{"id":"3de587d9-2ea4-4f78-aa3b-88bf13c5f1f2","arxiv_id":"2608.02426","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A static-Hamiltonian screening method (SEPR) using two descriptors Ξ and χ identifies 34 dynamically stable superconducting candidates with Tc>10 K from 36,377 materials.","lead":"This paper introduces a low-cost screening framework that estimates electron-phonon coupling strength directly from static orbital Hamiltonians, bypassing expensive phonon calculations. Applied to over 36,000 compounds, it identifies 34 dynamically stable superconducting candidates with predicted Tc above 10 K.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Universal decay length L=1.2 Å biases the screening metric by up to ~3x across families; the 34-candidate list is not robust to this documented assumption.","rationale":"The reader's weakest assumption — the universal BA decay length — is the same concern I identify, and the manuscript itself provides direct evidence for it in Sec. V. This is the most load-bearing issue because it sits at the exact step that converts static hoppings into displacement derivatives (Eq. 7), and the screening thresholds and candidate list inherit its bias. The paper is otherwise careful: the benchmarks against DFPT on eight superconductors (Table I) and the comparison for 150 dynamically stable screened compounds (Fig. 4) provide independent support for the overall correlation, and the DFPT verification of the final candidates is a real filter. However, neither of these validations addresses the effect of the uniform L on the absolute thresholds, because both are performed on the same family-biased Ξ/χ values. The proposed element-specific re-screening would settle whether the candidate list is robust. Since this concern reinforces the reader's conditional verdict rather than overturning it, I recommend UNCHANGED.","tokens_in":15973,"tokens_out":7472,"duration_ms":85087,"concrete_test":"Recompute SEPR for all 8,833 nonmagnetic metals using element-specific decay lengths, e.g., set L=0.8 Å for B,C,N,O and L=3.6 Å for alkali/alkaline-earth systems, with intermediate values for other elements fitted from the actual distance dependence of CWF hopping integrals in the MattKeyBond database. Then reapply the same thresholds (Ξ>15.0 meV, χ>125.0 meV·u^1/2) and compare the new candidate pool with the original 260. If the overlap is <80%, or if any compound enters/leaves the pool and the change corresponds to a DFPT Tc>10 K, the universal-L assumption is load-bearing and the 34-candidate list is not robust. If ≥28 of the 34 candidates survive and no new below-threshold compound yields DFPT Tc>10 K, the concern does not land.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central screening quantities Ξ and χ are derived from hopping derivatives via the BA model, Eq. 7: dHab/du ≈ −(R·e/|R|)/L · Hab. With a uniform L=1.2 Å (Sec. II C), these derivatives scale as 1/L. Yet Sec. V and Fig. 4 document that effective decay lengths vary by family: L≈0.8 Å for 2p covalent systems and L≈3.6 Å for alkali/alkaline-earth compounds. Thus Ξ and χ are overestimated by roughly a factor of 3 for alkali-containing materials and underestimated for 2p-covalent materials. Because the screening thresholds Ξ>15.0 meV and χ>125.0 meV·u^1/2 are absolute cutoffs applied to 8,833 metals, this family-dependent bias can admit or exclude compounds based on elemental identity rather than intrinsic EPC strength. The paper acknowledges this as 'predictable family-dependent bias' but does not correct it; the final 260-compound pool, the 34 DFPT-verified candidates, and the Vep–N* taxonomy in Fig. 5 all inherit this bias. This is load-bearing for the central claim of 'semi-quantitative' EPC scale across a broad chemical space: a factor-of-3 family-dependent shift violates semi-quantitative comparability.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a static electron-phonon response (SEPR) framework that estimates electron-phonon coupling (EPC) strength directly from static orbital-based Hamiltonians, bypassing DFPT at the screening stage. Three approximations are introduced: a dynamic isotropic average over phonon modes, a double-center approximation for hopping derivatives, and a bonding-attractivity (BA) model in which hopping integrals decay exponentially with bond length using a single universal decay length L = 1.2 Å. This yields two screening metrics, Ξ and χ. The method is benchmarked against DFPT for eight known superconductors (Table I), showing order-of-magnitude agreement at far lower computational cost. Applying thresholds Ξ > 15 meV and χ > 125 meV·u^{1/2} to 8,833 nonmagnetic metals in the MattKeyBond database gives 260 candidates; DFPT verification leaves 150 dynamically stable compounds, of which 99 have Tc > 2 K and 34 have Tc > 10 K. The paper further interprets the final candidates as following two routes to high-Tc superconductivity: metallized covalent σ-bonding and Fermi-level DOS accumulation.","tokens_in":16312,"tokens_out":7042,"duration_ms":66513,"significance":"If the central claims hold, SEPR would be a valuable low-cost, physically interpretable pre-filter for superconductor discovery in large chemical databases. The paper's strengths include the transparent decomposition of Ξ and χ into atom- and direction-resolved contributions, the DFPT benchmarks for eight systems at dramatically lower cost, and the open discussion of the method's systematic bias in Sec. V. The candidate list itself is DFPT-verified, so the 34 candidates are not mere artifacts of the screening metrics. However, the screening metrics rely on a single universal decay length and on absolute thresholds calibrated on known superconductors; the paper does not quantify how the documented family-dependent bias affects the final candidate list. This makes the 'semi-quantitative' cross-family claim currently under-supported, though the issue is addressable in revision.","major_comments":[{"comment":"The universal BA decay length L_A = 1.2 Å is load-bearing. Eq. (7) gives dH_ab/du ≈ -(R·e/|R|)/L · H_ab, so Ξ and χ scale approximately as 1/L. Sec. V and Fig. 4 report that alkali/alkaline-earth systems have effective L ≈ 3.6 Å and 2p-covalent systems L ≈ 0.8 Å. Thus SEPR overestimates the EPC response by about 3× for alkali-containing metals and underestimates it by about 1.5× for 2p-covalent metals. Because the thresholds Ξ > 15 meV and χ > 125 meV·u^{1/2} are absolute, this family bias directly changes the composition of the 260-compound pool and hence the 34 candidates. The paper acknowledges the bias but does not quantify its impact on the final list. Please provide a sensitivity analysis, e.g., re-running the screen with L = 0.8 Å and 3.6 Å or with element-dependent L, and state which candidates survive. Without this, the claim of a semi-quantitative EPC scale across broad chemica","section":"Sec. II C and Sec. V, Eq. (7)"},{"comment":"The criteria Ξ > 15.0 meV and χ > 125.0 meV·u^{1/2} are called 'benchmarked', but the benchmarking procedure is not reported. It is unclear whether these cutoffs were selected to include the known superconductors in Table I; if so, the screening success is partly in-sample. The absence of a threshold-sensitivity analysis makes it impossible to assess how robust the 260-compound pool is. I ask for the calibration data, a precision/recall analysis against a held-out set, or at least a demonstration that moderate threshold shifts do not substantially change the final conclusions.","section":"Sec. IV, threshold selection"},{"comment":"The main validation beyond the eight benchmarks is Fig. 4, where SEPR and DFPT values of Ξ for 150 screened compounds are described as 'clearly positively correlated'. The figure shows substantial family-dependent scatter, but no correlation coefficient, mean absolute error, or fit is reported. Since the paper's central claim is semi-quantitative agreement, please quantify the correlation and the family-resolved errors (alkali/alkaline-earth, 2p-covalent, transition-metal) so that the reader can judge the strength of the validation.","section":"Sec. V, Fig. 4"}],"minor_comments":[{"comment":"The phrase 'calculated Tc > 10 K after DFPT verification' should be reconciled with Table II, which lists T_c (exp.) for the experimentally known superconductors. Please report DFPT-computed T_c for all 34 candidates, not only for the newly identified ones.","section":"Abstract and Table II"},{"comment":"The legend mentions 'computationally difficult' cases, but this category is not defined or discussed in the text. Please clarify what makes a compound computationally difficult.","section":"Fig. 3 legend"},{"comment":"The table note says some materials are 'only slightly dynamically unstable at ambient pressure but become stable under applied pressure', yet they are labeled dynamically stable candidates. Specify the pressure and other conditions for each entry.","section":"Table II note"},{"comment":"The auxiliary matrices G, F, and cM are defined only in the Supplemental Material. To make the main text self-contained, provide at least their physical definitions or a brief derivation sketch.","section":"Eqs. (4)-(5)"},{"comment":"The definition of V_ep should be stated explicitly as V_ep = (Ξ^2/N*)^{1/3}; the text's reference to a 'cubic form' as a dimensional normalization is confusing.","section":"Sec. VI, Eq. (8)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses an important practical problem and the authors are commendably transparent about the systematic bias in their method. However, the central screening claim needs a robustness analysis with respect to the universal decay length and threshold selection. The bias documented in Sec. V is internally acknowledged, so the required revision is feasible. I also suggest the authors clarify in the revision what is new relative to Ref. [29] on sigma-bond screening, to avoid overlap concerns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper proposes a low-cost screening framework for phonon-mediated superconductors, and the core idea is sound. Instead of doing DFPT for every candidate, it estimates the electron-phonon coupling scale from a static orbital Hamiltonian via three approximations: isotropic averaging over phonons, keeping only two-center hopping derivatives, and assuming an exponential bond-length decay with a single universal length L=1.2 Å. That last assumption is the load-bearing one, and the paper is honest that it introduces a family-dependent bias. The stress-test note is correct: effective decay lengths vary from about 0.8 Å for 2p covalent systems to 3.6 Å for alkali/alkaline-earth compounds, so Ξ and χ can be off by roughly a factor of 3. Since the screening thresholds are absolute, this will systematically admit or exclude compounds based on elemental identity, not just intrinsic coupling strength. The paper acknowledges this in Sec. V and visualizes it in Fig. 4, but does not correct it. That is a real limitation for the claim of semi-quantitative comparability across all 8,833 metals.\n\nWhat is genuinely new and well done: the SEPR framework itself, the two screening quantities Ξ and χ, and the application to 36,377 compounds with DFPT verification of 34 surviving candidates. The benchmark against DFPT on eight representative superconductors — MgB2, BaBiO3, Nb3Ge, Nb, Al, SH3, LaH10 — shows the right order of magnitude, and the atom-resolved decomposition correctly identifies the B-B σ bonds in MgB2 and the Bi-O network in bismuthate. The two-route taxonomy (σ-bond metallization vs. DOS accumulation) is physically reasonable and provides a useful organizing picture for future searches.\n\nThe soft spots beyond the decay length are manageable but worth naming. The thresholds Ξ>15 meV and χ>125 meV·u^1/2 are benchmarked on known superconductors, which introduces a degree of circularity; the candidate pool is partially tuned to include the benchmark set. No code or data are provided to reproduce the screening, which makes independent validation harder. The paper also leans on an unpublished database paper (ref 30) for the MattKeyBond Hamiltonians. These are not fatal, but they mean the exact 34-candidate list should be treated as a set of predictions that need independent confirmation, not as a definitive catalog.\n\nWho is this for? Researchers working on high-throughput screening for superconductors, especially those who want a physically interpretable descriptor rather than a black-box machine-learning model. The paper deserves a serious referee — it is a new method with a clear derivation, honest benchmarking, and an open list of testable candidates. My recommendation is to send it to peer review, with the expectation that the authors will address the decay-length bias (e.g., element-dependent L or a two-parameter fit) and, ideally, release the code and screening data. The central idea holds up; the quantitative screening pipeline needs tightening.","headline":"SEPR is a genuinely cheap and physically transparent EPC screening heuristic, but the universal decay length and threshold tuning make the candidate list a starting point, not a discovery.","tokens_in":16822,"tokens_out":2834,"would_cite":false,"duration_ms":30774,"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":"A low-cost screening scheme estimates electron-phonon coupling directly from static orbital Hamiltonians, and applied to over 36,000 compounds it finds 34 dynamically stable superconducting candidates with calculated transition temperatures","keywords":["electron-phonon coupling","superconductor screening","static orbital Hamiltonian","SEPR","density functional perturbation theory","high-throughput materials discovery","σ-bond metallization","McMillan-Allen-Dynes"],"falsifier":"Measure the superconducting transition of a predicted candidate that already has synthetic records, such as B13C2 (calculated Tc ≈ 41 K), at ambient pressure; if no superconductivity above 10 K is found, the screening ranking is called into question. Alternatively, compute DFPT Ξ for a series of 2p-covalent and alkali-based compounds and check whether the uniform-L SEPR values reproduce the ordering; large deviations would falsify the universal decay-length assumption.","tokens_in":15872,"feed_emoji":"⚡","tokens_out":5035,"duration_ms":46893,"temperature":0.7,"pith_summary":"The paper claims that the electron-phonon coupling strength of a material can be estimated directly from its static orbital Hamiltonian, without computing phonon perturbations explicitly. It does this through three approximations—an averaged phonon frequency, bond-only hopping changes, and an exponential bond-length decay model—producing two screening quantities, Ξ and χ. Applied to 36,377 compounds, the scheme flags 260 candidates, of which 150 are dynamically stable and 34 have calculated Tc above 10 K after density functional perturbation theory verification. If correct, this makes high-throughput superconductor discovery much cheaper and reveals two physical routes to high Tc: metallized covalent σ-bonds and Fermi-level density-of-states accumulation.","feed_headline":"34 high-Tc superconductor candidates found via static orbital filter","feed_subtitle":"Skipping costly phonon runs, the screen ranks 36,000 compounds and flags 34 candidates for experimental follow-up.","key_machinery":"The central object is the static electron-phonon response (SEPR) scheme, which approximates the phonon-induced Hamiltonian derivative dH/du from static interatomic hopping integrals. Three approximations carry the argument: the dynamic isotropic approximation (mode-dependent phonon frequencies replaced by an average), the double-center approximation (only bonds directly attached to the displaced atom contribute), and the bonding-attractivity (BA) model (hopping integrals decay exponentially with bond length, with a uniform decay length L = 1.2 Å). These yield two screening quantities: Ξ, the mass-weighted EPC-related scattering scale, and χ, the mass-independent electronic deformation respon","core_discovery":"The central claim is that the static electron-phonon response (SEPR) framework, built from static orbital Hamiltonians, reproduces the scale and material trends of electron-phonon coupling at a fraction of the cost of density functional perturbation theory. Using the mass-weighted scattering scale Ξ and the mass-independent deformation response χ, with thresholds Ξ > 15.0 meV and χ > 125.0 meV·u^1/2, the screen reduces 8,833 nonmagnetic metals to 260 candidates; subsequent DFPT phonon and EPC calculations yield 150 dynamically stable systems, 99 with Tc > 2 K, and 34 with Tc > 10 K. The paper interprets the survivors as showing two recurring routes to high Tc and a systematic competition bet","pith_inferences":["If the uniform decay length were replaced by element-resolved decay lengths (e.g., about 0.8 Å for 2p elements and about 3.6 Å for alkali/alkaline-earth elements), the screening thresholds and candidate list could shift markedly; a quick test would be to recompute Ξ and χ with family-dependent L values.","The predicted dynamically unstable high-Ξ compounds, though excluded here, may become superconducting under pressure or with doping, exactly as the high-pressure hydrides did; this makes them a natural target for follow-up synthesis.","The SEPR decomposition into atom- and direction-resolved contributions could be used to identify the specific phonon modes that dominate coupling, potentially guiding isotope-substitution or strain experiments.","The same static-response logic could be paired with machine-learned or surrogate Hamiltonians to extend the screen far beyond the present 36,377-compound database, should such Hamiltonians become available."],"forward_implications":["EPC screening no longer requires explicit phonon perturbation at the initial stage, cutting evaluation time from hours to minutes.","The screen identifies 34 dynamically stable candidates with calculated Tc > 10 K, including known superconductors and previously untested boron/carbon-rich phases.","Two distinct routes to high Tc emerge: metallized covalent σ-bonds (larger per-state scattering V_ep) and Fermi-level DOS accumulation (larger N*(εF)).","The empty upper-right region of the V_ep–N*(εF) plane indicates a systematic competition between strong EPC and lattice stability, so optimal candidates should lie near the stability boundary.","Systematic bias analysis shows alkali/alkaline-earth systems tend to be overestimated and 2p-covalent systems underestimated, attributable to the uniform decay length."],"fun_headline_variants":["Static orbital screen finds 34 high-Tc superconductor candidates","Cheap static filter flags 34 superconductors with Tc > 10 K","Skip costly phonon runs: static orbitals predict 34 high-Tc superconductors","Low-cost static method uncovers 34 superconductor candidates"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The result rests on the assumption that every interatomic hopping integral decays exponentially with bond length with the same decay length L = 1.2 Å; if real decay lengths vary across chemical families (as the paper's own bias analysis suggests), the estimated Ξ and χ values—and hence the candidate list—are systematically skewed.","fun_headline_variants_meta":{"raw":{"variants":["Static orbital screen finds 34 high-Tc superconductor candidates","Cheap static filter flags 34 superconductors with Tc > 10 K","Skip costly phonon runs: static orbitals predict 34 high-Tc superconductors","Low-cost static method uncovers 34 superconductor candidates"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000197,"raw_usage":{"total_tokens":1188,"prompt_tokens":718,"completion_tokens":470,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":462,"completion_tokens_details":{"reasoning_tokens":391}},"tokens_in":462,"tokens_out":470,"duration_ms":4605,"temperature":1.0,"reasoning_tokens":391,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T07:35:22.675753+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the superconducting transition of a predicted candidate that already has synthetic records, such as B13C2 (calculated Tc ≈ 41 K), at ambient pressure; if no superconductivity above 10 K is found, the screening ranking is called into question. Alternatively, compute DFPT Ξ for a series of 2p-covalent and alkali-based compounds and check whether the uniform-L SEPR values reproduce the ordering; large deviations would falsify the universal decay-length assumption.","supporting_citations":[],"review_version":1}