{"id":"d3c2beb4-0a7f-4632-8b4b-ef4f5a81ad6c","arxiv_id":"1908.08334","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A high-throughput DFT and GW screen of exfoliable monolayers finds 13 candidate quantum spin Hall insulators, including a new Kane-Mele material Pd2HgSe3.","lead":"Using computer simulations, researchers screened over 1,800 exfoliable two-dimensional materials and found 13 candidate quantum spin Hall insulators, including the newly predicted Pd2HgSe3. The list gives experimentalists a short menu of promising atomically thin materials for low-dissipation electronics and spintronics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 13-candidate count and ~1% abundance include TiNI, a documented G0W0 false positive, plus several candidates with no G0W0 check; the abundance claim is not yet supported as stated.","rationale":"The reader's weakest assumption identifies the same load-bearing issue: PBE-level Z2 assignments for candidates not re-examined at G0W0, with TiNI as an explicit false positive. My read confirms this and adds two concrete aggravating details. First, TiNI is not merely discussed as a cautionary tale; it is still listed in Table I among the candidates, so the stated count of 13 includes a material the paper itself classifies as trivial at G0W0. Second, the count itself is numerically inconsistent: Table I has 14 rows, the text says 13, and the Fig. 2 caption says 13 candidates plus Pd2HgSe3. These inconsistencies make the headline \"13 monolayers\" and the derived \"around 1% abundance\" fragile. I agree with the reader that the screening protocol is otherwise sound and the G0W0-validated new materials, AsCuLi2 and Pd2HgSe3, are credible contributions. The appropriate remedy is the one the reader already proposed: make the validation status explicit, remove documented false positives from the count, and reframe the abundance statistic with a correct denominator and an acknowledgement of PBE-only uncertainty. Since the reader's conditional verdict already encodes exactly this requirement, no change in verdict is needed.","tokens_in":16581,"tokens_out":6413,"duration_ms":67410,"concrete_test":"Produce a corrected version of Table I with a G0W0/independent many-body validation status column, remove TiNI from the topological count, and recompute the abundance using as denominator the actual number of PBE band insulators identified in the screening funnel rather than the 1306 relaxed monolayers. If the corrected fraction drops below ~0.5% or the denominator changes by more than a factor of two, the \"around 1%\" claim should be revised in the abstract and conclusions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central abundance claim is load-bearing on the integrity of the candidate count, and that count is internally inconsistent. Table I lists 14 entries, while the text says 13 QSHI candidates are listed and the Fig. 2 caption describes \"13 candidates ... to which we add palladium jacutingaite.\" More importantly, one listed candidate, TiNI, is explicitly found trivial at G0W0 (Table I: inversion strength -705 meV), yet it remains in the candidate table and in the count. For seven further entries (TaIrTe4, TaRhTe4, NbIrTe4, Cu2Te3Ti, In2ZnS4, ZrBr, ZrCl) Table I gives no G0W0 inversion strength, so their Z2 labels rest on PBE-SOC. The paper itself shows with TiNI that PBE can produce a false positive even with a sizable PBE inversion strength (141 meV), well above the 20 meV screening cutoff. Thus the numerator of the \"around 1%\" estimate contains at least one known false positive and an unknown number of PBE-only assignments; with 13/1306 ≈ 1%, removing one or two false positives changes the headline figure materially. In addition, the denominator is all relaxed nonmagnetic monolayers (1306), not the number of band insulators, so the phrase \"in two-dimensional insulators\" is not directly supported by the reported ratio. The G0W0-validated new materials, notably AsCuLi2 and Pd2HgSe3, are not themselves invalidated by this concern, but the headline count and abundance statistic are.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a high-throughput first-principles screening of 1306 exfoliable monolayer materials for quantum spin Hall insulators (QSHIs). Using PBE-SOC band structures, Z2 invariants from Wannier charge centers, phonon stability checks via DFPT, strain engineering for direct-gap metals, and G0W0-with-SOC checks for five leading candidates, the authors identify a set of candidate QSHIs. The most notable new predictions are monolayer Pd2HgSe3, presented as a second Kane-Mele QSHI with a G0W0 inversion strength near 41 meV, and AsCuLi2, a new Bernevig-Hughes-Zhang-type prototype. The manuscript also reports that TiNI, previously proposed as a QSHI, is trivial at the G0W0 level, and it concludes that Z2 topological order has a relative abundance of around 1% among the screened exfoliable two-dimensional insulators.","tokens_in":16904,"tokens_out":4279,"duration_ms":44536,"significance":"If substantiated, this work would provide a valuable, systematically generated set of QSHI candidates and a useful methodological template for database-scale topological screening. Its strengths are the explicit first-principles computation of Z2 invariants (rather than descriptor-based proxies), the inclusion of phonon stability, the use of independent G0W0 quasiparticle corrections that are not fitted to the screening outcome, the honest reporting of a G0W0 false positive (TiNI), and the use of reproducible open-source codes and pseudopotential libraries. The two new material predictions, Pd2HgSe3 and AsCuLi2, are specific and falsifiable and are likely to be of interest to the experimental 2D-materials community. However, the headline candidate count and the associated abundance estimate are not currently supported in their stated form because the candidate table and text are internally inconsistent and because a large fraction of the candidates rest on PBE-level assignments that the paper itself shows can be reversed at G0W0.","major_comments":[{"comment":"The headline count of 13 QSHI candidates is internally inconsistent. The text states 'we find 13 QSHIs candidates... listed in Tab. I', but Table I lists 14 entries (Bi, Pt2HgSe3, Pd2HgSe3, TiNI, AsCuLi2, WTe2, MoTe2, TaIrTe4, TaRhTe4, NbIrTe4, Cu2Te3Ti, In2ZnS4, ZrBr, ZrCl). In addition, the Results paragraph on TiNI reports a G0W0 inversion strength of -705 meV and states explicitly that TiNI is a trivial insulator at G0W0, yet TiNI remains in the candidate table and in the count. This inconsistency directly affects the central claim and the 'around 1%' abundance estimate, and it must be corrected in a revised version.","section":"Results, Table I, and Abstract"},{"comment":"Only five candidates receive G0W0 validation, while seven entries in Table I (TaIrTe4, TaRhTe4, NbIrTe4, Cu2Te3Ti, In2ZnS4, ZrBr, ZrCl) have no G0W0 inversion strength. The TiNI case shows that a PBE-SOC inversion strength of 141 meV, well above the 20 meV screening cutoff, can nevertheless be reversed to a trivial assignment at G0W0. The PBE-only Z2 assignments should therefore be presented explicitly as tentative candidates, and the count of '13 QSHI candidates' and the deduced 1% abundance should be restricted to G0W0-validated cases or qualified with the number of PBE-only assignments.","section":"Methods (G0W0 subsection) and Results (TiNI paragraph)"},{"comment":"The abundance statement 'relative abundance of Z2 topological order in two-dimensional insulators of around 1%' is not directly supported by the reported ratio. The denominator is the 1306 relaxed non-magnetic monolayers, which includes metals and direct-gap metals, not only band insulators. The numerator counts candidates that include strain-induced insulators and at least one G0W0-trivial material. To support the stated claim, the authors should either compute the fraction with respect to the number of band insulators (or an otherwise clearly defined denominator) or rephrase the claim to describe abundance among the full set of non-magnetic exfoliable monolayers.","section":"Conclusions and Results (screening protocol)"}],"minor_comments":[{"comment":"The manuscript contains several typographical errors, including 'unpertubed', 'abudance', 'miminum', 'perfom', and 'hermaphrodite Wannier charge centers' (presumably 'hybrid Wannier charge centers'). These should be corrected.","section":"Throughout"},{"comment":"The TiNI paragraph gives a PBE inversion strength of 0.17 eV, while Table I lists 141 meV (0.141 eV) for the same quantity. Please reconcile this discrepancy.","section":"Results, TiNI paragraph and Table I"},{"comment":"The 20 meV inversion-strength cutoff and the 0.01 eV direct-gap-metal threshold are introduced without sensitivity analysis or justification. Since the abundance estimate depends on the candidate count, a brief statement on how the results change with these thresholds would strengthen the screening protocol.","section":"Methods, Eq. (2) and footnote 20"},{"comment":"The G0W0 inversion strengths obtained from the extrapolation in Eq. (1) are reported without uncertainty estimates; providing the fitted parameters or an error estimate would make the quoted values, especially the 41 meV result for Pd2HgSe3, more informative.","section":"Methods, Eq. (1)"}],"recommendation":"major_revision","confidential_remarks":"The core methodology and the specific new material predictions are sound and likely publishable after revision. The main issue is the inconsistency in the candidate count and the overreach of the abundance claim relative to the PBE-only assignments. These are fixable within the manuscript's scope, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a genuinely useful screening paper, and Pd2HgSe3 is likely the real prize — a second Kane-Mele QSHI with a G0W0 inversion strength of ~41 meV. But the headline numbers as written are not defensible. The text says 13 candidates, Table I lists 14 entries, and the 13 count includes TiNI, which the authors themselves show is trivial at G0W0. Seven more entries in the table have no G0W0 check at all, so their topological assignment rests on PBE-SOC alone. That's fine as a screening result, but it means the '13 candidates' and 'around 1% abundance' claims are not supported as stated. The denominator is also wrong for the phrasing: 1306 is the number of nonmagnetic monolayers that relaxed successfully, not the number of insulators. If you quote '1% of two-dimensional insulators,' you need insulators in the denominator.\n\nWhat the paper does well: the funnel is transparent and standard — Z2 from Wannier charge centers, phonons to confirm stability, strain tests for metals, and G0W0 for the five most interesting cases. The TiNI case is honestly reported, which is exactly the kind of calibration the community needs. AsCuLi2 is a plausible new BHZ prototype, and the Pd2HgSe3 prediction is concrete and falsifiable. The self-cited database and pseudopotential references are legitimate tools the authors built, not padding.\n\nThe fixes are not hard. Re-label the candidates by validation tier (G0W0-confirmed, PBE-only tentative, known false positive), recalculate the abundance using the actual number of insulators screened, and give error bars or convergence data for the inversion strengths. The arbitrary 20 meV cutoff and 0.01 eV DGM threshold are less worrying; they're standard screening choices and the paper is open about them. I'd also ask them to square the table count with the text, because that kind of internal inconsistency makes a reader wonder what else doesn't add up.\n\nMy take: the paper deserves a serious referee. The method is sound, the new materials are valuable, and the overclaim is correctable without new calculations. I'd desk-reject only if the journal doesn't want high-throughput screens, but for a topical or materials journal, send it out with clear instructions on the count and denominator.","headline":"Screening protocol is solid and Pd2HgSe3 is a real find, but the 13-candidate count and ~1% abundance claim are not supported as written.","tokens_in":17488,"tokens_out":3702,"would_cite":false,"duration_ms":35956,"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":"Thirteen exfoliable monolayers are quantum spin Hall insulator candidates, about 1% of the pool.","keywords":["quantum spin Hall insulators","Z2 topological order","two-dimensional materials","high-throughput screening","density functional theory","G0W0 approximation","spin-orbit coupling","exfoliable monolayers"],"falsifier":"Recompute the G0W0-with-spin-orbit $\\mathbb{Z}_2$ invariant and band inversion for the PBE-only candidates (TaIrTe4, TaRhTe4, NbIrTe4, ZrBr, ZrCl, Cu2Te3Ti, In2ZnS4); if any of them reverts to a trivial insulator as TiNI does, the candidate list and the 1% abundance claim need downward revision.","tokens_in":16363,"feed_emoji":"🔬","tokens_out":15299,"duration_ms":137914,"temperature":0.7,"pith_summary":"Using first-principles simulations, the paper screens 1,306 exfoliable monolayers drawn from a database of experimentally known layered compounds and asks which of them are quantum spin Hall insulators (QSHIs): two-dimensional insulators whose time-reversal-protected edge states can carry spin-momentum-locked current without dissipation. It identifies 13 candidate monolayers, with the newly synthesized Pd2HgSe3 presented separately as a second Kane-Mele-type example, that are non-magnetic, dynamically stable, and carry a non-trivial $\\mathbb{Z}_2$ invariant at the PBE-with-spin-orbit level, sometimes only after a few percent of strain. Standout new materials are AsCuLi2, a previously unknown structural prototype, and Pd2HgSe3, whose G0W0-calculated band inversion is about 41 meV. The screen concludes that the relative abundance of $\\mathbb{Z}_2$ topological order among two-dimensional insulators in this exfoliable set is around 1%, a benchmark useful because such materials are rare and promising for low-dissipation nanoelectronics and spintronics.","feed_headline":"1% of exfoliable monolayers host Z2 topological order","feed_subtitle":"A screen of 1,306 monolayers finds 13 quantum spin Hall candidates and two new material families.","key_machinery":"The machine that carries the argument is the screening funnel itself. Topological classification uses the $\\mathbb{Z}_2$ invariant obtained from hybrid Wannier charge centers, the evolution of Wannier centers across half the Brillouin zone, which the paper computes for every band insulator and, where needed, for direct-gap metals under strain. The robustness measure is the inversion strength, defined as the direct gap at the high-symmetry point where the band inversion occurs: at $K$ for Kane-Mele systems and at $\\Gamma$ for Bernevig-Hughes-Zhang systems. The final arbiter for the five most promising materials is the G0W0 approximation with spin-orbit coupling, extrapolated to an infinitely dense $\\mathbf{k}$-grid, because two-dimensional dielectric screening makes quasiparticle corrections slow to converge.","core_discovery":"The paper's central claim is that $\\mathbb{Z}_2$ topological order is not exotic within the pool of monolayer crystals that can be peeled from known layered compounds: of the 1,306 monolayers that survive the non-magnetic, structurally relaxed screening, a shortlist of 13 carries a non-trivial $\\mathbb{Z}_2$ invariant, and Pd2HgSe3 is added as a second Kane-Mele QSHI, for a relative abundance of roughly 1%. The result is obtained through a funnel: structural relaxation starting from experimental parent crystals, a band-insulator or direct-gap-metal selection (with isotropic strain of 1–3% allowed for the metals), the $\\mathbb{Z}_2$ invariant computed from hybrid Wannier charge centers, a magnetic-ground-state filter, and a phonon stability check. For five of the most interesting candidates the band inversion is recomputed at the G0W0 level with spin-orbit coupling; this confirms AsCuLi2 (inversion strength 169 meV) and Pd2HgSe3 (41 meV) as new QSHIs, while exposing TiNI as a PBE false positive that is trivial at G0W0 with a 0.7 eV direct gap. The authors read the 1% figure as the abundance of true two-dimensional bulk insulators whose entire occupied manifold is topologically nontrivial, as opposed to metals with well-defined interband gaps where non-trivial invariants coexist with gapless bulk.","pith_inferences":["The paper leaves the true abundance open: because only five of the shortlisted materials received G0W0 checks and one (TiNI) flipped to trivial, the 1% figure could creep below 1% if the unverified PBE-only candidates behave like TiNI.","The finding that every candidate has fewer than 12 atoms per unit cell and a third share the same space group suggests that future screens could pre-filter on structural motifs and cell size, trading a small risk of missed candidates for a large gain in cost.","For the strain-driven candidates (ZrBr, ZrCl, WTe2, MoTe2), a concrete testable route is to grow them on lattice-matched substrates that impose 1–3% isotropic strain and to measure the helical edge conductance predicted for the QSHI phase.","The 1% abundance refers to exfoliable monolayers from known layered crystals; a screen over broader classes of two-dimensional materials might give a different fraction, so the number is a property of this materials pool, not a universal constant."],"forward_implications":["Monolayer Pd2HgSe3 becomes a concrete experimental target: exfoliable, dynamically stable, and a Kane-Mele QSHI with a G0W0 inversion strength around 41 meV.","AsCuLi2 offers a new crystal prototype for spin-Hall physics, with a clean band inversion at $\\Gamma$ that strengthens at the G0W0 level (169 meV) compared with PBE (80 meV).","Small isotropic strain of 1–3% turns several screened structures into true QSHI insulators, so substrate-induced strain can be used to realize the topological phase in otherwise gapless or weakly gapped monolayers.","The 1% abundance estimate quantifies how likely a random exfoliable monolayer is to be a $\\mathbb{Z}_2$ topological insulator, giving future computational searches a statistical benchmark.","The TiNI case shows that PBE-level $\\mathbb{Z}_2$ assignments are provisional: candidates not yet re-examined at the G0W0 level should be treated as predictions pending quasiparticle checks."],"supporting_citations":[{"why":"Supplies the 1,825-monolayer database of experimentally known exfoliable compounds that the screen starts from, including the magnetic-ground-state screening used later.","marker":"[16]"},{"why":"Provides the hybrid-Wannier-centre method and its implementation for computing the Z2 invariant for every band insulator in the pool.","marker":"[22, 23]"},{"why":"Prediction of jacutingaite (Pt2HgSe3) as a Kane-Mele QSHI; the comparison baseline and the source of the quoted PBE inversion strength.","marker":"[33]"},{"why":"Documents false-positive and false-negative topological assignments in density functional theory, the justification for re-checking candidates at G0W0 and for the TiNI verdict.","marker":"[37]"},{"why":"Reports the synthesis and crystal structure of Pd2HgSe3, the material added to the database and identified as a second Kane-Mele QSHI.","marker":"[44]"},{"why":"Predicted the quantum spin Hall phase in 1T' transition-metal dichalcogenides and supplies the WTe2 and MoTe2 entries with their inversion strengths.","marker":"[13]"},{"why":"Predicted the MM'Te4 (Nb/Ta, Ir/Rh) monolayers as QSHIs, the previously known candidates reconfirmed in this screen.","marker":"[35]"},{"why":"Provides the density-functional perturbation theory used to compute phonon dispersions and enforce the dynamical-stability filter.","marker":"[24]"}],"fun_headline_variants":["13 of 1,306 exfoliable monolayers are quantum spin Hall","1% of exfoliable monolayers host Z2 topological order","Screen finds 13 new QSHIs in exfoliable monolayers","Z2 order in 1% of exfoliable monolayers","Pd2HgSe3: second Kane-Mele QSHI in exfoliable monolayers"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the PBE-with-spin-orbit $\\mathbb{Z}_2$ assignments, which were checked at the G0W0 level for only five of the shortlisted materials, stay correct for the candidates that were not re-examined; the paper itself shows that this premise failed for TiNI.","fun_headline_variants_meta":{"raw":{"variants":["13 of 1,306 exfoliable monolayers are quantum spin Hall","1% of exfoliable monolayers host Z2 topological order","Screen finds 13 new QSHIs in exfoliable monolayers","Z2 order in 1% of exfoliable monolayers","Pd2HgSe3: second Kane-Mele QSHI in exfoliable monolayers"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000885,"raw_usage":{"total_tokens":3923,"prompt_tokens":1146,"completion_tokens":2777,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":762,"completion_tokens_details":{"reasoning_tokens":2677}},"tokens_in":762,"tokens_out":2777,"duration_ms":17572,"temperature":1.0,"reasoning_tokens":2677,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:41:29.935388+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the G0W0-with-spin-orbit $\\mathbb{Z}_2$ invariant and band inversion for the PBE-only candidates (TaIrTe4, TaRhTe4, NbIrTe4, ZrBr, ZrCl, Cu2Te3Ti, In2ZnS4); if any of them reverts to a trivial insulator as TiNI does, the candidate list and the 1% abundance claim need downward revision.","supporting_citations":[{"cited_title":"Two-dimensional materials from high-throughput computational exfolia- tion of experimentally known compounds","cited_arxiv_id":null,"evidence_quote":"Supplies the 1,825-monolayer database of experimentally known exfoliable compounds that the screen starts from, including the magnetic-ground-state screening used later."},{"cited_title":"Prediction of a Large- Gap and Switchable Kane-Mele Quantum Spin Hall In- sulator","cited_arxiv_id":null,"evidence_quote":"Prediction of jacutingaite (Pt2HgSe3) as a Kane-Mele QSHI; the comparison baseline and the source of the quoted PBE inversion strength."},{"cited_title":"Vidal, X","cited_arxiv_id":null,"evidence_quote":"Documents false-positive and false-negative topological assignments in density functional theory, the justification for re-checking candidates at G0W0 and for the TiNI verdict."},{"cited_title":"Band inversion and topological aspects in a TiNI monolayer","cited_arxiv_id":null,"evidence_quote":"Reports the synthesis and crystal structure of Pd2HgSe3, the material added to the database and identified as a second Kane-Mele QSHI."},{"cited_title":"Quan- tum spin Hall effect in two-dimensional transition metal dichalcogenides","cited_arxiv_id":null,"evidence_quote":"Predicted the quantum spin Hall phase in 1T' transition-metal dichalcogenides and supplies the WTe2 and MoTe2 entries with their inversion strengths."},{"cited_title":"Van der Waals Stacking-Induced Topologi- cal Phase Transition in Layered Ternary Transition Metal Chalcogenides","cited_arxiv_id":null,"evidence_quote":"Predicted the MM'Te4 (Nb/Ta, Ir/Rh) monolayers as QSHIs, the previously known candidates reconfirmed in this screen."},{"cited_title":"Phonons and related crystal prop- erties from density-functional perturbation theory","cited_arxiv_id":null,"evidence_quote":"Provides the density-functional perturbation theory used to compute phonon dispersions and enforce the dynamical-stability filter."}],"review_version":1}