{"id":"bd6d7c4d-0328-40aa-9704-3e5d7ceebb6d","arxiv_id":"2502.05927","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Under hydrostatic pressure, β′-In2Se3 remains stable to 45 GPa with no β phase, and phase IV is a disordered orthorhombic solid solution.","lead":"Using helium as a pressure medium, this study tracks the crystal structure and vibrations of the ferroelectric layered material α-In2Se3 up to 60 GPa. It finds no intermediate β phase, contradicting earlier reports, and proposes a new disordered structure above 45 GPa.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The no-β conclusion rests on an absence-of-evidence argument: a second-order group-subgroup symmetrization could be hidden in both XRD and Raman, so the data above 12 GPa need a quantitative model comparison before the central claim is secure.","rationale":"The reader's CONDITIONAL verdict already centers on the same load-bearing premise: the experiments' ability to resolve a subtle second-order β′→β transition. I agree with that identification. The paper is careful, and it is honest about the ambiguity of indexing above 12 GPa and about the preliminary nature of the phase-IV model. However, the strongest claim — no β phase up to 45 GPa — is not positively established. The F−f linearity only rules out a volume anomaly; it cannot detect a pure symmetry change. The Raman mode-count argument is the best evidence, but without a quantitative line-shape analysis showing that no mode loses intensity discontinuously, the observed continuity could simply reflect the detection threshold. This is a resolvable, data-level question rather than a fundamental flaw: the same XRD and Raman datasets could be reanalyzed on purpose. Consequently the appropriate verdict remains CONDITIONAL; my stress-test does not change the reader's assessment. I would not raise the correctness risk to a rejection because the authors' prior hydrostatic measurements and the internal consistency of their XRD/Raman/F−f observations give meaningful, if not conclusive, support for their interpretation.","tokens_in":10834,"tokens_out":6206,"duration_ms":64030,"concrete_test":"Re-analyze the existing XRD patterns at 12, 20, 30 and 40 GPa with Le Bail fits using both C2/m (β′) and R-3m (β) models, and compare weighted-profile R-factors or χ². For each pattern, integrate the intensity of a reflection that is allowed in C2/m but forbidden in R-3m (identify it from the group-subgroup relation). If that diagnostic peak intensity falls to the instrumental background by ~12 GPa while the R-3m fit is statistically comparable, the data actually support a β′→β transition that the paper missed. If the diagnostic peak remains observable and C2/m fits significantly better at every pressure, the no-β claim is confirmed. Apply the same quantitative test to the Raman spectra: fit each spectrum and track the area of the lowest-frequency β′ mode; a continuous drop to zero above 12 GPa would indicate the transition.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the stability of β′-In2Se3 up to 45 GPa, i.e. the absence of the β′→β transition. The paper itself exposes the fragility: Sec. III B states that XRD patterns above 12 GPa can be indexed with either β′ (C2/m) or β (R-3m), that the transition is second-order with no volume discontinuity, and that the two space groups are in a group-subgroup relation. A second-order group-subgroup transition can be continuous in all thermodynamic variables while only the symmetry changes; consequently the F−f plot (Fig. 6) is not diagnostic, since it probes only P(V) and would remain linear through such a transition. The Raman argument is stronger but still depends on observing a 6→4 mode reduction; in unpolarized powder spectra the two modes that should disappear may simply be weak or unresolved, and the text does not show a quantitative mode-area or linewidth analysis across 12–45 GPa. Thus the data are fully consistent with β′ up to 45 GPa, but they are also consistent with β above 12 GPa if the diagnostic peaks/modes fall below the detection threshold. The distinction is therefore an absence-of-evidence argument rather than a positive identification, and the phase-IV assignment is explicitly preliminary.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a combined synchrotron X-ray diffraction and Raman spectroscopy study of α-In2Se3 in a diamond anvil cell with helium as the pressure-transmitting medium, from ambient pressure to above 60 GPa at room temperature. The authors observe the α→β′ transition at about 1 GPa, consistent with previous work, and find no evidence for the previously reported β′→β transition up to 45 GPa. Above 45 GPa they observe a transition to a phase they call phase IV, which they propose is a disordered, solid-solution-like orthorhombic structure. The paper compares its results with earlier studies, attributes earlier reports of β to non-hydrostatic conditions or insufficient resolution, and discusses the reversibility and hysteresis of the transitions.","tokens_in":11076,"tokens_out":5072,"duration_ms":55798,"significance":"If the no-β claim is correct, the paper revises the accepted pressure-temperature sequence of a technologically important layered ferroelectric and provides improved benchmark data under quasi-hydrostatic conditions. The simultaneous XRD and Raman measurements with helium as the pressure medium, the high maximum pressure, and the detailed comparison with previous work are notable strengths. The F−f analysis and the careful discussion of hydrostaticity add value. However, the central negative claim (stability of β′ and absence of β up to 45 GPa) rests on an absence-of-evidence argument whose load-bearing assumption—that a subtle second-order group-subgroup transition would be visible in the measured Raman mode count or in F−f linearity—is not quantified in the manuscript. The phase IV structural assignment is also explicitly preliminary. The significance of the paper therefore depends on closing this gap with a quantitative model comparison.","major_comments":[{"comment":"The conclusion that β′-In2Se3 remains stable up to 45 GPa is not established by the analyses presented. The manuscript states that XRD patterns above 12 GPa can be indexed with either β′ (C2/m) or β (R-3m), that the transition is characterized by a symmetrization without volume discontinuity, and that the two space groups hold a group-subgroup relation. For a second-order transition of this kind, the finite-strain F−f plot in Fig. 6 probes only volume as a function of pressure and cannot be expected to show a signature; a continuous symmetry change can leave P(V) perfectly smooth. To support the central claim, the authors should provide a quantitative comparison of the two structural models across the 12–45 GPa range, for example Le Bail/Rietveld refinements with both space groups and a report of Rwp, GooF, or an information criterion, together with an estimate of the minimum detectable difference in peak positions/intensities. Without such an analysis, the data are consistent with β′ but they do not rule out β.","section":"Sec. III B and Fig. 6"},{"comment":"The Raman evidence against the β′→β transition is also incomplete. The expected signature is a reduction from six Raman-active modes (4Ag+2Bg) to four (2A1g+2Eg), but in unpolarized powder spectra the two modes that should disappear could simply be weak or unresolved. The manuscript reports continuity of mode number and frequencies, but it does not provide a quantitative mode-fitting analysis (line positions, linewidths, integrated intensities, and their pressure dependence across 12–45 GPa) or an explicit detection-threshold estimate. Please add such an analysis and state whether the disappearance of two modes at the level of the experimental noise can actually be excluded.","section":"Sec. III A and Figs. 1-2"},{"comment":"The phase IV assignment as a disordered solid-solution-like orthorhombic structure is presented in the Abstract and Conclusion as a definite result, but it is inferred from powder-pattern indexing and volume-per-atom arguments rather than from a refined structural model. The proposed occupancies of 0.4 and 0.6 for In and Se are derived from stoichiometry and cell volume, not from diffraction intensities, and the paper concedes that no previously reported oP3-type topology fits the cell. This is a plausible hypothesis, but to be load-bearing it requires either a full Rietveld/occupancy refinement with the proposed model or a clear statement that the orthorhombic solid solution is only one of several possible interpretations. At minimum the word 'conclude' in Section V should be softened to match the explicitly preliminary nature of this assignment.","section":"Sec. IV and Table I"}],"minor_comments":[{"comment":"The wording 'remains stable up to 45 GPa' is stronger than the evidence described in Sec. III B, where the authors state that the data show no clear indication of the transition and that the patterns can be indexed with either phase. The abstract, body, and conclusion should adopt a consistent level of certainty, preferably the more cautious formulation.","section":"Abstract and Conclusion"},{"comment":"In the last row of Table I, the column entry '3/5' is unclear: it appears to be a garbled combination of the space group and Z value for phase IV. The space group should be written explicitly and Z separated from it.","section":"Table I"},{"comment":"The text compares the proposed phase IV with 'Bi2Ti3 under pressure'; given the following sentence about Bi2Te3 forming a bcc solid solution, this appears to be a typo for Bi2Te3. Please correct the compound name.","section":"Sec. IV"},{"comment":"The caption states that Le Bail refinements are shown for the β-In2Se3 phase, but the paper argues that the phase up to 45 GPa is β′. Please clarify which structural model is used in each panel, or label the phases consistently with the main text.","section":"Fig. S5"},{"comment":"There are several typographical and wording errors that should be fixed in a careful copyedit, including 'in agrement' in Sec. III B, 'specimens were grained' in Sec. II, and 'A detailed comparison' in Sec. I. None of these affect the science, but they detract from the presentation.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The dataset is solid and the comparison with previous non-hydrostatic studies is valuable. The main issue is the central negative claim: the absence of a β′→β transition is an absence-of-evidence argument, and the manuscript currently does not provide the quantitative model comparison needed to make it convincing. If the authors can add that comparison, or alternatively reframe the claim as 'no evidence for' rather than 'no transition', the paper would be suitable. The phase IV section should also be clearly marked as a tentative proposal unless a structural refinement is added."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a well-executed hydrostatic DAC study that deserves a serious referee. The genuinely new claims are (1) β′-In2Se3 stays stable to 45 GPa with no β′→β transition, and (2) phase IV is a disordered orthorhombic solid solution rather than the cubic defect Th3P4-type structure. Both contradict the sequences in Vilaplana and Zhao, and the He PTM plus simultaneous Raman/XRD is a real step up.\n\nWhat it does well: consistent Raman and XRD across three beamlines, helium as PTM, Le Bail refinements, clear comparison with earlier non-hydrostatic work, and the F–f plot is used appropriately as a descriptive check. The paper is also honest: it states outright that XRD patterns above 12 GPa can be indexed with either β′ or β, and it labels the phase IV structure as preliminary, even acknowledging a new topology with no known oP3 prototype.\n\nThe soft spot is exactly the one the authors themselves expose. The β′→β transition is second-order, group-subgroup, no volume change. That means the F–f plot is not diagnostic; it would stay linear through such a transition. The Raman mode count is the better probe, and the authors’ low-frequency access helps, but they do not show a quantitative mode-area or linewidth analysis across 12–45 GPa. So 'no evidence for the transition' is not the same as 'transition absent'; the data are fully consistent with β′ but also with β if the diagnostic modes are weak or unresolved. The phase IV assignment is based on indexing and volume arguments only, with no Rietveld/Le Bail refinement of the occupancies; the authors say so, but that limits its weight.\n\nProportionally: these are real gaps but they do not sink the paper. The hydrostatic data are the most careful to date on this material, and the contradictions with earlier work are worth publishing even if the final word on the β′/β question needs a quantitative model comparison or a higher-resolution study.\n\nWho it’s for: high-pressure experimentalists and anyone modeling strain effects in 2D ferroelectric In2Se3. I’d bring it to our reading group. I would cite it as the reference for the hydrostatic phase diagram, with a caveat on the no-β claim.\n\nRecommendation: send to peer review. The referee should ask for a quantitative mode-count analysis (or an explicit statement of detection limits) and a clearer separation between 'not observed' and 'absent' in the conclusions.","headline":"A careful hydrostatic re-study of In2Se3 that challenges the reported β′→β transition; the no-transition claim is an absence-of-evidence argument that needs a quantitative model comparison before it can be called secure.","tokens_in":11726,"tokens_out":1871,"would_cite":true,"duration_ms":17127,"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, under truly hydrostatic conditions, In2Se3 follows a simpler high-pressure sequence than previously claimed: the α phase converts to the β′ phase at about 1 GPa, and that β′ phase remains stable up to 45 GPa, with…","keywords":["In2Se3","high pressure","diamond anvil cell","Raman spectroscopy","X-ray diffraction","phase transition","equation of state","ferroelectric"],"falsifier":"Collect high-resolution XRD between 12 and 45 GPa under helium and search for weak Bragg peaks allowed in the monoclinic C2/m β′ cell but absent in the rhombohedral R-3m β cell; their disappearance would falsify the paper's central claim, while their persistence would confirm it.","tokens_in":10574,"feed_emoji":"🔬","tokens_out":5747,"duration_ms":54917,"temperature":0.7,"pith_summary":"This paper reports a combined X-ray diffraction and Raman spectroscopy study of layered ferroelectric α-In2Se3 compressed in helium up to 60+ GPa. It claims the ambient α phase transforms to the monoclinic β′ phase at about 1 GPa, and that this β′ phase—not the previously reported β phase—remains stable all the way to 45 GPa. Above 45 GPa, the material enters a disordered, solid-solution-like orthorhombic phase (phase IV), not the cubic defect Th3P4-type structure assigned in earlier work. The conclusion matters because it implies that earlier reports of a β′→β transition at 5–12 GPa were artifacts of non-hydrostatic pressure media, and that the high-pressure behavior of a candidate 2D ferroelectric is simpler than previously thought.","feed_headline":"In2Se3 skips a reported phase under hydrostatic pressure","feed_subtitle":"Helium-based XRD and Raman show the beta-prime phase holds to 45 GPa, overturning earlier beta-phase claims.","key_machinery":"The central analysis tool is the F–f finite-strain formalism: normalized pressure $F = P/[3f(1+2f)^{5/2}]$ plotted against Eulerian strain $f = [(V_0/V)^{2/3}-1]/2$, where a linear F–f relation indicates a single, stable equation of state without subtle structural modifications. It is used to argue that β′-In2Se3 undergoes no strain-induced change up to 45 GPa. Alongside it, helium as the pressure-transmitting medium maintains near-hydrostatic conditions beyond 50 GPa, and the group-subgroup relation between β′ (C2/m) and β (R-3m) frames why the supposed transition would be second-order and hard to detect.","core_discovery":"The paper establishes, on its own terms, a hydrostatic phase sequence for In2Se3: α (R3m) → β′ (C2/m) at about 1 GPa, with no distinct β (R-3m) phase, followed above 45 GPa by phase IV, a disordered orthorhombic solid solution. The evidence is that the number and frequencies of Raman modes are continuous across 12–40 GPa, that the XRD patterns above 12 GPa can be indexed equally well with β′ or β, and that an F–f equation-of-state analysis shows no deviation from linearity up to 45 GPa. The authors argue that the β′→β transition, being a second-order symmetrization with a group-subgroup relation and no volume change, would be difficult to see in XRD alone, and that previous claims of β rest on non-hydrostatic media and limited low-frequency Raman access. They also reject the cubic defect Th3P4 assignment for phase IV because more Bragg peaks are resolved here, including one low-angle peak the cubic cell cannot index, and the cubic cell gives a physically implausible volume.","pith_inferences":["If the β′ phase is truly stable to 45 GPa, then ferroelectric properties of α-In2Se3 may persist under pressures relevant to strained devices, and earlier reports of pressure-driven loss of ferroelectricity may need re-examination.","The proposed orthorhombic oP3-type phase IV, if confirmed by calculations, would be a new high-pressure topology for III–VI compounds; the Bi2Te3 analogy suggests other layered chalcogenides may form similar disordered solid solutions under hydrostatic compression.","A direct testable extension is to perform ab initio structure prediction for In2Se3 between 45 and 70 GPa to see whether a low-volume orthorhombic solid solution is energetically favored over the defect Th3P4-type cubic cell.","The authors' F–f analysis could be applied to other layered ferroelectrics to distinguish genuine second-order transitions from artifacts of non-hydrostatic stress."],"forward_implications":["The hydrostatic compression sequence of In2Se3 is α → β′ at about 1 GPa, with no further transition until about 45 GPa, so the ferroelectric β′ phase exists over a much wider pressure range than previously thought.","The previously reported β′→β transition at 5–12 GPa is likely an artifact of non-hydrostatic pressure media or limited spectral range, not an intrinsic structural transition.","Phase IV, stable by at least 67 GPa, is a disordered solid-solution-like orthorhombic structure with about 3 atoms per primitive cell, not the cubic defect Th3P4-type structure.","All pressure-induced transformations are reversible at room temperature on decompression, with considerable hysteresis in the reverse IV→β′ transition.","A broad amorphous-like XRD feature above 45 GPa suggests partial amorphization accompanying formation of the disordered IV phase, while Raman continuity rules out decomposition."],"supporting_citations":[{"why":"Previous combined experimental and theoretical high-pressure study of α-In2Se3; supplies the main comparison for Raman mode frequencies, lattice parameters, and the claimed β′→β transition at 12 GPa.","marker":"[11]"},{"why":"Previous XRD study to 59.5 GPa that reported β, an isostructural transition, and cubic defect Th3P4-type phase IV; provides the phase-IV assignment this paper argues against.","marker":"[12]"},{"why":"Earlier structural-evolution study of β′-In2Se3 under pressure; part of the previous α→β′→β sequence that this paper re-examines.","marker":"[10]"},{"why":"Reported a direct α→β transition based on a low number of observed Bragg peaks; used to illustrate that distinguishing β′ from β by XRD alone is resolution-dependent.","marker":"[13]"},{"why":"Documents hydrostatic limits of pressure-transmitting media; justifies using helium to avoid the non-hydrostatic artifacts blamed for earlier β-phase reports.","marker":"[14]"},{"why":"Introduces the finite-strain F–f equation-of-state formalism used to show the absence of subtle structural modification up to 45 GPa.","marker":"[29]"},{"why":"Provides zero-Kelvin compression isotherms of the elements, used for atomic-volume arguments showing the cubic Th3P4-type cell has an implausibly high volume.","marker":"[30]"},{"why":"Shows pressure-induced transition of Bi2Te3 to a bcc structure, used as the analogous case for the disordered solid-solution interpretation of phase IV.","marker":"[34]"}],"fun_headline_variants":["In2Se3 skips β phase under hydrostatic pressure","No β phase for In2Se3 up to 45 GPa","Hydrostatic pressure reveals In2Se3's true phases","In2Se3's β phase missing under helium pressure","β phase of In2Se3 never appears under hydrostatic pressure"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument breaks if the β′→β transition would be invisible in the measured Raman spectra and F–f plot; the authors assume that a subtle second-order symmetrization with no volume change would still produce a detectable change in the number of Raman modes or in the pressure-volume relation.","fun_headline_variants_meta":{"raw":{"variants":["In2Se3 skips β phase under hydrostatic pressure","No β phase for In2Se3 up to 45 GPa","Hydrostatic pressure reveals In2Se3's true phases","In2Se3's β phase missing under helium pressure","β phase of In2Se3 never appears under hydrostatic pressure"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000818,"raw_usage":{"total_tokens":3610,"prompt_tokens":1004,"completion_tokens":2606,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":620,"completion_tokens_details":{"reasoning_tokens":2519}},"tokens_in":620,"tokens_out":2606,"duration_ms":19473,"temperature":1.0,"reasoning_tokens":2519,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T17:21:18.495884+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Collect high-resolution XRD between 12 and 45 GPa under helium and search for weak Bragg peaks allowed in the monoclinic C2/m β′ cell but absent in the rhombohedral R-3m β cell; their disappearance would falsify the paper's central claim, while their persistence would confirm it.","supporting_citations":[{"cited_title":"Vilaplana , author S","cited_arxiv_id":null,"evidence_quote":"Previous combined experimental and theoretical high-pressure study of α-In2Se3; supplies the main comparison for Raman mode frequencies, lattice parameters, and the claimed β′→β transition at 12 GPa."},{"cited_title":"Zhao \\ and\\ author L","cited_arxiv_id":null,"evidence_quote":"Previous XRD study to 59.5 GPa that reported β, an isostructural transition, and cubic defect Th3P4-type phase IV; provides the phase-IV assignment this paper argues against."},{"cited_title":"Liang , author H","cited_arxiv_id":null,"evidence_quote":"Earlier structural-evolution study of β′-In2Se3 under pressure; part of the previous α→β′→β sequence that this paper re-examines."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reported a direct α→β transition based on a low number of observed Bragg peaks; used to illustrate that distinguishing β′ from β by XRD alone is resolution-dependent."},{"cited_title":"Klotz , author J","cited_arxiv_id":null,"evidence_quote":"Documents hydrostatic limits of pressure-transmitting media; justifies using helium to avoid the non-hydrostatic artifacts blamed for earlier β-phase reports."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides zero-Kelvin compression isotherms of the elements, used for atomic-volume arguments showing the cubic Th3P4-type cell has an implausibly high volume."},{"cited_title":"Einaga , author A","cited_arxiv_id":null,"evidence_quote":"Shows pressure-induced transition of Bi2Te3 to a bcc structure, used as the analogous case for the disordered solid-solution interpretation of phase IV."}],"review_version":1}