{"id":"61022a6e-38f8-4238-8767-9316282a5b17","arxiv_id":"2505.22392","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"DFT calculations show weak valley-layer coupling in centrosymmetric FeCl2 that allows electric-field-switchable valley polarization, and a PT-symmetric bilayer that polarizes spontaneously.","lead":"This paper simulates a single layer of the magnetic material FeCl2 and finds a weak connection between its electronic valleys and its layers, so an electric field can create and flip valley polarization. The authors also find that stacking two layers makes this polarization appear without any field, a step toward smaller valley-based memory devices.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Both headline results rest on magnetic energy differences (95 µeV/Fe MAE; 3.3 meV interlayer exchange) below PBE+U accuracy; without U/vdW/hybrid checks, the out-of-plane easy axis and PT-antiferromagnetic bilayer ordering remain unverified.","rationale":"I read the paper in good faith as a DFT-based proposal of weak valley-layer coupling in centrosymmetric FeCl2 and a stacking-induced valley polarization in its bilayer. The symmetry arguments for the sign of the valley splitting under reversal of E and M are sound, and the calculated band structures are internally consistent. The genuinely load-bearing vulnerability is not the conceptual framework but the numerical stability of the magnetic state: the monolayer valley-layer coupling requires out-of-plane magnetization, which is fixed by a 95 µeV/Fe MAE, and the bilayer spontaneous valley polarization requires interlayer AFM ordering, which is favored by only 3.3 meV. Both are within the noise of PBE+U and are not tested against U, vdW, or hybrid functionals. The reader's weakest_assumption identifies the same fragility, so agreement is 'agree'. The recommended verdict is unchanged: the paper merits conditional acceptance pending standard functional-robustness checks, but the central claims are not yet established at the accuracy level the narrative implies.","tokens_in":9138,"tokens_out":13640,"duration_ms":172754,"concrete_test":"Recompute with the same VASP settings: (a) monolayer MAE and valley splitting at E = ±0.25 V/Å for Ueff = 3, 4, and 5 eV, and (b) bilayer AA FM-AFM vs FM-FM total-energy difference with DFT-D3 and with HSE06 (or at least with Ueff = 3 and 5 eV). If the easy axis remains out-of-plane and the FM-AFM ordering remains the ground state by a margin larger than about 10 meV, the concern is resolved; if either ordering flips, the central valley-polarization mechanism is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claims are two linked results: (i) monolayer FeCl2 exhibits weak valley-layer coupling, so an out-of-plane electric field produces a valley splitting of 57.2 meV at 0.25 V/Å, and (ii) AA-stacked bilayer FeCl2 is a PT-antiferromagnet that spontaneously develops about 6 meV of valley polarization. Both results depend on magnetic configurations whose stability is fixed by very small energy differences. In the Bilayer stacking section, FM-AFM interlayer coupling is said to be the ground state by only 3.3 meV relative to FM-FM; AA is preferred over AB by 2.8 meV. In the Computational detail section, the monolayer's out-of-plane easy axis rests on an MAE of 95 µeV/Fe. These values are at or below the typical accuracy of PBE+U; no U-dependence, van der Waals correction, hybrid-functional calculation, or zero-point-energy estimate is reported. If a more accurate treatment flips the easy axis to in-plane, the valley-layer coupling disappears because the mechanism explicitly requires out-of-plane magnetization. If a better functional reverses the interlayer ordering, the PT symmetry is lost and the spontaneous valley polarization vanishes. The paper provides no code or raw data, so the numerical ordering cannot be independently checked. The concept is internally consistent, but the quantitative foundation of both headline results is fragile.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript reports first-principles PBE+U+SOC calculations on a centrosymmetric 1T-FeCl2 monolayer and its AA-stacked bilayer. The authors find that the monolayer exhibits 'weak valley-layer coupling': at the valence-band -K and K valleys, Fe orbitals dominate both valleys, while the upper and lower Cl layers give small opposite contributions. An out-of-plane electric field E breaks inversion symmetry and produces a valley splitting of 57.2 meV at E = 0.25 V/Å, with the sign of the polarization controlled by the product of E and the magnetization M; reversing either E or M reverses the polarization, while reversing both leaves it unchanged. In-plane magnetization removes the effect. The paper introduces a reduction factor α = 0.081 and approximates the splitting as αeEd. For the AA-stacked bilayer with intralayer FM and interlayer AFM order, described as a PT-antiferromagnet, a spontaneous valley splitting of about 6 meV is predicted without an external field.","tokens_in":9532,"tokens_out":9818,"duration_ms":96613,"significance":"If the predicted magnetic ground states and easy axes are robust, the paper provides a conceptually interesting extension of valley-layer coupling to the weak-coupling regime and a stacking-based route to spontaneous valley polarization in zero-net-magnetization systems. The symmetry arguments connecting E and M reversals to valley polarization are internally consistent, and the linear field dependence in Fig. 5(g) is a clear DFT result. The bilayer idea, treating the adjacent layer as a built-in field, is simple and potentially useful for device proposals. The computational setup (PBE+U with U = 4 eV, 21×21×1 k-mesh, 500 eV cutoff) is standard for this class of materials. The paper does not provide input files or raw data, and the headline quantitative claims rest on very small energy differences; nevertheless, the qualitative predictions are falsifiable, including the 57 meV monolayer splitting at 0.25 V/Å and the 6 meV bilayer splitting.","major_comments":[{"comment":"The out-of-plane easy axis is a load-bearing assumption: the authors state that valley-layer coupling and the resulting electric-field valley splitting exist only for out-of-plane magnetization and vanish for in-plane magnetization (main text and FIG.S2). The computed MAE is 95 µeV/Fe, which is at or below the typical accuracy of PBE+U for magnetic anisotropy, and no U-dependence, van der Waals correction, hybrid-functional test, or zero-point estimate is reported. If a more accurate treatment yielded an in-plane easy axis, the monolayer valley-splitting mechanism would not operate in the ground state. Please provide robustness checks of the MAE (e.g., U = 3 and 5 eV, or a hybrid functional) and report the MAE values as a function of electric field from FIG.S4 in the main text.","section":"Computational detail; Crystal structures and valley-layer coupling"},{"comment":"The spontaneous valley polarization in the AA-stacked bilayer relies on the FM-AFM interlayer configuration being the ground state, favored by only 3.3 meV over FM-FM for AA stacking and by 2.7 meV for AB stacking; AA is preferred over AB by 2.8 meV. These energy differences are of the same order as typical PBE+U errors, and the paper does not report vdW-corrected calculations or a U sweep. Because the PT symmetry, and hence the 6 meV valley splitting, disappears if the FM-FM ordering becomes the ground state, the bilayer claim is not yet quantitatively secured. Please test the interlayer ordering with vdW-corrected functionals (e.g., DFT-D3 or optB88-vdW) and a range of U values, and report the resulting valley splitting.","section":"Bilayer stacking-induced valley polarization"},{"comment":"The parameter α = 0.081 is introduced as a reduction factor in the estimate αeEd, but it is not an independent parameter: with d = 2.82 Å and E = 0.25 V/Å, eEd = 705 meV, and the computed splitting is 57.2 meV, so α is exactly the ratio of the two. Using αeEd to 'estimate' the splitting therefore merely restates the DFT point used to define α and cannot validate the linear dependence. The linearity in Fig. 5(g) is a legitimate computational result; please reframe α as an empirical fit to the DFT data or derive it from a microscopic model, and state explicitly which data were used to obtain it.","section":"Electric field-induced valley polarization"}],"minor_comments":[{"comment":"The text 'the valley splitting of FeCl2 will be 4α/0' should read '4α for out-of-plane and 0 for in-plane magnetization'; as written it suggests division by zero.","section":"Electric field-induced valley polarization"},{"comment":"The symbol α is overloaded: it denotes the reduction factor α = 0.081 and also the SOC-related parameter in ΔEV = 4α cosθ. Please use separate symbols for these two quantities.","section":"Electric field-induced valley polarization"},{"comment":"Please specify how the out-of-plane electric field is applied in the slab supercell (e.g., sawtooth potential with dipole correction); the computational detail section gives cutoffs and k-mesh but no field-application method.","section":"Computational detail"},{"comment":"The qualitative claim of weak valley-layer coupling is based on visual inspection of projected band weights in Fig. 3(e–h); quantitative orbital and layer weights at -K and K would make the distinction between weak and strong coupling explicit.","section":"Crystal structures and valley-layer coupling"},{"comment":"The statement that the AFM2 configuration 'converges to a non-magnetic solution' means its energy cannot be compared with the other magnetic orderings as an AFM state; please clarify this in the text.","section":"Crystal structures and valley-layer coupling"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a standard computational prediction paper within the scope of a computational materials science journal. The main concerns are the numerical fragility of the MAE and interlayer exchange energies, and the circular use of the α parameter. If the requested robustness checks and reframing are provided, the paper would be suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid, modest DFT paper. The electric-field valley splitting in monolayer FeCl2 was already reported in the paper's ref 37, and the authors say so. The genuinely new pieces are the weak valley-layer coupling classification (both valleys dominated by the same Fe layer, with small opposing Cl contributions), the E·M sign rule, and the prediction of spontaneous valley polarization in the AA-stacked bilayer as a PT-antiferromagnet. The bilayer result is interesting partly because it contradicts the authors' own earlier proposal that PT-bilayer valley polarization needs a valley-polarized building block; they flag that themselves, which is honest.\n\nWhat it does well: the DFT setup is standard for this subfield (PBE+U with U=4 eV, 21x21x1 k-mesh, SOC, 500 eV cutoff). The layer-projected bands in Fig. 3 do show the weak coupling they claim, and the sign rule under reversal of E and M is clean. The statement that valley-layer coupling vanishes for in-plane magnetization is a clear, testable claim.\n\nNow the soft spots, in proportion. First, both headline results rest on small energy differences: the bilayer interlayer FM-AFM preference is 3.3 meV and the monolayer easy axis rests on a 95 µeV/Fe MAE. Those are below what PBE+U can resolve, and there are no U-dependence, vdW, or hybrid checks. If a better treatment flips the easy axis to in-plane, the coupling vanishes; if the interlayer ordering reverses, the PT symmetry and the 6 meV polarization disappear. That is a genuine robustness gap, but it is a referee question, not an internal inconsistency. Second, the α=0.081 reduction factor is fitted to the same computed splitting it later \"estimates,\" which is mildly circular; the estimate is cosmetic, since the real results are the direct DFT splittings. Also, the same symbol α is reused later for the SOC parameter in ΔE=4αcosθ, which is sloppy notation. Third, no code or data is provided, so the energy orderings cannot be checked independently; this is common in the field but still limits confidence.\n\nWho it is for: researchers in 2D valleytronics, especially the groups working on electric-field control and antiferromagnetic valley physics. It is a useful worked example and an explicit design rule, not a breakthrough. I would send it to peer review, asking for robustness checks — U-dependence of the splitting and MAE, and a better functional for the interlayer energy — and a cleaner separation of fitted versus computed quantities. I would not build my own next project on the 6 meV bilayer polarization without that evidence.","headline":"A competent DFT study whose genuinely new claims — the weak valley-layer coupling framing and the PT-bilayer spontaneous valley polarization — are plausible but rest on small energy differences that need functional checks.","tokens_in":9985,"tokens_out":4332,"would_cite":false,"duration_ms":42049,"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":"Centrosymmetric monolayer FeCl2 shows weak valley-layer coupling, so an out-of-plane electric field produces a 57.2 meV valley splitting at 0.25 V/Å, and AA-stacked bilayer FeCl2 spontaneously polarizes its valleys by about 6 meV without…","keywords":["valleytronics","valley polarization","valley-layer coupling","FeCl2 monolayer","first-principles calculations","PT-antiferromagnet","electric field control","bilayer stacking"],"falsifier":"Measure the interlayer exchange of mechanically stacked or epitaxial bilayer FeCl2 (for example by torque magnetometry or inelastic neutron scattering on bulk analogs) and check whether the antiferromagnetic AA stacking is really preferred; if the ferromagnetic stacking is the ground state, the spontaneous 6 meV valley splitting cannot occur. Alternatively, compute the monolayer with a functional that captures van der Waals and correlation effects more accurately (e.g., a hybrid functional or RPA) and see whether the out-of-plane easy axis and the linear $\\alpha e E d$ splitting survive; a sign change of the magnetic anisotropy energy would remove the valley-layer coupling entirely.","tokens_in":8978,"feed_emoji":"⚡","tokens_out":6682,"duration_ms":60335,"temperature":0.7,"pith_summary":"This paper uses first-principles density functional theory to argue that a centrosymmetric monolayer of iron chloride (FeCl2), which has no inversion-symmetry breaking of its own, still exhibits valley-layer coupling: both valleys are anchored to the middle Fe layer while the upper and lower Cl layers contribute small opposite weights. Because of this weak coupling, an out-of-plane electric field shifts the -K and K valleys in opposite directions, producing a valley splitting of 57 meV at 0.25 V/Å that grows linearly with field and can be switched by reversing either the field or the magnetization. The paper further shows that an AA-stacked bilayer, an antiferromagnet with combined space-time-reversal (PT) symmetry, spontaneously develops about 6 meV of valley polarization without any external field, because each layer acts as a built-in electric field on the other. The broader stake is a valleytronic control mechanism that needs an electric field rather than a magnetic field or broken lattice inversion, and a zero-net-magnetization route to spontaneous valley polarization.","feed_headline":"Electric field splits FeCl2's valleys by 57 meV","feed_subtitle":"Centrosymmetric FeCl2 needs only out-of-plane magnetization; its stacked bilayer polarizes valleys on its own.","key_machinery":"The central object is the weak valley-layer coupling: a layer-resolved orbital character in which the two valleys (-K and K) are both dominated by the middle Fe layer, with opposite small weights from the upper and lower Cl layers. The paper's quantitative machinery is the valley-splitting expression $\\Delta E_V = 4\\alpha\\cos\\theta$ for $d_{x^2-y^2}/d_{xy}$ valence bands, where $\\theta$ measures the magnetization angle from the plane normal, and the linear-in-field estimate $\\alpha e E d$ with the dimensionless reduction factor $\\alpha = 0.081$, which encodes how much weaker the coupling is than the ideal $eEd$ limit. For the bilayer, the machinery is the AA stacking of two ferromagnetic monolayers with antiferromagnetic interlayer coupling, which makes the combined system a PT-symmetric antiferromagnet; the layer-dependent built-in potential plays the role of the external electric field.","core_discovery":"In the language of the paper, monolayer FeCl2 in the 1T phase is centrosymmetric with P symmetry and a ferromagnetic ground state whose easy axis is out of plane. Its valence band maxima at -K and K are dominated by Fe $d_{x^2-y^2}+d_{xy}$ states, with both valleys receiving essentially equal Fe contributions; the upper and lower Cl layers contribute oppositely but with small weight, which is exactly the weak valley-layer coupling regime. An out-of-plane electric field therefore breaks the equivalence of the two valleys linearly, with the splitting described by $\\alpha e E d$ with $\\alpha = 0.081$ and $d = 2.82$ Å; reversing either E or M flips the polarization. When two monolayers are stacked in AA order with antiferromagnetic interlayer coupling, the bilayer is a PT-antiferromagnet whose spin-degenerate bands nonetheless show a spontaneous valley splitting of about 6 meV, because each layer experiences the other as a built-in field with reversed E and M, so both layers polarize the same valley. This contradicts the authors' earlier rule that a PT-antiferromagnetic bilayer's building block must itself have valley polarization.","pith_inferences":["The reduction factor $\\alpha = 0.081$ suggests that the electric field mostly couples to the Cl-derived tail of the wavefunction; substituting Cl with Br or I might tune $\\alpha$ and hence the required field strength, a testable extension the paper does not explore.","At the large fields achievable by dual ionic gating (above 0.4 V/Å), the predicted linear scaling implies a splitting approaching 90 meV, competitive with magnetic-field-induced valley Zeeman splittings; measuring this would connect the calculation to device experiments.","The bilayer result suggests a design rule opposite to the one the authors previously proposed: a PT-antiferromagnetic bilayer can polarize valleys even when its monolayer building block has no valley polarization, provided the stacking reverses both magnetization and built-in field between layers; this could be tested in other 1T transition-metal dihalides."],"forward_implications":["An out-of-plane electric field alone can split the valleys of centrosymmetric FeCl2 by about 57 meV at 0.25 V/Å, with the splitting proportional to field strength.","Valley polarization in FeCl2 can be switched by reversing either the electric field or the magnetization, giving two independent control knobs.","AA-stacked bilayer FeCl2 exhibits spontaneous valley polarization of about 6 meV without an external field, as a PT-antiferromagnet with spin-degenerate bands.","Centrosymmetric materials with out-of-plane magnetization and layer-resolved d-orbital valleys are candidate platforms for weak valley-layer coupling, and can be tuned by stacking engineering.","Because valley-layer coupling vanishes for in-plane magnetization, the effect can serve as a probe of the magnetization direction."],"supporting_citations":[{"why":"Introduces the valley-layer coupling design principle and the strong-coupling picture that the paper extends to the weak regime.","marker":"[34]"},{"why":"Previous first-principles result that an electric field induces valley splitting in FeCl2, which the paper reinterprets through weak valley-layer coupling.","marker":"[37]"},{"why":"Supplies the Hubbard U = 4 eV value and the reference for the optimized lattice constants and magnetic ground state of FeCl2.","marker":"[43]"},{"why":"Experimental growth of monolayer and bilayer 1T-FeCl2 and identification of AA stacking as the ground state.","marker":"[47]"},{"why":"Comparison with 2H-FeCl2, where only magnetization reversal flips the valley polarization, highlighting the extra electric-field control in 1T-FeCl2.","marker":"[49]"},{"why":"Provides the valley-splitting formula $\\Delta E_V = 4\\alpha\\cos\\theta$ used to connect orbital character to the out-of-plane/in-plane magnetization dependence.","marker":"[50]"},{"why":"The authors' earlier rule that PT-antiferromagnetic bilayer valley polarization requires a polarized building block, which this paper's bilayer result overturns.","marker":"[51]"},{"why":"Demonstrates experimentally feasible electric fields above 0.4 V/Å via dual ionic gating, making the predicted field strengths realistic.","marker":"[52]"}],"fun_headline_variants":["Flip E or M to flip FeCl2's valley polarization","Bilayer FeCl2: spontaneous valley polarization without a field","FeCl2's weak coupling: out-of-plane spin enables valley splitting"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The bilayer's spontaneous valley polarization exists only if the antiferromagnetic interlayer coupling is the true ground state of the stacked system, but the calculation places it just 3.3 meV below the ferromagnetic stacking, an energy difference small enough that a different exchange-correlation treatment could reverse it; similarly, the monolayer's out-of-plane easy axis rests on a magnetic anisotropy energy of only 95 µeV per Fe atom.","fun_headline_variants_meta":{"raw":{"variants":["Flip E or M to flip FeCl2's valley polarization","Bilayer FeCl2: spontaneous valley polarization without a field","FeCl2's weak coupling: out-of-plane spin enables valley splitting"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001068,"raw_usage":{"total_tokens":4516,"prompt_tokens":1029,"completion_tokens":3487,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":645,"completion_tokens_details":{"reasoning_tokens":3436}},"tokens_in":645,"tokens_out":3487,"duration_ms":32971,"temperature":1.0,"reasoning_tokens":3436,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:08:21.115853+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the interlayer exchange of mechanically stacked or epitaxial bilayer FeCl2 (for example by torque magnetometry or inelastic neutron scattering on bulk analogs) and check whether the antiferromagnetic AA stacking is really preferred; if the ferromagnetic stacking is the ground state, the spontaneous 6 meV valley splitting cannot occur. Alternatively, compute the monolayer with a functional that captures van der Waals and correlation effects more accurately (e.g., a hybrid functional or RPA) and see whether the out-of-plane easy axis and the linear $\\alpha e E d$ splitting survive; a sign change of the magnetic anisotropy energy would remove the valley-layer coupling entirely.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the valley-layer coupling design principle and the strong-coupling picture that the paper extends to the weak regime."},{"cited_title":"Zhu, J.-T","cited_arxiv_id":null,"evidence_quote":"Previous first-principles result that an electric field induces valley splitting in FeCl2, which the paper reinterprets through weak valley-layer coupling."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Hubbard U = 4 eV value and the reference for the optimized lattice constants and magnetic ground state of FeCl2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Experimental growth of monolayer and bilayer 1T-FeCl2 and identification of AA stacking as the ground state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Comparison with 2H-FeCl2, where only magnetization reversal flips the valley polarization, highlighting the extra electric-field control in 1T-FeCl2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the valley-splitting formula $\\Delta E_V = 4\\alpha\\cos\\theta$ used to connect orbital character to the out-of-plane/in-plane magnetization dependence."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates experimentally feasible electric fields above 0.4 V/Å via dual ionic gating, making the predicted field strengths realistic."}],"review_version":1}