{"id":"d208a23e-f60c-498d-aba6-20d3b947cbd2","arxiv_id":"2607.15581","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Vapor HCl gas selectively etches MAX ceramics into a library of 36 two-dimensional carbides, nitrides, and carbonitrides, including semiconducting Hf2CTx.","lead":"Hydrogen chloride gas can selectively carve sheets of two-dimensional carbides and nitrides out of layered ceramic MAX phases, producing 36 materials including the first semiconducting hafnium-based MXene. The vapor process is fast, solvent-free, and scalable to kilograms, which could make MXene production practical for electronics, catalysis, and energy storage.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'general' claim rests on a binary DFT stability map that ignores kinetics and side reactions; the map has been validated only against the 12 demonstrated MXenes, leaving the other 24 advertised compositions untested.","rationale":"The paper's headline claim is a general and scalable vapor-phase platform producing a library of 36 2D carbides, nitrides, and carbonitrides. The generality hinges on the DFT stability map: it is the only evidence that extends the proof-of-concept beyond the 12 experimentally shown compositions. The map's binary competition (MXene formation vs. complete disintegration) is an idealized thermodynamic construct; it does not include kinetic barriers, partial side reactions, or volatility limits of ACl_y species. The main text confirms only a handful of failures, which the map predicts, but it does not test any predicted success outside the 12. Without such counterfactual tests, the claim that HCl vapor is a 'general' platform is underdetermined—favorable ΔG does not guarantee that a specific MAX phase will actually etch to a clean MXene at a practical rate. The reader's weakest_assumption already identified exactly this point, so I agree. The concern is substantive but does not overturn the paper: the 12 demonstrated MXenes, including the novel semiconducting Hf2CTx, are credible and the platform clearly works for a useful subset. The correct verdict remains CONDITIONAL, contingent on broader experimental validation of the map's predictions and on full enumeration/characterization of the 36-material library. No change to the reader's verdict is needed.","tokens_in":12121,"tokens_out":9539,"duration_ms":118674,"concrete_test":"Select 2–3 MAX phases from table S1 that Fig. 3A places below the diagonal (predicted MXene formation) and that are not among the 12 experimentally realized, e.g., a Si-, Ge-, Sc-, or Y-containing MAX. Subject each to the same HCl vapor etch (923 K, 25 min) and check for the characteristic MXene (002) XRD peak and loss of A by EDS. If any fails to convert to MXene, the binary thermodynamic model misses kinetic or volatility constraints and the 'general' claim must be narrowed; if all three convert, the thermodynamic map is strengthened.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim of a 'general and scalable' platform rests on the DFT thermodynamic stability map (Fig. 3A), which compares only two end-member reactions: selective etching to MXene + ACl_y(g) vs. complete disintegration to volatile chlorides. This model assumes that every A-site chloride is volatile at the reaction temperature and that no kinetic barrier or competing side reaction (M-site chlorination, passivating oxide from trace H2O, sluggish A diffusion) overturns the thermodynamic ranking. The map has been validated only against the 12 demonstrated MXenes plus a few expected failures (V2AlC, Cr2AlC, Mo-containing MAX phases); the other 24 compositions in the advertised 36-material library are not enumerated in the main text. No kinetic model or time-resolved measurement is provided. The paper's own mention that Mo-based MAX phases show 'low reactivity' despite the thermodynamic framework hints that kinetics matter but are not captured. If any predicted-success composition fails in practice—for example, because its ACl_y is insufficiently volatile at 923 K or because a side reaction passivates the M-X framework—the 'general' claim overreaches.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a vapor-phase etching and transformation platform for MXene synthesis. The central reaction is MAX + HCl(g) → MX (MXene) + ACl_y(g) + H2(g), where HCl vapor selectively removes A-layers from MAX phases at 873–1073 K, producing layered MXenes without liquid etching or washing. The authors support this with DFT free-energy calculations comparing MXene formation versus MAX disintegration, presenting a thermodynamic stability map for many MAX chemistries (36 materials advertised). They experimentally demonstrate 12 MXenes, including first-time Hf2CTx and Zr2CTx, and report a six-order-of-magnitude spread in electrical conductivity, with Tauc-derived optical bandgaps of 2.04 eV and 2.24 eV for Zr2CTx and Hf2CTx, respectively. They further show post-synthetic termination exchange (O, S, Se, Te, P), X-site substitution (C/N), and transformation into Sn- or Al-intercalated 'MAXene' structures. Scalability is claimed through single-batch 285 g and routine 10 kg production.","tokens_in":12365,"tokens_out":2756,"duration_ms":37707,"significance":"If the claims hold, this would be a substantial advance in MXene synthesis: a solvent-free, scalable route that expands the accessible MXene family to previously elusive Zr/Hf-based semiconducting members and enables postsynthetic compositional and structural tailoring. The thermodynamic framework is independent (computed using DFT and thermochemical data, not fit to the measured conductivity or bandgap), which is a strength, and the experimental demonstrations cover a range of chemistries. The work combines synthesis, characterization, and computational screening in a way that could reframe MXene production. However, the central generality and 'library of 36' claim currently rest on only 12 demonstrated phases and a thermodynamic map that omits kinetic and side-reaction pathways. The semiconducting assignment is based on optical measurements and low powder conductivity rather than transport measurements. These load-bearing points need strengthening before the platform can be considered established at the claimed breadth.","major_comments":[{"comment":"The abstract and Section 3 claim a 'library of 36 2D carbides, nitrides, and carbonitrides' and a universal synthesis, but the main text demonstrates only 12 MXenes experimentally. The remaining 24 compositions are not enumerated in the main text, and the thermodynamic stability map (Fig. 3A) is only a binary comparison of MXene formation versus total disintegration. The paper itself notes that Mo-based MAX phases show 'low reactivity' despite the thermodynamic framework, indicating that kinetics or side reactions can overturn the thermodynamic ranking. Since 'general' is a central claim, the authors should provide a full list of the 36 predicted compositions, mark which are experimentally verified, and discuss or measure kinetic limitations (e.g., time-dependent conversion, competing chlorination of M, passivation by trace H2O-derived oxide) for representative cases.","section":"Abstract and §3 (Figs. 3A, 3E)"},{"comment":"The semiconducting assignment for Zr2CTx and Hf2CTx rests on Tauc plots from UV-vis diffuse reflectance (optical bandgaps 2.04 and 2.24 eV) and on a drop in electrical conductivity to 10^-2 S m^-1. No temperature-dependent conductivity, Hall-effect, or field-effect transistor data are provided to confirm semiconducting behavior. The conductivity values in Fig. 3E appear to come from pressed powders or films without stated measurement geometry or error bars; contact resistance and porosity can dominate. Given that the discovery of semiconducting MXenes is a headline result, transport measurements on delaminated flakes or at least carefully described pellet measurements are needed to distinguish intrinsic semiconductivity from poor interparticle contact.","section":"§3, Fig. 3E–G"},{"comment":"Scalability is claimed from a 285 g single-batch synthesis and routine 10 kg production, but no process details are given: reactor design, gas flow rates, temperature uniformity, batch-to-batch reproducibility, yield, or purity. A single photograph (Fig. 1B) is not sufficient evidence for a 'scalable platform' central claim. Provide quantitative scale-up data, including reaction time, conversion efficiency, and characterization (XRD, composition) at the largest scale, to substantiate the scalability.","section":"§5 'Scalable production' and Fig. 1B"},{"comment":"The 'MAXene' structures (Cl-MX-[A-MX]_n-Cl) are claimed from atomic-resolution STEM images after Sn vapor treatment, but the XRD evidence (Fig. S72) shows only a weak broad peak, and no diffraction pattern or quantitative composition analysis is presented to confirm the stacking sequences and the absence of unreacted Sn or Sn-containing secondary phases. The claim that n can be 1, 2, and 4 is structural but not statistically supported. Given that this is presented as a new class of non-vdW layers, more evidence (e.g., EELS/EDX line profiles, simulated STEM images, or diffraction) is required.","section":"§4 and Fig. 4D–G"}],"minor_comments":[{"comment":"No error bars or measurement uncertainty are shown for the conductivity values. At least three independent samples or measurements should be reported to support the six-order-of-magnitude spread.","section":"Fig. 3E"},{"comment":"The caption contains an incomplete phrase: 'as a function of HCl-to-MAX molar (nHCl/nMAX)' — likely missing 'ratio'.","section":"Fig. 1C caption"},{"comment":"The notation M-X-T′ with T′ = O, S, Se, Te, P is used; clarify whether these are terminal sites or substitutional X-site doping, as both are discussed in the same section.","section":"§4, Fig. 4A"},{"comment":"The term 'MAXenes' is introduced without a formal definition or comparison to existing nomenclature. Define it explicitly and note its relation to 'MAX phases' and 'MXenes'.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically ambitious and the core vapor-etching reaction is plausible and well evidenced for the demonstrated phases. My main concern is the gap between the advertised 'library of 36' and the 12 experimentally verified compositions, combined with a purely thermodynamic selection map that explicitly fails for Mo-containing phases. The semiconductor claim also needs transport confirmation. These are fixable with additional data and careful scoping of claims, not fundamental errors. The manuscript would be suitable after major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, this is a genuinely useful synthesis paper. The core result—reactive HCl vapor selectively removes A-layers from MAX phases to give MXenes—is well supported by XRD, STEM, XPS, XAS, and conductivity data, and it delivers two things the field has wanted: a washing-free scalable route (285 g per batch) and the first experimental Hf2CTx, with a Tauc-derived bandgap of 2.24 eV. The DFT stability map is an independent screening tool, not fit to the experiments, and it correctly anticipates the failed V2AlC/Cr2AlC cases. The termination swapping (Cl to O/S/Se/Te/P), X-site C/N substitution, and Sn/Al 'zippering' into MAXene stacks are all credible extensions, backed by atomically resolved STEM. That is a substantial batch of new results.\n\nThe soft spots are real but mostly presentational. The headline 'library of 36' overstates what is shown: the main text demonstrates 12 MXenes, and the other 24 are not enumerated or characterized at all in the main text. Even with the supplement, the reader cannot verify the 36 without the table. The semiconductor claim for Zr2CTx and Hf2CTx rests on Tauc plots and a conductivity drop to 10^-2 S/m, which is suggestive but not proof of a transport gap; hopping or grain-boundary effects could give the same numbers. I'd want at least temperature-dependent conductivity before calling them semiconductors. Error bars are missing from the conductivity and bandgap data. The stress-test concern about the DFT map is fair: comparing MXene formation vs. total disintegration is a two-reaction model, and Mo-containing MAX phases already show low reactivity that the map doesn't predict. That doesn't sink the paper; it means 'general' should be softened to 'broadly demonstrated across the tested compositions.'\n\nThe citation pattern is fine; the Björk/Rosen self-citations are directly on termination thermodynamics and are appropriate. No circular fitting.\n\nVerdict: this deserves a serious referee, not a desk reject. Ask for the full 36-material table with per-material characterization, error bars/statistics, a transport measurement on Hf2CTx if possible, and a title/abstract that matches what is actually demonstrated.","headline":"Vapor HCl etching works, Hf2CTx is new and credible, but the '36-material library' and 'semiconductor' claims run ahead of the demonstrated data.","tokens_in":12943,"tokens_out":2304,"would_cite":true,"duration_ms":29395,"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":"This paper reports that hydrogen chloride vapor can selectively strip the A-element layers out of MAX phases, yielding a library of 36 two-dimensional carbides, nitrides, and carbonitrides—including the first semiconducting Hf2CTx—without l","keywords":["MXenes","MAX phases","vapor etching","hydrogen chloride","two-dimensional materials","semiconducting Hf2CTx","surface termination","scalable synthesis"],"falsifier":"Take a MAX phase that the thermodynamic map places below the diagonal but that is not among the 12 experimentally demonstrated, run HCl vapor etching at 923 K, and check for a layered MXene. If the product is metal chlorides, unreacted MAX, or a non-layered residue, the universality claim fails. A narrower falsifier: measure four-probe resistance of a delaminated Hf2CTx film versus temperature; a metallic temperature dependence would contradict the claimed semiconductor.","tokens_in":11986,"feed_emoji":"🧪","tokens_out":9645,"duration_ms":101372,"temperature":0.7,"pith_summary":"The paper's central claim is that a gas-solid reaction, rather than liquid acids, can make MXenes at scale. At 873–1073 K, HCl vapor reacts with the A element in a MAX phase and carries it away as a volatile chloride, leaving the transition-metal carbide/nitride layers intact; because the byproduct is a gas, nothing accumulates to block the etch. The authors support this with a thermodynamic map that predicts which MAX phases will etch selectively, and they demonstrate 12 MXenes, including the previously elusive semiconducting Hf2CTx. The broader assertion is that this solvent-free vapor platform is general: a library of 36 carbides, nitrides, and carbonitrides with conductivities spanning six orders of magnitude, and routine production to ten kilograms. If correct, it would turn MXene synthesis into a dry, cheap, scalable step.","feed_headline":"Vapor etching yields 36 2D carbides, nitrides, and carbonitrides","feed_subtitle":"No acid baths or washing: HCl gas strips A-layers from MAX phases to yield MXenes spanning metallic to semiconducting.","key_machinery":"The load-bearing machinery is the reaction MAX + HCl(g) → MX + ACl_y(g) + H2(g), combined with a thermodynamic stability map built from density functional theory. The map places each MAX phase on a plot whose axes are the computed free-energy changes for MXene formation and for full disintegration; the diagonal sets the selectivity boundary. This is what turns a single etch into a library prediction: any MAX phase below the diagonal should yield the corresponding MXene, and experiments on 12 compounds confirm the trend. The volatility of ACl_y is equally important—it removes the byproduct continuously, which is why the etch proceeds in minutes and can be scaled to kilogram batches.","core_discovery":"The paper discovers that HCl vapor selectively removes A-layers from MAX phases via MAX + HCl(g) → MX (MXene) + ACl_y(g) + H2(g). The volatile ACl_y leaves the solid, preventing byproduct accumulation and allowing deep, sustained etching—fast enough to convert Ti4AlN3 to Ti4N3Tx in 25 minutes. A DFT-based thermodynamic stability map compares formation of the MXene against complete disintegration of the MAX phase; compositions below the diagonal etch selectively, whereas V2AlC and Cr2AlC, above the diagonal, do not. Twelve MXenes are experimentally made, including Zr2CTx and first Hf2CTx, whose optical absorption gives indirect bandgaps of 2.04 and 2.24 eV, and the ensemble spans six orders o","pith_inferences":["Because the selectivity criterion is thermodynamic and the byproduct is volatile, the same design rule should extend to other non-van der Waals laminates whose A element forms a volatile chloride—this is my inference, not a claim the paper tests.","The semiconducting character of Zr2CTx and Hf2CTx is inferred from optical absorption; a direct four-probe transport measurement on delaminated flakes would confirm whether the bandgap is intrinsic and whether carrier mobility is useful.","The paper's batches are still batch processes; a continuous-flow reactor that feeds HCl gas and removes ACl_y vapor continuously is the obvious next scale-up step and is not demonstrated here."],"forward_implications":["MXene production becomes a solvent-free, washing-free gas-solid process; the paper demonstrates 285 g in a single batch and routine 10 kg production, with lower water use and cost than wet or molten-salt routes.","The first synthesis of semiconducting Hf2CTx opens access to a predicted class of MXenes beyond metallic conductors, with reported indirect optical bandgaps of 2.04 eV for Zr2CTx and 2.24 eV for Hf2CTx.","Exposing MXenes to O2, H2S, Se, Te, or P vapor after etching changes their surface terminations, turning Ti4N3Tx from metallic Cl-terminated to semiconductor-like O-terminated and making it stable in base for three weeks and to 1100 K.","The platform edits the interior of the layers: CH4 or NH3 swaps C and N at the X-site, enabling carbide–nitride interconversion, and Sn or Al vapor inserts A layers between MX slabs to make 2D MAX phases with conductivities up to about 48,000 S/m."],"fun_headline_variants":["HCl vapor strips MAX phases into 36 2D materials","No acids needed: vapor etching makes 2D MXenes","Vapor platform yields semiconducting Hf2CTx and more","36 2D carbides, nitrides from vapor etching","Scalable vapor etching opens 2D material library"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The generality claim depends on the DFT thermodynamic ranking being the whole story: that HCl selectively volatilizes the A element while leaving the M-X framework intact, and that kinetic barriers, side chlorination of the metal, or trace-water oxidation do not overturn that ranking for any of the 36 library members.","fun_headline_variants_meta":{"raw":{"variants":["HCl vapor strips MAX phases into 36 2D materials","No acids needed: vapor etching makes 2D MXenes","Vapor platform yields semiconducting Hf2CTx and more","36 2D carbides, nitrides from vapor etching","Scalable vapor etching opens 2D material library"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000523,"raw_usage":{"total_tokens":2355,"prompt_tokens":725,"completion_tokens":1630,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":469,"completion_tokens_details":{"reasoning_tokens":1544}},"tokens_in":469,"tokens_out":1630,"duration_ms":11970,"temperature":1.0,"reasoning_tokens":1544,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T22:51:38.368734+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a MAX phase that the thermodynamic map places below the diagonal but that is not among the 12 experimentally demonstrated, run HCl vapor etching at 923 K, and check for a layered MXene. If the product is metal chlorides, unreacted MAX, or a non-layered residue, the universality claim fails. A narrower falsifier: measure four-probe resistance of a delaminated Hf2CTx film versus temperature; a metallic temperature dependence would contradict the claimed semiconductor.","supporting_citations":[],"review_version":1}