{"id":"1e6fa04e-adc9-40c3-b177-bc48fc1058c8","arxiv_id":"2608.03478","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"First-principles calculations find that tetragonal GeS2 and GeSe2 suppress cross-plane lattice heat flow to about 1.2 and 1.5 W/m/K at 300 K, yielding a moderate cross-plane zT up to 0.257 for n-type GeS2 at 800 K.","lead":"A computational study reports that layered tetragonal GeS2 conducts heat about 20 times better within its planes than across them, and estimates a thermoelectric figure of merit zT near 0.26 along the cross-plane direction at 800 K. The result, if correct, suggests these cheap germanium chalcogenides are strongly anisotropic heat managers worth testing experimentally.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Cross-plane kappa is the load-bearing quantity: the production cell has c=11.002 Å (1.8% below the quoted reference c=11.20 Å) and the 4×4×2/7th-neighbor third-order cutoff has no convergence test, so the 22:1 anisotropy and zT_c=0.257 are not yet quantitatively secure.","rationale":"The paper is a standard, mostly transparent first-principles study. The qualitative directional story—layered P42/nmc framework gives large in-plane vs cross-plane stiffness and suppressed cross-plane heat flow—is plausible and internally supported by the elastic constants and low-frequency phonon analysis. But the abstract's quantitative claims (κ_c = 1.19 W m−1 K−1, κ_ab/κ_c ≈ 22, zT_c = 0.257) are load-bearing, and the weakest link is precisely the cross-plane lattice thermal conductivity. The reader identified this correctly: the production cell's 1.8% shorter c-axis and the lack of any third-order force-constant convergence test make κ_c, and therefore the anisotropy ratio and zT_c, numerically fragile. A 1.8% change in c may seem small, but in layered compounds the interlayer force constants and anharmonic scattering that set κ_c vary steeply with interlayer separation; this is not an external-consensus disagreement but an internal numerical-convergence gap. The paper does provide useful independent checks—elastic stability, phonon stability, explicit statement of RTA vs iterative tensors, and transparent acknowledgment of GeSe2's qualitative status—so I do not think the central directional claim is wrong. However, the specific values in the abstract should not be taken as converged until either a c-spacing sensitivity test or a third-order supercell/cutoff convergence check is reported. Since the reader already assigned CONDITIONAL on essentially this basis, my stress-test does not move the verdict; it reinforces it.","tokens_in":25558,"tokens_out":4122,"duration_ms":51311,"concrete_test":"Recompute κ_c for GeS2 at 300 K with the same ShengBTE/DFT-D2 workflow using (i) the quoted reference c=11.20 Å with re-optimized internal coordinates, and (ii) the production c=11.002 Å cell, while also increasing the third-order supercell to 6×6×2 or at least extending the neighbor cutoff beyond the seventh shell. If κ_c changes by more than ~30%—or the anisotropy ratio by more than ~20%—between these settings, the quantitative headline values are not converged; if it changes by less than ~10%, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claims—the ~22:1 in-plane/cross-plane lattice thermal conductivity anisotropy and the headline zT_c=0.257 for n-type GeS2—both hinge on κ_c, which is 20–25 times smaller than κ_ab. Yet κ_c is exactly the component least protected by the calculations. In Sec. 2.1 the production Quantum ESPRESSO cell is fixed by celldm(3)=3.133016, giving c=11.002 Å, while Sec. 1 quotes the reference structure as c=11.20 Å; the 1.8% interlayer contraction is not tested for its effect on cross-plane transport, even though van der Waals–dominated interlayer coupling is highly sensitive to this distance. The third-order force constants used for the ShengBTE calculation are generated from 4×4×2 supercells and truncated at the seventh-nearest-neighbor shell, with no reported convergence test against larger supercells or longer cutoffs anywhere in the main text or Supplementary Information. Because κ_c is dominated by weak interlayer anharmonic interactions, a modest under-convergence in third-order interlayer force constants—or a small overbinding from the Grimme D2 correction—can change κ_c by a large factor. Since κ_e is small at the reported doping (κ_e+κ_l = 0.496 W m−1 K−1 at 800 K, with κ_c=0.46 W m−1 K−1), zT_c is almost inversely proportional to κ_c: if κ_c increased by 2×, the headline zT would drop from 0.257 to roughly 0.13. No experimental or independent computational benchmark for κ_c is provided, so the absolute cross-plane values rest on a single, untested computational setting.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a first-principles study of tetragonal GeS2 and GeSe2 (space group P42/nmc) combining PBE+DFT-D2 structural and electronic calculations, HSE03/Wannier band-structure references, DFPT phonons, ShengBTE lattice thermal conductivity (RTA and iterative), AMSET carrier-dependent transport, LOBSTER bonding analysis, and a component-wise zT estimate. The central quantitative claims are that GeS2 has a roughly 22:1 in-plane to cross-plane lattice thermal conductivity anisotropy (26.86 versus 1.19 W m−1 K−1 at 300 K; 10.22 versus 0.46 W m−1 K−1 at 800 K), and that n-type cross-plane GeS2 reaches zT_c = 0.257 at 800 K and 1×10^19 cm−3 because the suppressed cross-plane lattice thermal conductivity dominates the zT denominator. GeSe2, predicted in the same tetragonal structure by S-to-Se substitution, gives smaller values (zT_c = 0.066 p-type at 800 K), with the authors explicitly labeling the GeSe2 transport results as qualitative owing to the near-closure of the PBE gap. The paper also reports elastic constants, phonon dispersions, gap values, and bonding descriptors as supporting evidence.","tokens_in":25841,"tokens_out":2560,"duration_ms":35283,"significance":"If the quantitative results are correct, the paper identifies a strikingly anisotropic phonon-transport behavior in a relatively little-studied polymorph family and shows that the suppressed cross-plane lattice conductivity is the key factor directing the best thermoelectric response along the c-axis. The work is also useful as a first computational characterization of a hypothetical tetragonal GeSe2 phase. Strengths include the internally consistent workflow using standard, well-tested codes (Quantum ESPRESSO, ShengBTE, AMSET, Wannier90, LOBSTER), the clear disclosure of the grid-point selection for zT maxima, the explicit component-wise combination of electronic and lattice tensors, and the honest caveats about the PBE-based GeSe2 transport. The qualitative direction of the anisotropy is consistent with the layered structure and the elastic constants, so the central physics is plausible. However, the quantitative headline numbers—the anisotropy ratio and zT_c—depend on cross-plane lattice thermal conductivity, which is the least converged quantity in the calculation. No experimental benchmark or independent computational benchmark for κ_c is provided.","major_comments":[{"comment":"The production Quantum ESPRESSO cell is defined by celldm(1)=6.636200 and celldm(3)=3.133016, giving c=11.002 Å, while the reference structure quoted in the Introduction and Sec. 3.1 has c=11.20 Å. The 1.8% contraction in the interlayer spacing is not tested for its effect on the cross-plane transport. Since κ_c is 20–25 times smaller than κ_ab and is governed by weak interlayer anharmonic interactions, a contraction of this size can plausibly change κ_c and therefore the headline anisotropy ratio and zT_c by a large factor. I request a convergence test over the cell (e.g., using the fully relaxed c under the same Grimme D2 correction, or computing κ_c at the experimental/reference c) and a statement of the resulting change in κ_c and zT_c.","section":"Sec. 2.1 and Sec. 3.1"},{"comment":"The third-order force constants used in ShengBTE were generated from 4×4×2 supercells truncated at the seventh-nearest-neighbor shell (thirdorder_espresso.py scf.in reap 4 4 2 -7) with no reported convergence test against larger supercells or longer cutoffs. Cross-plane thermal transport is especially sensitive to long-range anharmonic interlayer coupling, so the absence of a cutoff/supercell convergence test is a load-bearing gap. I ask for explicit convergence data (κ_ab, κ_c, and the ratio) for at least one larger supercell and/or one longer cutoff, or a detailed justification for why the chosen cutoff is sufficient.","section":"Sec. 2.1 and SI S1"},{"comment":"The headline zT_c = 0.257 for n-type GeS2 is obtained from AMSET electronic coefficients computed from PBE bands with a 0.99 eV gap and with 'placeholder weights' for orbital projections, while the HSE03/Wannier gap is 2.48 eV. The paper does not quantify how a gap-corrected electronic structure would change the carrier-density–chemical-potential mapping, mobilities, and therefore zT_c. Given that the zT maximum occurs at a relatively low carrier concentration (1×10^19 cm−3) near the band edge, this sensitivity should be assessed. If a full HSE03 AMSET calculation is too costly, a scissor-shift or constant-relaxation-time comparison at the same carrier densities would help establish robustness. The paper already labels GeSe2 as qualitative because of the PBE near-overlap, but the same issue needs to be addressed for the GeS2 headline number.","section":"Sec. 3.7 and Sec. 2.1"}],"minor_comments":[{"comment":"Typo: 'gives azT value' should read 'gives a zT value'.","section":"Abstract"},{"comment":"The text says the final variable-cell relaxations gave residual pressures of −0.03 kbar for GeS2, but the production cell has c=11.002 Å, which differs from the quoted reference c=11.20 Å. Clarify whether the relaxation genuinely found this contracted c under Grimme D2, and state that the reference structure is not the one used in the transport calculations.","section":"Sec. 3.1"},{"comment":"The temperature dependence of κ_l is described as approximately T^{−0.97} to T^{−0.99}; state explicitly how these exponents were extracted (e.g., fit over 300–800 K) and give the corresponding R² or uncertainty.","section":"Sec. 3.4"},{"comment":"The text states 'IMP scattering is included in the transport tensors but not in the plotted electron–phonon lifetime.' The figure label in SI Fig. S9 only mentions ADP and POP; make the plot label consistent with the main text.","section":"Sec. 3.6"},{"comment":"Reference 23, 'Gang Tse' appears to be a malformed author name; check the actual authorship of the HSE03 GeS2 study.","section":"References"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing you should know first: this is a transparent, well-documented first-principles study of two tetragonal GeX2 phases that, as far as I can tell, have no prior phonon or thermoelectric transport calculations. The central qualitative claim—a roughly 22:1 in-plane to cross-plane lattice thermal conductivity anisotropy in GeS2, rooted in the layered P42/nmc structure—is internally consistent with the computed elastic anisotropy (C11/C33 ≈ 5.5) and is likely right. The numbers attached to it are less secure.\n\nWhat is actually new: the specific phonon and electronic transport tensors, anisotropy ratios, and zT estimates for bulk tetragonal GeS2 and for the substituted tetragonal GeSe2 phase. The workflow is standard—QE, DFPT, ShengBTE, AMSET, LOBSTER—but executed at a reasonable level. The authors are more honest than many: they disclose the placeholder orbital-overlap factors in AMSET, the near-band-overlap PBE limit for GeSe2, and they explicitly label the GeSe2 zT as qualitative in the main text. No constants are fitted to experimental targets, so there is no circularity.\n\nThe soft spots, in order of importance. Cross-plane κ_c is the load-bearing quantity. It is 20–25 times smaller than κ_ab, and the headline zT_c = 0.257 is nearly inversely proportional to it. Yet κ_c is exactly the component least protected by the calculation: the production cell has c = 11.002 Å while the quoted reference is 11.20 Å (a 1.8% contraction that matters for the van der Waals–dominated interlayer coupling), and the third-order force constants are generated from 4×4×2 supercells with a seventh-nearest-neighbor cutoff and no reported convergence test. No experimental or independent computational benchmark for κ_c is provided. If κ_c were off by a factor of two, zT_c would drop to roughly 0.13. The abstract also presents the GeSe2 zT of 0.066 without the qualitative caveat that appears in the main text. These are real weaknesses, but they are quantitative uncertainties, not a fatal flaw: the directional anisotropy claim does not depend on the exact magnitude.\n\nWho is this for: people working on Ge-chalcogenide thermoelectrics or on anisotropic phonon transport in layered materials. It is a useful, though not definitive, computational data point.\n\nRecommendation for peer review: send it out. The work is new and honestly presented, and the central claim is physically plausible. A referee can produce a concise request—convergence tests on the third-order force constants, a cell-geometry check, and a gap-corrected AMSET calculation for GeSe2—rather than a rejection. Desk rejection would be wrong.","headline":"Transparent first-principles study of two new tetragonal GeX2 phases: the 22:1 anisotropy claim likely holds, but the headline zT and κ_c numbers rest on untested anharmonic-force-constant settings.","tokens_in":26528,"tokens_out":4844,"would_cite":false,"duration_ms":51013,"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":"Tetragonal GeS2 conducts heat about 22 times better within its layers than across them, and its strongest thermoelectric response runs along the stacking direction.","keywords":["tetragonal GeS2","lattice thermal conductivity anisotropy","phonon Boltzmann transport","thermoelectric figure of merit","ShengBTE","AMSET","germanium diselenide","first-principles transport"],"falsifier":"Measure the cross-plane thermal conductivity of a single-crystal or high-quality film of tetragonal GeS2 at 300 K: if it comes out near 25 W m-1 K-1 rather than about 1 W m-1 K-1, the predicted 22:1 anisotropy collapses. Computationally, a convergence test of kappa_c against larger supercells or higher-order interlayer force constants would settle whether the seventh-nearest-neighbor truncation is adequate.","tokens_in":25271,"feed_emoji":"⚡","tokens_out":4005,"duration_ms":47753,"temperature":0.7,"pith_summary":"The paper aims to establish that tetragonal GeS2 and GeSe2 are strongly anisotropic heat conductors, with in-plane lattice thermal conductivity far exceeding cross-plane values. For GeS2 at 300 K it predicts 26.86 W m-1 K-1 in-plane versus 1.19 W m-1 K-1 cross-plane, an anisotropy ratio near 22:1 that persists up to 800 K. The suppressed cross-plane heat flow is attributed to the layered framework of corner-sharing GeX4 tetrahedra: low-frequency phonons carry most heat in plane, while out-of-plane transport is kinematically restricted. Combining these lattice tensors with scattering-aware electronic transport gives a cross-plane figure of merit zT = 0.257 for n-type GeS2 at 800 K and 10^19 cm-3, far above its in-plane value. The paper also argues that low lattice thermal conductivity alone is not enough, since the power factor and electronic heat conduction still limit performance.","feed_headline":"Heat flows 22x faster within GeS2 layers than across them","feed_subtitle":"Layered GeS2's suppressed cross-plane heat flow puts its best thermoelectric response along the stacking axis.","key_machinery":"The load-bearing machinery is the phonon Boltzmann transport equation solved by ShengBTE in the relaxation-time approximation, supplied with DFPT harmonic force constants and finite-displacement third-order force constants, which yields the anisotropic lattice thermal conductivity tensor. Electronic transport comes from AMSET with state-dependent acoustic-deformation-potential, polar-optical-phonon, and ionized-impurity scattering. The RTA lattice tensor and the AMSET electronic tensor are combined component-wise (xx paired with kappa_ab, zz with kappa_c) to produce directional zT values. The mechanism behind the low kappa_c is the layered corner-sharing GeX4 framework, where most heat-carry","core_discovery":"The central claim is that tetragonal GeS2 (and GeSe2 to a lesser extent) has a strong, persistent anisotropy in phonon heat conduction: at 300 K, GeS2 gives kappa_ab = 26.86 W m-1 K-1 and kappa_c = 1.19 W m-1 K-1, and GeSe2 gives 18.74 and 1.52 W m-1 K-1, with anisotropy ratios near 22 and 12.3 respectively. This separation is traced to the quasi-two-dimensional tetrahedral network and weak interlayer coupling, which restricts out-of-plane phonon velocities and heat-carrying modes. Because the cross-plane lattice conductivity is so small, pairing the ShengBTE RTA lattice tensor with AMSET electronic coefficients makes the cross-plane direction the best thermoelectric channel: zT_c = 0.257 fo","pith_inferences":["I infer that few-layer or thin-film forms of tetragonal GeS2, with even weaker interlayer coupling than the bulk, could show an even larger cross-plane suppression, though the paper does not test this.","A direct testable extension is a thermal-conductivity measurement on a high-quality single crystal or oriented film: a cross-plane value near 1 W m-1 K-1 would support the third-order force-constant treatment, while a value near 25 W m-1 K-1 would invalidate it.","The paper's own warning that the GeSe2 zT is qualitative suggests a gap-corrected AMSET calculation could substantially change the GeSe2 numbers, and this is the most natural follow-up.","The small C66 shear constant hints that shear strain, not just layer sliding, may be an efficient phonon-scattering lever, but the paper does not compute strain-dependent conductivity."],"forward_implications":["If the predicted anisotropy is correct, orientation control alone could reduce the relevant lattice thermal conductivity by more than a factor of 20 in tetragonal GeS2, without nanostructuring or alloying.","The cross-plane channel becomes the target for n-type GeS2 thermoelectric design, with a calculated zT near 0.26 at 800 K in the pristine material.","Se substitution lowers the in-plane conductivity and the anisotropy ratio but keeps a useful ~12:1 contrast, while shifting phonon frequencies downward.","Because the Lorenz number varies with carrier concentration and direction, using a fixed Wiedemann-Franz value would have misestimated the electronic contribution to heat transport.","The roughly T^-1 temperature dependence of kappa indicates standard crystal-like anharmonic transport, suggesting defects or alloy scattering could reduce kappa further."],"supporting_citations":[{"why":"Supplies the ShengBTE phonon Boltzmann transport solver used to compute the anisotropic lattice thermal conductivity tensor.","marker":"[29]"},{"why":"Supplies the AMSET scattering-aware electronic transport formalism used for the electrical conductivity, Seebeck coefficient, and electronic thermal conductivity.","marker":"[26]"},{"why":"Provides the tetragonal GeS2 reference structure and the experimental evidence of strong in-plane anisotropy that motivates the study.","marker":"[17]"},{"why":"Reports weak interlayer interaction in layered GeSe2, supporting the assumption that cross-plane phonon transport is suppressed.","marker":"[20]"},{"why":"The independent Phono3py finite-displacement calculation used to verify the harmonic phonon spectrum and the non-analytic splitting at Gamma.","marker":"[48]"}],"fun_headline_variants":["Heat flows 22x faster in GeS2 layers than across them","GeS2's layered heat flow boosts cross-plane thermoelectric performance","Anisotropic phonon flow in GeS2 sets up cross-plane zT boost","GeS2: 22x heat anisotropy steers thermoelectric response","Tetragonal GeS2 heat flow anisotropy favors cross-plane zT"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The load-bearing premise is that third-order interatomic forces computed from a 4x4x2 supercell truncated at the seventh-nearest-neighbor shell capture the weak interlayer coupling accurately enough that the cross-plane lattice conductivity is trustworthy, yet no convergence test or experimental benchmark for kappa_c is supplied.","fun_headline_variants_meta":{"raw":{"variants":["Heat flows 22x faster in GeS2 layers than across them","GeS2's layered heat flow boosts cross-plane thermoelectric performance","Anisotropic phonon flow in GeS2 sets up cross-plane zT boost","GeS2: 22x heat anisotropy steers thermoelectric response","Tetragonal GeS2 heat flow anisotropy favors cross-plane zT"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000696,"raw_usage":{"total_tokens":3135,"prompt_tokens":1051,"completion_tokens":2084,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":795,"completion_tokens_details":{"reasoning_tokens":1996}},"tokens_in":795,"tokens_out":2084,"duration_ms":15760,"temperature":1.0,"reasoning_tokens":1996,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T18:18:05.054625+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the cross-plane thermal conductivity of a single-crystal or high-quality film of tetragonal GeS2 at 300 K: if it comes out near 25 W m-1 K-1 rather than about 1 W m-1 K-1, the predicted 22:1 anisotropy collapses. Computationally, a convergence test of kappa_c against larger supercells or higher-order interlayer force constants would settle whether the seventh-nearest-neighbor truncation is adequate.","supporting_citations":[{"cited_title":"Ganose, Junsoo Park, Alireza Faghaninia, Rachel Woods-Robinson, Kristin A","cited_arxiv_id":null,"evidence_quote":"Supplies the AMSET scattering-aware electronic transport formalism used for the electrical conductivity, Seebeck coefficient, and electronic thermal conductivity."},{"cited_title":"Sub-angstrom characterization of the structural origin for high in-plane anisotropy in 2d GeS _2","cited_arxiv_id":null,"evidence_quote":"Provides the tetragonal GeS2 reference structure and the experimental evidence of strong in-plane anisotropy that motivates the study."},{"cited_title":"Weak interlayer interaction in 2D anisotropic GeSe _2","cited_arxiv_id":null,"evidence_quote":"Reports weak interlayer interaction in layered GeSe2, supporting the assumption that cross-plane phonon transport is suppressed."}],"review_version":1}