{"id":"c4b9ad87-85c9-4ab9-9726-d8de8df9ee6d","arxiv_id":"2411.13330","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Cubic NaK2Sb and hexagonal Na2KSb, two predicted polymorphs, are computed to be indirect-gap semiconductors with near-infrared absorption and weak excitonic effects, favorable for photocathode applications.","lead":"Researchers calculated the electronic and optical properties of two predicted crystal forms of sodium-potassium-antimony, a material used in photocathodes in particle accelerators. Both forms absorb near-infrared light with small exciton binding, suggesting they would not hurt photocathode performance if present in samples.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Internal inconsistency in NaK2Sb: reported 0.81 eV indirect gap vs 0.64 eV BSE peak plus 65 meV binding implies a 0.705 eV direct Gamma transition.","rationale":"The most load-bearing claim is the quantitative electronic structure of the two predicted crystals, because the photocathode conclusion follows directly from those numbers. The reader's weakest assumption concerns whether the OQMD polymorphs can be realized or appear in samples; that is relevant to the applied conclusion but does not test the internal correctness of the computed properties. The inconsistency I identify is internal to the paper and checkable from the released data. For NaK2Sb, the reported fundamental gap, first optical peak, and binding energy cannot all be true simultaneously. If the direct Gamma transition is actually 0.705 eV, the fundamental gap is direct and 0.1 eV smaller than the abstract states, and the indirect-gap characterization and the 0.81 eV value in the abstract would need correction. This would not necessarily destroy the photocathode suitability, but it changes the central quantitative message. Thus the verdict remains conditional: the paper should not be accepted as is, but the open data make the discrepancy easy to resolve. I agree only partially with the reader: the structural-relevance concern is valid but secondary, while the internal inconsistency is more urgent and more concrete.","tokens_in":14246,"tokens_out":9778,"duration_ms":99888,"concrete_test":"Retrieve the open data (Zenodo DOI 10.5281/zenodo.14100233) and directly evaluate for cubic NaK2Sb: (i) the G0W0 band energy difference CBM(Gamma) - VBM(Gamma); (ii) the lowest BSE eigenvalue E_1; (iii) the exciton binding energy E_b defined relative to the corresponding independent-QP transition at Gamma. Check whether E_1 + E_b equals the direct QP gap. If it equals ~0.705 eV, the reported 0.81 eV gap or the VBM-L assignment is wrong; if it equals ~0.81 eV, the reported 0.64 eV peak or 65 meV binding is wrong. A simpler rerun with the published settings and an 8x8x8 BSE grid should reproduce one of the two scenarios and settle which number is in error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sections IV B and IV C of the manuscript contain an internal numerical inconsistency for cubic NaK2Sb. Section IV B reports a G0W0 fundamental (indirect) gap of 0.81 eV, with the VBM at L and the CBM at Gamma. Section IV C reports the first bright BSE excitation at 0.64 eV, with binding energy 65 meV, and explicitly states that this excitation stems from transitions between the topmost valence states at Gamma and the CBm. If the QP vertical transition at Gamma has energy E_QP(Gamma), then E_QP(Gamma) = 0.64 eV + 0.065 eV = 0.705 eV. This is ~0.1 eV below the claimed minimum gap of 0.81 eV. Since the CBM is at Gamma and the VBM is at L, the indirect gap must be less than or equal to the direct Gamma gap; a 0.705 eV Gamma transition would place the VBM at Gamma and make the fundamental gap direct at 0.705 eV, contradicting the stated VBM-at-L character. Thus at least one of (0.81 eV gap, 0.64 eV excitation, 65 meV binding) is incorrect or misreported for NaK2Sb. The abstract and conclusion rely on these precise values for the near-IR onset and the photocathode assessment, so this discrepancy is load-bearing. No analogous contradiction appears for Na2KSb (0.70 eV gap vs 0.64 eV + 50 meV binding is consistent within a few meV), which makes the NaK2Sb case stand out as a likely error rather than a method-wide bias.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents first-principles calculations (DFT-PBEsol, G0W0, and BSE) of two computationally predicted Na-K-Sb polymorphs: cubic NaK2Sb and hexagonal Na2KSb. The authors find that both crystals have indirect fundamental gaps (0.81 eV for NaK2Sb, 0.70 eV for Na2KSb) with the CBM at Γ and the VBM at L/M, respectively, and that the direct gap at Γ is only slightly larger. Optical spectra from BSE show near-infrared absorption onsets at 0.64 eV in both materials, with exciton binding energies of about 50–100 meV. On this basis, the authors argue that these phases are suitable photocathode materials and that their presence in polycrystalline samples would not be detrimental.","tokens_in":14547,"tokens_out":6253,"duration_ms":59718,"significance":"If the reported values are correct, the paper provides useful quantitative predictions for an emerging photocathode material class, complementing prior studies of the experimentally known phases. The use of all-electron G0W0+BSE is state-of-the-art, and the data are openly available, which strengthens reproducibility. The main significance lies in the prediction of near-infrared response with low exciton binding energies, which is relevant for accelerator applications. However, an internal inconsistency in the NaK2Sb numbers and the lack of stability and convergence analysis currently undermine confidence in the quantitative conclusions.","major_comments":[{"comment":"For cubic NaK2Sb, the reported data are internally inconsistent. §IV B states a G0W0 fundamental (indirect) gap of 0.81 eV with the VBM at L and the CBM at Γ. §IV C reports the first bright BSE excitation at 0.64 eV with a binding energy of 65 meV, arising from vertical transitions between the topmost valence states at Γ and the CBm. These numbers imply a quasiparticle vertical transition at Γ of 0.64 eV + 0.065 eV = 0.705 eV, which is smaller than the claimed fundamental gap of 0.81 eV. Because the CBM is at Γ, the direct Γ transition provides an upper bound on the fundamental gap; a 0.705 eV direct gap would place the VBM at (or very near) Γ and make the fundamental gap direct, contradicting the stated VBM-at-L character. At least one of the three reported values (gap, excitation energy, binding energy) is incorrect or mislabeled. This discrepancy is load-bearing because the abstract and conclusions rest on these precise near-infrared onset values and on the indirect-gap characterization.","section":"§IV B and §IV C"},{"comment":"The conclusion that the presence of cubic NaK2Sb and hexagonal Na2KSb in polycrystalline samples 'is not detrimental' depends on the assumption that these computationally predicted phases are actually realizable as metastable polymorphs under synthesis conditions. The paper provides no thermodynamic or kinetic evidence for their viability; §IV A only reports the OQMD structure parameters, and the stability analysis of related multi-alkali antimonides in Ref. [30] is not extended to these polymorphs. Without such evidence (e.g., formation energies relative to the experimentally known phases, phonon calculations, or nucleation arguments), the applied conclusion is unsupported. The authors should either provide a stability analysis or soften the conclusion to a conditional statement.","section":"§V and §IV A"},{"comment":"No convergence tests are presented for the G0W0 and BSE calculations. The quantitative claims (fundamental gaps of 0.81 and 0.70 eV, exciton binding energies of 50–100 meV) are sensitive to the k-mesh, the number of empty states, and the BSE transition space (3 valence/9 conduction bands for NaK2Sb; 6/12 for Na2KSb). In particular, the G0W0 self-energy is computed via analytic continuation, an approximation whose accuracy should be benchmarked. The authors should demonstrate convergence of the reported quantities with respect to these parameters, or at least quantify the expected uncertainty, before the meV-level numbers are taken at face value.","section":"§III"}],"minor_comments":[{"comment":"There is a typo: 'band strcturess' should be 'band structures'.","section":"§IV B"},{"comment":"In the description of the hexagonal Na2KSb spectrum, it would be helpful to explicitly state that the out-of-plane component is shown in Fig. 5b and the in-plane component in Fig. 5a, to avoid ambiguity.","section":"§IV C"},{"comment":"A compact summary table comparing the computational parameters with those of Ref. [25] would make the claimed overlap more transparent.","section":"§III"},{"comment":"The paper does not discuss the possible influence of spin-orbit coupling on the Sb p-derived valence bands; given the meV-level quantitative claims, the authors should at least justify its neglect.","section":"§IV B"},{"comment":"The sentence 'we are confident that this work may stimulate research in this direction' is informal; consider a more neutral formulation.","section":"§V"}],"recommendation":"major_revision","confidential_remarks":"The internal inconsistency for NaK2Sb (major comment 1) is the key obstacle. If the authors can reconcile the G0W0 gap with the BSE excitation and binding energy, the paper's central claims may survive. The other major comments (stability analysis, convergence) are also addressable. The paper is likely publishable after major revision, but the inconsistency must be resolved rather than merely discussed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, this is the first G0W0+BSE study of the two computationally predicted polymorphs cubic NaK2Sb and hexagonal Na2KSb, and the structural and spectral analysis is careful and consistent with the group’s earlier work on the stable phases. Second, there is a clear internal inconsistency in the NaK2Sb numbers that the authors need to fix before this can be used.\n\nThe paper reports an indirect G0W0 gap of 0.81 eV for cubic NaK2Sb (VBM at L, CBM at Gamma), and then reports the first bright BSE excitation at 0.64 eV with a binding energy of 65 meV, explicitly stating it comes from vertical transitions at Gamma between the topmost valence state and the CBm. That implies a QP direct gap at Gamma of 0.705 eV, which contradicts the claimed 0.81 eV indirect gap (the direct gap must be larger than the indirect gap). Either the gap, the peak energy, the binding energy, or the transition assignment is wrong. The abstract and conclusion lean on these values for the photocathode suitability claim, so this is load-bearing. For Na2KSb the analogous numbers (0.70 eV gap, 0.64 eV peak, ~50 meV binding) are consistent within a few meV, so this looks like a specific error rather than a systematic bias.\n\nThe good parts: the methodology is standard and well documented (PBEsol, G0W0, BSE with the exciting code), the comparison with the experimentally known phases from Ref. [25] is appropriate, and the orbital character analysis is useful. The paper honestly notes the polymorphs are not yet experimentally confirmed. The main soft spots beyond the inconsistency are the absence of convergence tests (k-mesh, empty states, analytic continuation), the limited BSE transition space (3 valence/9 conduction bands for NaK2Sb), and the leap from ideal bulk crystals to the claim that the presence of these phases in polycrystalline samples is not detrimental. That last step needs thermodynamic or experimental evidence.\n\nWho is this for: people working on multi-alkali antimonide photocathodes will find the near-IR spectra and band structure useful once the numbers are corrected. As written, the NaK2Sb inconsistency undermines the quantitative claims. A serious referee should see this, but it should not be accepted without revision. I would not cite it in its current form.","headline":"First many-body characterization of two predicted Na-K-Sb polymorphs, but a numerical inconsistency in the NaK2Sb gap/BSE values makes the central claim unreliable as written.","tokens_in":15080,"tokens_out":4484,"would_cite":false,"duration_ms":45472,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Two computationally predicted Na-K-Sb phases are calculated to be suitable photocathodes.","keywords":["photocathodes","multi-alkali antimonides","Na-K-Sb","GW approximation","Bethe-Salpeter equation","excitons","density functional theory","near-infrared absorption"],"falsifier":"Synthesize Na-K-Sb films under standard co-deposition conditions and look for an infrared absorption onset near $0.6$ to $0.7$ eV and for diffraction signatures of cubic NaK$_2$Sb or hexagonal Na$_2$KSb; their absence would undermine the prediction. A cheaper falsifier is a phonon calculation: imaginary frequencies in the phonon dispersion of either structure would show it is not even metastable.","tokens_in":14041,"feed_emoji":"🔆","tokens_out":6608,"duration_ms":62515,"temperature":0.7,"pith_summary":"This paper sets out to show that two computationally predicted sodium-potassium-antimonide phases, cubic NaK$_2$Sb and hexagonal Na$_2$KSb, have the electronic and optical properties needed for photocathodes. Both are indirect-gap semiconductors with fundamental gaps of $0.81$ eV and $0.70$ eV that sit very close to the direct gap at $\\Gamma$, and both absorb strongly in the near-infrared with exciton binding energies of roughly $50$ to $100$ meV. Because these crystals are grown as polycrystalline films that often contain multiple phases, establishing that these polymorphs are not optically harmful matters for interpreting and improving photocathode performance. The paper concludes that if such phases appear in samples, their presence should not be detrimental.","feed_headline":"Predicted Na-K-Sb phases suit infrared photocathodes","feed_subtitle":"Quantum calculations give 0.81 and 0.70 eV gaps with weak exciton binding, so these polytypes likely don't hurt emitters.","key_machinery":"The argument is carried by a density-functional-theory plus many-body-perturbation-theory workflow: PBEsol DFT produces the ground-state band structure, the $G_0W_0$ approximation supplies quasiparticle corrections that raise the gap by about $0.43$ eV in both crystals, and the Bethe-Salpeter equation (BSE) yields the optical absorption spectrum and exciton binding energies. The central objects are the quasiparticle band structures, the imaginary part of the macroscopic dielectric function with and without excitons, and the exciton weights that map each absorption peak onto specific transitions between Sb $p$-dominated valence states and Sb-Na $s$-hybridized conduction states. Comparison with earlier same-level-of-theory results for the experimentally known phases (cubic Na$_2$KSb and hexagonal NaK$_2$Sb) anchors the prediction: the predicted polymorphs have smaller gaps, lower absorption onsets, and weaker exciton binding.","core_discovery":"The central claim is that cubic NaK$_2$Sb and hexagonal Na$_2$KSb, two polymorphs taken from the Open Quantum Materials Database, are viable photocathode candidates. At the $G_0W_0$ level, NaK$_2$Sb has an indirect fundamental gap of $0.81$ eV and Na$_2$KSb of $0.70$ eV, with the conduction-band minimum at $\\Gamma$ and the valence-band maximum at $L$ (NaK$_2$Sb) or $M$ (Na$_2$KSb), so the indirect and direct gaps at $\\Gamma$ are nearly equal. Solving the Bethe-Salpeter equation on top of the quasiparticle band structure gives optical absorption dominated by near-infrared peaks starting around $0.64$ eV, with exciton binding energies of $50$ to $100$ meV and no strong excitonic reshaping of the lowest-energy features. The paper argues that these characteristics align with the requirements for efficient vacuum electron sources and that the presence of these phases in polycrystalline samples would not degrade photocathode performance.","pith_inferences":["A natural test of the metastability premise is to compute phonon dispersion curves for both phases; imaginary phonon modes would directly contradict the 'realizable polymorph' assumption.","One could extend the paper's claim by estimating the photoemission threshold and quantum efficiency from the computed dielectric functions, which the authors do not do.","Since the predicted polymorphs have markedly smaller gaps than the known phases, their presence should shift the measured absorption edge to lower energy; growing Na-K-Sb films and measuring the infrared onset would be an experimental check.","The bulk OQMD structures ignore surface and interface effects relevant to thin-film growth, so surface calculations would be a natural follow-up to strengthen the photocathode conclusion."],"forward_implications":["If the predictions hold, cubic NaK$_2$Sb and hexagonal Na$_2$KSb should be considered alongside the known phases when modeling photoemission from Na-K-Sb photocathodes.","The near-infrared absorption onset near $0.6$ eV means these polymorphs would respond to infrared drive lasers, supporting the push toward infrared-operated electron sources.","Exciton binding energies of only $50$ to $100$ meV imply that photoelectrons come from weakly bound excitations, favorable for efficient room-temperature photoemission.","In both polymorphs the lowest-energy excitation is optically active, unlike in hexagonal NaK$_2$Sb where the first excitation is dark, so these phases add allowed transitions at the band edge.","The band character is essentially the same as in the known phases (Sb $p$ valence, mixed Sb-Na $s$ conduction), indicating that the electronic fingerprint of Na-K-Sb films is robust to polytypism."],"supporting_citations":[{"why":"Earlier same-level-of-theory study of the experimentally known cubic Na$_2$KSb and hexagonal NaK$_2$Sb, providing the reference data and direct comparison.","marker":"[25]"},{"why":"Open Quantum Materials Database that supplies the two predicted crystal structures used throughout the paper.","marker":"[69]"},{"why":"Hedin's GW approximation, the method behind the quasiparticle corrections to the band gap.","marker":"[54]"},{"why":"Bethe-Salpeter equation, the formalism used for the optical absorption and exciton binding energies.","marker":"[55]"},{"why":"Prior BSE calculation on CsK$_2$Sb, providing the comparison value for exciton binding energy.","marker":"[21]"},{"why":"Review of bright electron source requirements that frames the relevance of infrared photocathodes.","marker":"[11]"},{"why":"The exciting all-electron code in which all DFT and MBPT calculations are implemented.","marker":"[62]"}],"fun_headline_variants":["Na-K-Sb polymorphs predicted as IR photocathodes","Predicted NaK2Sb and Na2KSb suit photocathodes","Calculated Na-K-Sb phases show near-IR gaps for photocathodes","Computed Na-K-Sb polytypes: near-IR photocathode candidates","NaK2Sb (0.81 eV) and Na2KSb (0.70 eV) predicted for IR photocathodes"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the two OQMD structures are realizable metastable phases that can actually form within polycrystalline Na-K-Sb samples; if they cannot be synthesized or do not nucleate under deposition conditions, the paper's applied conclusion loses its object.","fun_headline_variants_meta":{"raw":{"variants":["Na-K-Sb polymorphs predicted as IR photocathodes","Predicted NaK2Sb and Na2KSb suit photocathodes","Calculated Na-K-Sb phases show near-IR gaps for photocathodes","Computed Na-K-Sb polytypes: near-IR photocathode candidates","NaK2Sb (0.81 eV) and Na2KSb (0.70 eV) predicted for IR photocathodes"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000757,"raw_usage":{"total_tokens":3423,"prompt_tokens":1065,"completion_tokens":2358,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":681,"completion_tokens_details":{"reasoning_tokens":2262}},"tokens_in":681,"tokens_out":2358,"duration_ms":21539,"temperature":1.0,"reasoning_tokens":2262,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:32:14.580439+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Synthesize Na-K-Sb films under standard co-deposition conditions and look for an infrared absorption onset near $0.6$ to $0.7$ eV and for diffraction signatures of cubic NaK$_2$Sb or hexagonal Na$_2$KSb; their absence would undermine the prediction. A cheaper falsifier is a phonon calculation: imaginary frequencies in the phonon dispersion of either structure would show it is not even metastable.","supporting_citations":[{"cited_title":"Amador, H.-D","cited_arxiv_id":null,"evidence_quote":"Earlier same-level-of-theory study of the experimentally known cubic Na$_2$KSb and hexagonal NaK$_2$Sb, providing the reference data and direct comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Open Quantum Materials Database that supplies the two predicted crystal structures used throughout the paper."},{"cited_title":"Cocchi, S","cited_arxiv_id":null,"evidence_quote":"Prior BSE calculation on CsK$_2$Sb, providing the comparison value for exciton binding energy."},{"cited_title":"Musumeci, J","cited_arxiv_id":null,"evidence_quote":"Review of bright electron source requirements that frames the relevance of infrared photocathodes."},{"cited_title":"Gulans, S","cited_arxiv_id":null,"evidence_quote":"The exciting all-electron code in which all DFT and MBPT calculations are implemented."}],"review_version":1}