) the exact identity (55) is used to project 1/q onto partial waves, yielding 1/(2ℓ+1). The problem is that step (iii) is dominated by z=1 (q→0), not by z=−1 (θ=π): the integrand Pℓ(z)/√(1−z) has an integrable singularity at z=1, while z=−1 is regular. The paper's own assertion that the projection is dominated by z=−1 is false. Moreover, step (i) is not valid for q→0: there the Θ constraints restrict the loop phase space to a strip of width ~q, so I(Λ,q) is finite and essentially Λ-independent for q≲Λ, meaning Λ dI/dΛ does not behave as −N0Λ ln2/(2q). Applying the exact 1/q projection across all angles therefore mixes a regime where the quoted beta function is inapplicable, and the resulting 1/(2ℓ+1) coefficient is not a derived consequence of the hard-cutoff calculation. A correct treatment would need either a q-dependent vertex or a properly cut-off partial-wave projection; neither is given. The ℓ=1 RG comparison in Sec. VI agrees with known perturbative coefficients near λ∼0.5, but it cannot discriminate between −ℓ and −ℓ⁴ scaling at large ℓ, and the proposed two-loop test in Sec. VII is only a proposal. Thus the central claim lacks a secure derivation, and the reader's CONDITIONAL verdict is appropriate: the angular projection must be corrected and the beta function recomputed before the −ℓ exponent can be accepted.","tokens_in":24199,"tokens_out":15862,"duration_ms":178538,"concrete_test":"Numerically evaluate the exact hard-cutoff integral I(Λ,q) in Eq. (47) for fixed ratios Λ/kF = 0.01, 0.1, and 0.5, for q from 0 to 2kF, obtain Λ∂I/∂Λ by finite differencing in Λ, and project onto Pℓ(z) over the full angular range. Compare the resulting βZS,ℓ with Eq. (54). If the projected coefficient is not −N0V²(π)ln2/(2ℓ+1)Λ/kF but instead decays faster in ℓ (e.g. ℓ⁻² or ℓ⁻⁴), or if the dominant angular contribution is not θ=π, then the −ℓ scaling in Eq. (71) is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim Eq. (71) rests on βZS,ℓ = −N0 V²(π) ln2/(2ℓ+1) Λ/kF in Eq. (54). That coefficient is obtained by writing Λ dI/dΛ ∝ 1/q as in Eq. (52) and then using the exact identity (55) to project 1/q onto partial waves. But Eq. (52) was derived under q̃ ≡ 2kF − q ≪ Λ, i.e. q ≈ 2kF, while the projection integral (55) is dominated by the opposite limit, q→0: for q(z)=kF√(2(1−z)), the integrand Pℓ(z)/q(z) has an integrable singularity at z=1, where Pℓ(1)=1, and is regular at z=−1. The paper's assertion after Eq. (55) that the integral is dominated by z=−1 is therefore incorrect. In the q→0 region, step (50)–(52) also fails: for q ≲ Λ the Θ-function phase space makes I(Λ,q) essentially q-independent, so Λ dI/dΛ is not −N0Λ ln2/(2q). The global 1/q projection thus mixes in an angular region where the quoted beta function is not valid, and the 1/(2ℓ+1) coefficient in Eq. (54) is not established. Since the sign, magnitude, and ℓ dependence of this ZS beta function control the denominator in Eqs. (69)–(71), the replacement of the −ℓ⁴ scaling by −ℓ is currently unsupported. The ℓ=1 comparison in Sec. VI is suggestive but does not test the large-ℓ exponent; the two-loop projection proposed in Sec. VII is not carried out.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper revisits the Kohn-Luttinger mechanism for superconductivity from repulsive interactions and proposes that, when the zero-sound (ZS) diagram is included in the one-loop RG beta function, the pairing-gap exponent changes from ln(Δ/μ) ∝ −ℓ^4 to ln(Δ/μ) ∝ −ℓ. After reviewing the standard KL argument and the RG setup, the author computes the ZS beta function with a hard cutoff in Sec. IV, projects it onto partial waves using an exact Legendre identity, solves the resulting Riccati equation in Sec. V, and extracts the new exponent. The paper then applies the formalism to the triplet p-wave gap from a repulsive s-wave contact interaction and compares with known perturbative results, arguing that the RG result effectively captures the poorly converging next-to-leading-order corrections. The central quantitative claim is Eq. (71), and the p-wave comparison in Sec. VI is presented as a consistency check rather than as a test of the large-ℓ exponent.","tokens_in":24673,"tokens_out":15683,"duration_ms":167787,"significance":"If the claimed result were established, it would be significant: replacing the extremely small KL gap exp(−cℓ^4) by the much larger exp(−c'ℓ) would change quantitative predictions for higher-partial-wave pairing in neutron-star matter, cold atoms, and other weakly coupled Fermi systems. The paper is self-contained and has genuine strong points: the RG equation is solved explicitly with no fitted parameters, the derivation produces a falsifiable prediction for the partial-wave projection of a two-loop diagram in Eq. (92), and the ℓ=1 comparison with available perturbative coefficients is a useful sanity check. However, the magnitude and even the sign of the central effect hinge on a subleading, cutoff-dependent term in the ZS beta function, and the projection step used to obtain that term has a technical flaw that must be repaired before the central claim is credible.","major_comments":[{"comment":"The derivation of the partial-wave ZS beta function, Eq. (54), is not valid as written. Equation (52) is obtained under the assumption q̃ ≡ 2kF − q ≪ Λ, i.e. q ≈ 2kF, but the projection identity (55) is then used for all q. The statement immediately after Eq. (55) that the integral is dominated by z = −1 is incorrect in the large-ℓ regime. Writing z = cos θ, Eq. (55) becomes (1/kF)∫_0^π P_ℓ(cos θ) cos(θ/2) dθ / (2?) — more precisely the full integral is (1/kF)∫_0^π P_ℓ(cos θ) cos(θ/2) dθ, and for large ℓ the asymptotic P_ℓ(cos θ) ≈ J_0((ℓ+1/2)θ) gives ∫_0^π P_ℓ(cos θ) cos(θ/2) dθ ≈ 1/(ℓ+1/2), exactly the value claimed in Eq. (55). This contribution comes from θ ≈ 0, i.e. q → 0, where the weight cos(θ/2) is maximal; near z = −1 the weight cos(θ/2) vanishes. Therefore for large ℓ, the exact identity (55) is dominated by the q → 0 region, precisely the region where Eq. (52) is not valid: for q ≲ Λ the phase-space restrictions make I in Eq. (50) essentially q-independent, so Λ dI/dΛ is not −N0 Λ ln2/(2q). The coefficient 1/(2ℓ+1) in Eq. (54), and hence the −ℓ exponent in Eq. (71), are not established by the calculation presented.","section":"§IV, Eqs. (50)–(55)"},{"comment":"The O(Λ/kF) term in Eq. (52) is a subleading term in an expansion of an irrelevant operator computed with a hard cutoff. Beta-function coefficients for irrelevant operators are generally scheme dependent, and the comparison with Shankar's soft-cutoff calculation in Sec. IV only addresses the O((Λ/kF)^0) contribution, which vanishes upon differentiation with respect to Λ. No argument is given that the coefficient ln2/(2ℓ+1) in Eq. (54) is independent of the regularization scheme, nor that the truncated RG flow with only this single irrelevant term produces a scheme-invariant physical gap. Because the central result in Eq. (71) is directly proportional to this coefficient, the paper needs either a scheme-independent derivation of the subleading ZS beta function or an explicit demonstration that the gap exponent is independent of the cutoff procedure.","section":"§IV, Eq. (52) and comparison with Ref. [41]"},{"comment":"The replacement V(π; t) ≈ V(π; 0) is an uncontrolled approximation. The ZS term is multiplied by e^{−t}, but the BCS pole in Eq. (69) occurs at t* ≈ 2(2ℓ+1)/(V^2(π) ln2), which grows linearly with ℓ. For large ℓ, t* can be much larger than 1, so the flow is not confined to a small neighborhood of t = 0. Since V(π; t) itself runs through the BCS and ZS couplings, the approximation could change the denominator in Eq. (71) and hence the coefficient, and potentially the ℓ-scaling, of the gap. At minimum, the paper should estimate the size of this correction or identify a parametric reason why it is negligible for large ℓ.","section":"§V, paragraph after Eq. (59)"}],"minor_comments":[{"comment":"The phrase \"Bethe-Salpater equation\" appears twice; it should read \"Bethe-Salpeter equation.\"","section":"Sec. I, paragraph 5"},{"comment":"The sentence \"The emergence of superconducting/superfluid pairing is associated with a singularity in the two-particle vertex function Γ in the particle-particle (BCS) channel ar zero total momentum and frequency\" contains a typo: \"ar\" should be \"at.\"","section":"Sec. II, paragraph 1"},{"comment":"The sentence \"Our intension here is to point out the issue...\" contains a typo: \"intension\" should be \"intention.\"","section":"Sec. VI, paragraph after Eq. (86)"},{"comment":"The figure caption does not specify the horizontal-axis quantity or the values of λ used; the text says the plot range is λ > 0.35, but the axis labels are not visible in the manuscript text and should be given explicitly in the caption.","section":"Fig. 3 caption"},{"comment":"The denominator in Eq. (78) would be easier to read if the spin-channel combination in brackets were explicitly parenthesized, as it is in Eq. (84); as written, the reader must infer the intended grouping from the later equation.","section":"Eq. (78)"}],"recommendation":"major_revision","confidential_remarks":"The paper addresses an interesting and potentially important question, and the author has made a serious attempt to go beyond the standard KL analysis. However, the main quantitative claim currently rests on a projection step that is internally inconsistent: the exact identity (55) is dominated, for large ℓ, by exactly the q → 0 region in which the hard-cutoff beta function (52) was derived to be invalid. This is a load-bearing technical issue, not a matter of presentation. I would encourage the editor to send the manuscript back for a major revision in which the partial-wave projection is performed using the full hard-cutoff expression, or an equivalent scheme-independent calculation, so that the coefficient in Eq. (54) is either established or corrected. The ℓ=1 comparison is interesting but does not test the large-ℓ exponent. I recommend major revision rather than rejection because the proposed mechanism may survive a corrected calculation, but as it stands the central claim is not supported."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's core claim is that the zero-sound diagram, ordinarily dropped as an irrelevant operator, shifts the BCS pole before the RG flow reaches the fixed point and changes the Kohn-Luttinger gap exponent from −ℓ^4 to −ℓ. That claim is genuinely new to me and to the literature as cited. Credit where it is due: the derivation is explicit, self-contained, and parameter-free. The beta function is computed, the RG equation is solved in terms of Bessel functions, and the ℓ=1 case is checked against known perturbative coefficients. This is serious work, and if the result were right it would matter for neutron-star 3P2 pairing and weak-coupling repulsive Fermi systems generally.\n\nThe soft spot is load-bearing, not cosmetic. Equation (54) obtains the ZS beta function coefficient 1/(2ℓ+1) by projecting 1/q onto partial waves using the exact identity (55). The paper says that identity is dominated by z=−1, i.e. q≈2kF, where the Kohn anomaly sits. That is not what happens: the integral is dominated by the integrable q→0 singularity at z=1, where Pℓ(1)=1. But the hard-cutoff expression (50)–(52) was derived in the opposite regime, q̃=2kF−q≪Λ, and is invalid near q→0, where the Θ-function phase space makes Λ dI/dΛ essentially q-independent. So the coefficient that controls the denominator of Eq. (71) is not established, and neither is the −ℓ scaling. The stress-test note is correct on this point. I checked the two-loop argument in Sec. VI C as well: the factorized 1/|k1−k3| projection is exact, but the factorization itself was justified for Kohn-anomaly kinematics, and the proposed two-loop projection is not carried out. The ℓ=1 comparison is suggestive, but it tests one value, not the large-ℓ exponent.\n\nA smaller worry: V(π;t) is frozen at t=0 because the ZS contribution is concentrated at small t. That may be acceptable for a leading estimate, but it deserves a check rather than an assertion.\n\nBottom line: this paper is for theorists working on KL pairing or on RG near the Fermi surface; they will find the RG formulation instructive even if the final exponent fails. It deserves a serious referee, not a desk rejection. I would send it to review with instructions that the author either correct the partial-wave projection or demonstrate directly that the βZS partial-wave coefficient is scheme-independent. As it stands, I would not cite the −ℓ claim, and I would flag it as unproven.","headline":"The zero-sound resummation idea is genuinely new and the calculation is explicit, but the partial-wave projection that produces the 1/(2ℓ+1) coefficient is wrong, so the claimed −ℓ gap exponent is not yet established.","tokens_in":25085,"tokens_out":3856,"would_cite":false,"duration_ms":42215,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that the Kohn-Luttinger pairing gap, long thought to scale as $e^{-c\\ell^4}$, actually scales as $e^{-c'\\ell}$ once the zero-sound contribution enters the RG flow.","keywords":["Kohn-Luttinger mechanism","renormalization group","BCS superconductivity","Kohn anomaly","partial-wave expansion","pairing gap","repulsive interactions","zero-sound diagram"],"falsifier":"Evaluate the two-loop skeleton integrals $\\Pi^{(a)}_{\\mathrm{ppph}}$ and $\\Pi^{(c)}_{\\mathrm{phph}}$ in odd partial waves, as proposed in Eq. (92): if the partial-wave projection scales as $1/\\ell^4$ rather than $1/\\ell$, the predicted exponent fails. A second test is to recompute the zero-sound $\\beta$ function with a smooth cutoff; if the $\\Lambda/q$ term with projection $1/(2\\ell+1)$ disappears, the enhancement is a hard-cutoff artifact.","tokens_in":23995,"feed_emoji":"⚛️","tokens_out":8944,"duration_ms":83988,"temperature":0.7,"pith_summary":"This paper claims that the Kohn-Luttinger mechanism produces pairing gaps far larger than the classic estimate: instead of the exponent of the gap scaling as $-\\ell^4$ with angular momentum $\\ell$, it scales as $-\\ell$. The difference comes from a subleading term in the renormalization-group $\\beta$ function, the zero-sound contribution, which is usually discarded as an irrelevant operator but cannot be neglected before the BCS singularity forms. If this is right, superconductivity from purely repulsive interactions is exponentially easier to reach at high $\\ell$, and the poorly convergent perturbative series for the $p$-wave gap is explained as a sign that the Kohn anomaly must be resummed separately from the BCS logarithm.","feed_headline":"Pairing gap from repulsion scales as ℓ, not ℓ⁴","feed_subtitle":"Zero-sound resummation of the Kohn anomaly makes high-partial-wave pairing exponentially stronger.","key_machinery":"The machinery is the zero-sound (ZS) diagram, a one-loop particle-hole diagram with momentum transfer $q=k_1-k_3$, evaluated with a hard cutoff $\\Lambda$. Its partial-wave $\\beta$ function is $\\beta_{\\mathrm{ZS},\\ell}=-N(0)V^2(\\pi)(\\ln 2)/(2\\ell+1)\\,\\Lambda/k_F$, obtained with the exact identity $\\frac12\\int_{-1}^{1}dz\\,P_\\ell(z)/q(z)=1/((2\\ell+1)k_F)$; the $1/(2\\ell+1)$ factor is what converts the $\\ell^4$ law into $\\ell$. Combined with the BCS $\\beta$ function $-\\frac{N(0)}{2}V_\\ell^2$, the flow equation has a solution built from Bessel functions, and locating its pole gives the BCS singularity. The ZS term is formally irrelevant and renormalizes to zero, but the singularity occurs before the flow reaches the fixed point, so the term shifts where the instability happens.","core_discovery":"The central discovery is Eq. (71): for large $\\ell$, the pairing gap obeys $\\ln(\\Delta/\\mu)\\simeq -2(2\\ell+1)/(V^2(\\pi)\\ln 2)$, so the BCS singularity occurs at RG time $t^*\\propto (2\\ell+1)$ and the gap is exponentially larger than the conventional $\\ln\\Delta\\propto -\\ell^4$. The paper argues that this follows once the zero-sound diagram's subleading contribution $\\beta_{\\mathrm{ZS},\\ell}=-N(0)V^2(\\pi)(\\ln 2)/(2\\ell+1)\\,\\Lambda/k_F$ is included in the $\\beta$ function alongside the BCS term. It identifies the source of the old result as the $\\Lambda^0/\\ell^4$ term, which vanishes when differentiated with respect to the cutoff, while the $\\Lambda/\\ell$ term, previously dropped, controls the flow. The paper illustrates the mechanism with the spin-triplet $p$-wave gap from a repulsive $s$-wave contact interaction, where the RG result matches the next-to-leading-order perturbative result at moderate coupling.","pith_inferences":["Extension: if the $\\ell$ scaling survives scrutiny, the $^3P_2$ pairing gap in neutron-star matter could be orders of magnitude larger than one-loop estimates, strengthening predicted neutron-star cooling rates.","Extension: the hard-cutoff dependence is not settled in the paper; recomputing the zero-sound beta function with a smooth cutoff would test whether the $\\Lambda/q$ term is scheme-independent.","Extension: the same logic in two dimensions, where partial waves are $\\cos(\\ell\\theta)$ and the angular measure differs, may give a different exponent; a 2D version of this calculation would be a direct next step.","Extension: the paper's identification of the divergent integrand suggests a practical diagnostic: a two-loop partial-wave coefficient enhanced by $1/\\ell$ relative to one-loop signals that the Kohn anomaly must be resummed."],"forward_implications":["For large $\\ell$, the Kohn-Luttinger gap is exponentially larger than the classic $\\ell^4$ estimate, making high-partial-wave pairing much more accessible in weak-coupling systems.","In the $p$-wave channel with repulsive $s$-wave contact interaction, the leading-order RG result reproduces the next-to-leading-order Bethe-Salpeter result at moderate coupling, so one-loop inputs effectively capture two-loop physics.","The poor convergence of the perturbative gap series is explained: diagrams containing both a BCS loop and a Kohn anomaly have divergent integrands that must be resummed together.","The RG approach provides a portable way to compute pairing gaps in nuclear and quark matter, where higher-partial-wave pairing such as $^3P_2$ may be amplified."],"supporting_citations":[{"why":"It establishes the Kohn-Luttinger mechanism and the standard $1/\\ell^4$ gap exponent that this paper revises.","marker":"[1]"},{"why":"It supplies the Fermi-surface RG framework and the original zero-sound diagram calculation, re-evaluated here with a hard cutoff.","marker":"[41]"},{"why":"It provides the leading-order $p$-wave pairing gap from a repulsive $s$-wave contact interaction, a baseline for comparison.","marker":"[6]"},{"why":"It gives an independent leading-order derivation of the $p$-wave Kohn-Luttinger instability, also a baseline.","marker":"[7]"},{"why":"It is the two-loop numerical calculation showing poor convergence of the perturbative gap series, which motivates the resummation.","marker":"[32]"},{"why":"It is the earlier RG treatment of the Kohn-Luttinger mechanism that includes irrelevant operators in the beta function.","marker":"[53]"},{"why":"It identifies the Kohn anomaly at momentum transfer $2k_F$, the singular integrand whose resummation drives the enhanced exponent.","marker":"[54]"},{"why":"It is the recent two-loop evaluation of the pairing gap whose BCS-subdiagram integrals the paper analyzes to explain the $1/\\ell$ enhancement.","marker":"[57]"}],"fun_headline_variants":["RG reroute boosts pairing gap from ℓ⁴ to ℓ exponent","Kohn-Luttinger gap: zero sound turns ℓ⁴ into ℓ","Pairing gap amplification: RG yields ℓ instead of ℓ⁴","Zero-sound RG resummation shrinks gap exponent to ℓ","Kohn-Luttinger mechanism strengthened: gap scales as ℓ"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the subleading $\\Lambda/q$ zero-sound term being a genuine, scheme-independent contribution of the hard-cutoff calculation, and on the interaction vertex in the diagram being well approximated by its back-to-back value $V(\\pi)$.","fun_headline_variants_meta":{"raw":{"variants":["RG reroute boosts pairing gap from ℓ⁴ to ℓ exponent","Kohn-Luttinger gap: zero sound turns ℓ⁴ into ℓ","Pairing gap amplification: RG yields ℓ instead of ℓ⁴","Zero-sound RG resummation shrinks gap exponent to ℓ","Kohn-Luttinger mechanism strengthened: gap scales as ℓ"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000823,"raw_usage":{"total_tokens":3663,"prompt_tokens":1072,"completion_tokens":2591,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":688,"completion_tokens_details":{"reasoning_tokens":2491}},"tokens_in":688,"tokens_out":2591,"duration_ms":19974,"temperature":1.0,"reasoning_tokens":2491,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T16:24:31.397767+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evaluate the two-loop skeleton integrals $\\Pi^{(a)}_{\\mathrm{ppph}}$ and $\\Pi^{(c)}_{\\mathrm{phph}}$ in odd partial waves, as proposed in Eq. (92): if the partial-wave projection scales as $1/\\ell^4$ rather than $1/\\ell$, the predicted exponent fails. A second test is to recompute the zero-sound $\\beta$ function with a smooth cutoff; if the $\\Lambda/q$ term with projection $1/(2\\ell+1)$ disappears, the enhancement is a hard-cutoff artifact.","supporting_citations":[{"cited_title":"Kohn and J","cited_arxiv_id":null,"evidence_quote":"It establishes the Kohn-Luttinger mechanism and the standard $1/\\ell^4$ gap exponent that this paper revises."},{"cited_title":"vFGbUYHKLX3FOO0/17bYLN5ZPDw=","cited_arxiv_id":null,"evidence_quote":"It provides the leading-order $p$-wave pairing gap from a repulsive $s$-wave contact interaction, a baseline for comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"It gives an independent leading-order derivation of the $p$-wave Kohn-Luttinger instability, also a baseline."},{"cited_title":"Polarization contributions to the spin-dependence of the effective interaction in neutron matter","cited_arxiv_id":"nucl-th/0307089","evidence_quote":"It is the two-loop numerical calculation showing poor convergence of the perturbative gap series, which motivates the resummation."}],"review_version":1}