{"id":"237ee19e-20ec-4303-acf6-8f922e53a4e7","arxiv_id":"2505.24186","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Photodoping of bilayer 3R-MoS2 is predicted to cause up to 5% in-plane lattice expansion, enhanced ferroelectric polarization, bandgap reduction, and softening of the interlayer A1g phonon.","lead":"This computational study predicts that shining light on bilayer 3R-MoS2, a sliding ferroelectric, expands its crystal lattice in-plane, strengthens its out-of-plane polarization, and shrinks its electronic bandgap. The results suggest light could be used to control ferroelectric memory states and electronic properties in atomically thin devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted 5% photostrictive expansion depends on an unspecified orbital-occupation scheme in the constrained DFT, so the central claim is not yet reproducible.","rationale":"The reader's weakest assumption correctly identifies the missing specification of OCDFT occupations. This is the most load-bearing point because the entire photostriction magnitude is extracted from OCDFT energy curves. The abstract's contradictory statement about interlayer spacing is a presentational error and does not affect the main numerical claim. The method section is otherwise standard (PBE+optB86b-vdW, Berry phase, NEB), and the qualitative trend of strain-dependent polarization and bandgap is consistent with prior work. Therefore, the appropriate verdict remains conditional on the authors supplying the exact occupation scheme and demonstrating that the predicted expansion is independent of that choice. We agree with the reader's assessment.","tokens_in":107,"tokens_out":3149,"duration_ms":50370,"concrete_test":"Redo the OCDFT calculation for nph = 0.8 e/u.c. using at least two distinctly different occupation schemes that both yield the same total carrier density—for example, (a) electrons added to the lowest unoccupied conduction band at the K point with holes at the valence-band maximum, and (b) a Fermi-Dirac smearing that mimics a hot carrier distribution. Compare the resulting equilibrium in-plane lattice constants. If the lattice constant differs by more than about 0.5% between schemes, the reported 5% expansion is not robust to the occupation choice and the central photostriction claim must be re-evaluated; if it is stable, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim—about 5% in-plane expansion at 0.8 e/u.c.—rests on orbital-constrained DFT (OCDFT) calculations whose occupation scheme is never specified. Supplement Eq. (1) defines ρ_OCDFT(r) with fixed target occupations n_i^OCDFT, but the paper does not state which orbitals (band index, k-point, spin) are emptied and filled to reach the target carrier densities. The real-time TDDFT simulations only determine the total number of excited electrons (nph); they do not determine a unique occupation pattern for a static OCDFT calculation. Different choices—for example, promoting electrons from the valence-band maximum at Γ to the conduction-band minimum at K versus a uniform occupation shift—can produce different charge redistributions and hence different forces and equilibrium lattice constants. Since the 5% expansion is derived from the energy minimum of the OCDFT energy surface (Fig. 2d), an arbitrary occupation choice could either exaggerate or suppress the photostrictive response. Without the explicit occupation pattern or a sensitivity analysis, the headline result is not reproducible and its physical mapping to a laser-induced state is ambiguous.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports first-principles calculations on bilayer 3R-MoS2 as a sliding ferroelectric. Using real-time TDDFT to estimate photodoping carrier densities (nph = 0.1, 0.2, 0.8 e/u.c. for increasing laser intensity) and orbital-constrained DFT (OCDFT) to compute total energies as a function of in-plane lattice constant, the authors claim a pronounced photostrictive response: in-plane expansion up to ~5% at 0.8 e/u.c., a decrease in interlayer spacing, enhanced out-of-plane polarization (0.84 to 1.09 pC/m), a strong bandgap reduction (1.30 to 0.22 eV), and phonon softening of the interlayer A1g mode. The paper further argues that these effects arise from the non-centrosymmetric rhombohedral stacking and proposes applications in optically programmable ferroelectric memories.","tokens_in":13139,"tokens_out":3055,"duration_ms":36947,"significance":"The work addresses an interesting and timely question: whether light-induced carrier injection can dynamically tune both the structural and ferroelectric properties of sliding ferroelectrics. The computational setup is standard for the community (DFT with PBE+optB86b-vdW, 550 eV cutoff, 21x21x1 k-grid; TDDFT with converged real-time propagation) and the inclusion of a comparison with the 2H phase is a strength. If the central claim is substantiated, the predicted 5% photostrictive strain and the associated polarization and bandgap changes would be a notable step toward optical control of slidetronics devices. However, the quantitative predictions depend on an under-specified mapping between the TDDFT carrier density and the OCDFT occupation scheme, and the internal inconsistency between the abstract and Fig. 3(a) currently undermines the phenomenological narrative.","major_comments":[{"comment":"The orbital-constrained DFT occupation scheme is never specified. Eq. (1) defines rho_OCDFT(r) = sum_i n_i^OCDFT phi_i^*(r) phi_i(r), but the manuscript does not state which orbitals (band index, k-point, spin) are emptied and filled to realize the target carrier densities nph = 0.1, 0.2, and 0.8 e/u.c. The real-time TDDFT simulations determine only the total number of excited electrons, not a unique static occupation pattern. Different occupation choices (e.g., excitations at specific k-points versus a uniform occupation shift, spin-conserving versus spin-flip) can lead to different charge redistributions, forces, and equilibrium lattice constants. Since the central claim of a ~5% lattice expansion at 0.8 e/u.c. is derived from the minimum of the OCDFT energy surface (Fig. 2d), the result is not reproducible without this information. Please specify the occupation pattern used or provide a sensitivity analysis over plausible occupation schemes.","section":"Supplement, Eq. (1)"},{"comment":"The abstract states that electron-hole excitation leads to 'substantial in-plane expansion, increased interlayer spacing, and enhanced ferroelectric polarization,' but Fig. 3(a) and the associated main-text discussion report the opposite: the interlayer spacing decreases from 3.06 Å at zero expansion to 2.95 Å at 5% expansion. This is a direct contradiction in the central physical description. The authors should correct the abstract (or, if the interlayer distance actually increases in the dynamically photoexcited state, clearly explain the difference between the static strained geometries in Fig. 3 and the TDDFT dynamics). As written, the inconsistency makes it difficult to interpret the polarization and phonon results.","section":"Abstract and Fig. 3(a)"},{"comment":"There is an internal numerical inconsistency in the reported interlayer distance values. The text first states the relaxed interlayer distance is 3.06 Å and decreases to 2.95 Å at 5% expansion (consistent with the Fig. 3(a) axis label), but later, when comparing 3R and 2H stacking, it states the 3R interlayer distance decreases 'from 3.04 Å to 2.94 Å.' These two sets of values should be reconciled, as the reported polarization and phonon softening depend on the quantitative interlayer relaxation.","section":"Main text, paragraph after Fig. 3(a)"}],"minor_comments":[{"comment":"Typo: 'combing' should be 'combining'.","section":"Conclusion"},{"comment":"Typo: 'The stars denote he energy minima' should be 'the energy minima'.","section":"Figure 2 caption"},{"comment":"The text refers to 'the fluctuation reaches 80.6 fs' and later 'the oscillation period under tensile strain,' but it is not clear whether the quoted quantities are oscillation periods or something else. Please label the quantity explicitly.","section":"Fig. 4(a) and surrounding text"},{"comment":"The abstract mentions 'increased interlayer spacing' while the main text reports a decrease; beyond the major issue, please ensure all occurrences are consistent after revision.","section":"Abstract and main text"}],"recommendation":"major_revision","confidential_remarks":"The central claim is potentially valuable, but the unspecified OCDFT occupation scheme is a load-bearing reproducibility issue that must be resolved. The authors should be asked to state the occupation pattern explicitly and to show that the predicted expansion is robust to reasonable variations in that pattern. The internal inconsistency in the interlayer spacing values also needs correction before the paper can be considered for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The paper is a competently executed first-principles study of photostriction in bilayer 3R-MoS2, a sliding ferroelectric. The new results are the specific numbers: photodoping expands the in-plane lattice by up to 5% at 0.8 e/u.c., raises the out-of-plane polarization from 0.84 to 1.09 pC/m, cuts the bandgap from 1.30 to 0.22 eV, softens the A1g mode from 405 to 369 cm^-1, and modestly raises the switching barrier from 18 to 22 meV. Those trends are physically plausible, and the calculations are careful (550 eV cutoff, 21x21x1 k-grid, optB86b-vdW, Berry phase). The combination of real-time TDDFT to set carrier densities and constrained DFT to map out the energy landscape is a sensible approach.\n\nThe two problems that matter. First, the abstract claims 'increased interlayer spacing' under photodoping, but the main text and Fig. 3(a) show the interlayer distance decreases from 3.06 to 2.95 Å with 5% expansion. That is a direct contradiction that needs correcting. Second, and more important for reproducibility, the constrained-DFT occupation scheme is never specified. The supplement defines rho_OCDFT with target occupations n_i^OCDFT but never says which orbitals are emptied and filled to realize n_ph = 0.1, 0.2, and 0.8 e/u.c. Since the OCDFT energy surface determines the equilibrium lattice constant, different occupation choices could shift the results. The real-time TDDFT only fixes the total number of excited electrons, not the occupation pattern. This is a genuine gap, not a manufactured one. The qualitative photostriction is probably robust, but the quantitative 5% expansion could be sensitive to the occupation choice. The paper would be much stronger with the occupations stated and a sensitivity test (e.g., promoting from VBM to CBM vs. a uniform shift).\n\nThere are also minor inconsistencies: the period from the TDDFT dynamics at 5% expansion (86.6 fs) corresponds to ~385 cm^-1, while Fig. 4(b) reads ~369 cm^-1; and the interlayer distance at 0% is 3.06 Å in Fig. 3(a) but 3.04 Å in the text about the 2H comparison. These are small but should be fixed.\n\nThe novelty is moderate, not high: Gao and Bellaiche (PRL 2024) and Yang and Meng (PRL 2024) already demonstrated optical tuning of polarization in sliding ferroelectrics. But the specific numbers for 3R-MoS2 and the strain-dependence analysis are new. The paper deserves a serious referee, but the authors should be asked for the missing method details and a corrected abstract before it is published.","headline":"A solid but under-specified computational study: the abstract contradicts its own interlayer-spacing data, and the constrained-DFT occupation scheme is never given.","tokens_in":13778,"tokens_out":4910,"would_cite":true,"duration_ms":52898,"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":"Photodoping stretches bilayer 3R-MoS2 and strengthens its sliding ferroelectric polarization.","keywords":["photostriction","sliding ferroelectricity","bilayer 3R-MoS2","real-time time-dependent density functional theory","constrained density functional theory","lattice expansion","ferroelectric polarization","bandgap renormalization"],"falsifier":"Measure the in-plane lattice constant of bilayer 3R-MoS2 with ultrafast electron or X-ray diffraction immediately after a 2 eV pump pulse that creates roughly 0.1 to 0.8 electrons per unit cell: the calculations predict a transient expansion of about 1% to 5%, with the interlayer A1g mode softening from about 405 cm-1 toward 369 cm-1. Seeing no expansion, or an expansion far below 1% at low fluence, would contradict the central claim.","tokens_in":12683,"feed_emoji":"⚡","tokens_out":8873,"duration_ms":86638,"temperature":0.7,"pith_summary":"This paper argues that photoexcited electron-hole pairs in bilayer 3R-MoS2, a sliding ferroelectric whose polarization comes from a lateral shift between layers, do not just change the electronic occupation: they stretch the crystal lattice. Using first-principles calculations, the authors find an in-plane lattice expansion up to about 5% at a photodoping level of 0.8 electrons per unit cell, together with an increase in the out-of-plane ferroelectric polarization from 0.84 to 1.09 pC/m and a drop in the bandgap from 1.30 to 0.22 eV. They also find that the interlayer distance contracts as the lattice expands, a sign of strong coupling between in-plane strain and out-of-plane layer relaxation. The significance is that light, strain, and switchable electric polarization become one coupled system, so an optical pulse could serve as a non-thermal knob for memory, band-structure, and vibrational properties.","feed_headline":"Light stretches sliding ferroelectric MoS2 by up to 5 percent","feed_subtitle":"Photodoping expands the lattice, lifts polarization to 1.09 pC/m, and cuts the bandgap to 0.22 eV.","key_machinery":"The central objects are the rhombohedral (3R) stacking of bilayer MoS2, whose broken inversion symmetry produces a net out-of-plane sliding polarization, and the photostriction mechanism that couples photoexcited carriers to the lattice. The argument is carried by two computational routes: real-time time-dependent density functional theory, which simulates a 2 eV laser pulse and the subsequent coupled electron-ion motion, and orbital-constrained density functional theory, which freezes orbital occupations to model photodoped states and maps their total energy against the in-plane lattice constant. The load-bearing quantities are the energy-versus-lattice-constant curves at fixed carrier densities, the geometric-phase polarization and interlayer distance as functions of biaxial strain, nudged-elastic-band barriers between the two sliding stacking orders, and the real-time interlayer oscillation period of the A1g mode.","core_discovery":"The central claim is that photodoping in sliding ferroelectric bilayer 3R-MoS2 produces a pronounced photostrictive response, and that the strain it creates feeds back into the ferroelectric and electronic degrees of freedom. Orbital-constrained density functional theory places the energy minimum of the photoexcited system at an in-plane lattice constant of 3.30 Å at 0.8 e/u.c., compared with 3.15 Å in the ground state, a roughly 5% expansion; at the same carrier density the geometric-phase polarization rises monotonically from 0.84 to 1.09 pC/m and the indirect gap falls from 1.30 eV to 0.22 eV. The same expansion softens the interlayer A1g breathing-like mode, lengthening its period from 76.0 fs to 86.6 fs and shifting its frequency from about 405 cm-1 to about 369 cm-1, and it raises the sliding energy barrier only modestly, from 18 to 22 meV/u.c., so polarization switching remains accessible. The mechanism is the strong electromechanical coupling inherent to the non-centrosymmetric rhombohedral stacking: photoexcited carriers redistribute interlayer charge, and the weak van der Waals interlayer bonding makes the out-of-plane response unusually sensitive to in-plane expansion.","pith_inferences":["By extension, the same photostriction mechanism should appear in other 3R-stacked transition metal dichalcogenides and in rhombohedral boron nitride, wherever sliding polarization exists; the paper does not compute those cases.","A concrete experimental test would be ultrafast electron or X-ray diffraction after a 2 eV pump: the calculations predict a transient in-plane expansion of about 1% at low fluence and several percent at high fluence, together with a red-shift of the interlayer Raman mode.","The paper's abstract states that photodoping increases interlayer spacing, while the main-text numbers show the relaxed interlayer distance decreasing from 3.06 Å to 2.95 Å under 5% expansion; the sign of this sub-effect is one place the manuscript is internally inconsistent.","The supplemental methods acknowledge a limitation of the coupled electron-ion dynamics: nuclear motion is represented by a single averaged trajectory, which can miss multi-path effects; the authors argue the laser-driven path dominates here."],"forward_implications":["The paper's results imply that a single laser pulse can dynamically shrink the bandgap of bilayer 3R-MoS2 by about 1 eV, moving the optical response from visible to near-infrared.","The strain-enhanced polarization means an in-plane stretch can be read out as a change in ferroelectric polarization, creating a direct strain-light-memory coupling.","Because the interlayer A1g mode softens with expansion, photostriction also tunes phonon frequencies and thus vibrational and thermal properties on ultrafast timescales.","The sliding energy barrier stays low under 5% expansion, so ferroelectric switching remains energetically feasible while the bandgap is being renormalized, making optically programmable ferroelectric memory a plausible target.","The contrast with 2H stacking, which shows much weaker interlayer relaxation and bandgap response under the same strain, identifies broken inversion symmetry as the ingredient that turns lattice expansion into strong ferroelectric and electronic tuning."],"supporting_citations":[{"why":"Provides the experimental realization of fatigue-resistant sliding ferroelectricity in bilayer 3R-MoS2, the material class and system this paper studies.","marker":"[1]"},{"why":"Supplies the density functional theory framework for photostriction in ferroelectrics, the effect this paper transplants to sliding ferroelectrics.","marker":"[18]"},{"why":"Shows light-induced polarization reversal in sliding ferroelectrics, setting the context for optical control of ferroelectric order.","marker":"[21]"},{"why":"Demonstrates ultrafast light-induced switching of sliding polarization in van der Waals bilayers, another photo-ferroelectric coupling baseline.","marker":"[22]"},{"why":"Establishes that photoexcitation can tune out-of-plane polarization in 3R-MoS2, the specific system studied here.","marker":"[23]"},{"why":"Provides experimental observation of photoinduced lattice expansion and carrier diffusion in WS2, motivating the nonthermal expansion mechanism.","marker":"[24]"},{"why":"Foundational formulation of time-dependent density functional theory, the method used for the laser-excitation dynamics.","marker":"[25]"},{"why":"Describes the real-time local-basis implementation of time-dependent density functional theory used to propagate coupled electron-ion dynamics.","marker":"[29]"},{"why":"Provides the experimental lattice constant of 3R-MoS2 used to validate the ground-state structure.","marker":"[34]"}],"fun_headline_variants":["Photodoping stretches bilayer MoS2, boosting ferroelectric polarization","Photodoping stretches sliding ferroelectric, boosting polarization and shrinking gap","Carrier doping expands MoS2 lattice, tunes ferroelectric polarization","Photostriction in bilayer MoS2: light stretches lattice, boosts polarization","Light-induced lattice expansion enhances polarization in sliding ferroelectrics"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that a fixed pattern of orbital occupations in constrained density functional theory faithfully represents the electronic state a real laser pulse creates, and the paper never specifies which orbitals are emptied and filled for each carrier density, so a different occupation pattern could change the predicted expansion and polarization shift.","fun_headline_variants_meta":{"raw":{"variants":["Photodoping stretches bilayer MoS2, boosting ferroelectric polarization","Photodoping stretches sliding ferroelectric, boosting polarization and shrinking gap","Carrier doping expands MoS2 lattice, tunes ferroelectric polarization","Photostriction in bilayer MoS2: light stretches lattice, boosts polarization","Light-induced lattice expansion enhances polarization in sliding ferroelectrics"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000988,"raw_usage":{"total_tokens":4224,"prompt_tokens":1018,"completion_tokens":3206,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":634,"completion_tokens_details":{"reasoning_tokens":3122}},"tokens_in":634,"tokens_out":3206,"duration_ms":24080,"temperature":1.0,"reasoning_tokens":3122,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:31:59.409355+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the in-plane lattice constant of bilayer 3R-MoS2 with ultrafast electron or X-ray diffraction immediately after a 2 eV pump pulse that creates roughly 0.1 to 0.8 electrons per unit cell: the calculations predict a transient expansion of about 1% to 5%, with the interlayer A1g mode softening from about 405 cm-1 toward 369 cm-1. Seeing no expansion, or an expansion far below 1% at low fluence, would contradict the central claim.","supporting_citations":[{"cited_title":"Bian et al","cited_arxiv_id":null,"evidence_quote":"Provides the experimental realization of fatigue-resistant sliding ferroelectricity in bilayer 3R-MoS2, the material class and system this paper studies."},{"cited_title":"Paillard, B","cited_arxiv_id":null,"evidence_quote":"Supplies the density functional theory framework for photostriction in ferroelectrics, the effect this paper transplants to sliding ferroelectrics."},{"cited_title":"Yang and S","cited_arxiv_id":null,"evidence_quote":"Shows light-induced polarization reversal in sliding ferroelectrics, setting the context for optical control of ferroelectric order."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates ultrafast light-induced switching of sliding polarization in van der Waals bilayers, another photo-ferroelectric coupling baseline."},{"cited_title":"Gao and L","cited_arxiv_id":null,"evidence_quote":"Establishes that photoexcitation can tune out-of-plane polarization in 3R-MoS2, the specific system studied here."},{"cited_title":"Wang et al., Lattice Expansion Enables Large Surface Carrier Diffusion in WS2 Monolayer, ACS Energy Lett","cited_arxiv_id":null,"evidence_quote":"Provides experimental observation of photoinduced lattice expansion and carrier diffusion in WS2, motivating the nonthermal expansion mechanism."},{"cited_title":"Runge and E","cited_arxiv_id":null,"evidence_quote":"Foundational formulation of time-dependent density functional theory, the method used for the laser-excitation dynamics."},{"cited_title":"Schonfeld, J","cited_arxiv_id":null,"evidence_quote":"Provides the experimental lattice constant of 3R-MoS2 used to validate the ground-state structure."}],"review_version":1}