{"id":"a1d127f5-403d-4cc1-93dd-c7c2db2202c5","arxiv_id":"2411.15493","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In bilayer CrSBr, intralayer excitons bind at about 200 meV and direct trions at 18-20 meV, while the interlayer trion with electrons on opposite layers is essentially unbound at realistic spacings.","lead":"This paper calculates how strongly electrons and holes bind into excitons and trions in CrSBr, a layered magnetic semiconductor with strongly directional electronic properties. It finds intralayer trion binding near 18-20 meV, matching recent measurements, and predicts which interlayer trion configurations are stable.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Acknowledged breakdown of the macroscopic screening model at the relevant length scales leaves the 18–20 meV direct-trion binding and the X'_i^- unbound prediction insufficiently secured.","rationale":"The paper's central claim is the assignment of the experimentally observed trion in Ref. [15] to the intralayer X_d^- state with binding energy 18–20 meV, and the prediction that the split-layer trion X'_i^- is unbound at realistic interlayer distances. For this claim to hold, the effective-mass Hamiltonian with the bilayer Rytova–Keldysh potential must be quantitatively reliable for states whose radii (r_x≈0.6 nm, r_y≈1.8 nm) and interlayer distance (d≈0.8 nm) are comparable to the lattice constant. The authors explicitly disclaim this in Sec. III B. Within the model, the most avoidable source of error is the neglect of the strongly anisotropic in-plane polarizability (factor ≈2), which the paper dismisses with a qualitative statement. Because the trion binding energy is a small difference of large Coulomb energies, a 10–20% change in the short-range potential along x could shift the claimed 18–20 meV window or alter the near-zero binding of X'_i^-, potentially changing the species assignment. This is an internal correctness risk, not a disagreement with consensus. The reader's CONDITIONAL verdict already accounts for the broader model uncertainty, so the verdict need not change; the proposed anisotropic-screening calculation would settle whether the specific neglect is benign. No code or data were provided, so an independent numerical rerun is not possible; the anisotropic-screening test is the most direct reproducible check using formulas already in the paper.","tokens_in":20633,"tokens_out":15030,"duration_ms":128008,"concrete_test":"Recompute the direct (X_d^-) and split-layer (X'_i^-) trion binding energies of Fig. 7(b) using the anisotropic screening potential from Eq. (13) with the anisotropic polarizabilities of Ref. [3] (αx/αy ≈ 2), keeping all other parameters fixed (ε=4.5, r0=3.9 nm, d=0.8 nm). If X_d^- binding moves outside the 14–20 meV range or X'_i^- becomes bound, the paper's identification of the trion in Ref. [15] as intralayer X_d^- is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that the trion observed in Ref. [15] is the intralayer X_d^- with binding energy 18–20 meV, while the split-layer trion X'_i^- is unbound—rests on a model the authors explicitly disclaim in Sec. III B: 'the effective mass approximation and macroscopic description of the screening by a dielectric between the layers are not entirely applicable.' At the computed exciton radii (r_x≈0.6 nm, r_y≈1.8 nm) and d≈0.8 nm, the short-range part of the Rytova–Keldysh potential, which controls the tightly bound X_d^- and the delicate balance for X'_i^-, is not derived from a microscopic dielectric function. The paper also neglects the in-plane anisotropy of the CrSBr polarizability (a factor ≈2 per Ref. [3]) while asserting it 'does not change the results much' without showing a calculation. Since the trion binding energy is only ~5% of the exciton binding energy, a localized change in the potential along x could shift the 18–20 meV window significantly and could bind X'_i^-, which was found to have near-zero binding for d≳0.25a_B^2D. This is a correctness risk, not a consensus disagreement: the same claim could fail even if all parameters are exactly as stated.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a theory of neutral and charged excitons in anisotropic two-dimensional semiconductors, applied to CrSBr monolayers and bilayers. Using an effective-mass Hamiltonian with Rytova-Keldysh and bilayer-screened Coulomb interactions, the authors compute exciton and trion binding energies variationally and by finite-element methods, obtaining a direct exciton binding energy around 200 meV and a direct intralayer trion binding energy of 18–20 meV for the assumed parameters. They also study indirect (interlayer) excitons and trions, finding X_i^- binding of about 2–3 meV and X_i'^- essentially unbound at realistic interlayer distances. The paper additionally addresses radiative decay, long-range exchange effects on exciton dispersion, and Fermi-polaron/Suris-tetron correlations, and compares the trion binding estimates with the experiments of Ref. [15].","tokens_in":20832,"tokens_out":7780,"duration_ms":68160,"significance":"If the quantitative assignments are correct, the paper resolves the nature of the trion observed in CrSBr bilayers (intralayer X_d^-) and provides a falsifiable prediction that the split-layer trion X_i'^- is unbound. The work is also useful as a general framework for excitons and trions in strongly anisotropic 2D semiconductors. Strengths include the cross-check of variational trion energies against finite-element calculations for isotropic cases, the explicit discussion of the model's limitations, and the analytic treatment of quasi-1D asymptotics and Fermi-polaron correlations. The main quantitative claim, however, rests on a model whose short-range screening and effective-mass description the authors themselves state are not entirely applicable at the relevant length scales, so the specific 18–20 meV assignment and the X_i'^- unbound prediction require additional robustness checks before they can be considered definitive.","major_comments":[{"comment":"This paragraph explicitly states that 'the effective mass approximation and macroscopic description of the screening by a dielectric between the layers are not entirely applicable' because d and r0 are on the order of the unit cell. The central quantitative claim—that the trion observed in Ref. [15] is the intralayer X_d^- with 18–20 meV binding and that X_i'^- is unbound—is controlled precisely by the short-range part of the Rytova-Keldysh potential and by small differences of large energies. The manuscript should provide a quantitative robustness test, for example by varying r0 and d over their plausible ranges, by evaluating the anisotropic polarizability case of Eq. (13), or by benchmarking against a microscopic dielectric function. Without such a test, the specific assignment to Ref. [15] is not secured to the precision implied by the conclusion.","section":"Sec. III B (paragraph beginning 'We stress that here both the interlayer distance d...')"},{"comment":"The authors state that the variational accuracy of the absolute trion energy is better than 5–10% but that the binding-energy accuracy is only 20–30% because it is a difference of two large energies. For the quoted X_d^- binding of 18–20 meV, this implies an uncertainty of roughly ±4–6 meV. The conclusion that the measured trion in Ref. [15] is compatible with this value, and the contrast with the 2–3 meV X_i^- binding, should be presented with this uncertainty explicitly, for instance as error bars in Fig. 7(b). Without that, the 'reasonable agreement' claim is stronger than the stated numerical accuracy warrants.","section":"Sec. III A (text following Eq. (19))"},{"comment":"The neglect of the in-plane anisotropy of the CrSBr polarizability is justified only by the assertion that it 'does not change the results much,' citing a factor-of-2 variation of the susceptibility in Ref. [3]. No calculation is shown. Since the trion binding energy is only about 5% of the exciton binding energy and the potential enters in its logarithmic short-range regime (r0≈3.9 nm > rx≈0.6 nm), a factor-of-2 change in r0 along x could shift the 18–20 meV window by an amount comparable to the claimed accuracy. The authors should either perform the anisotropic calculation using Eq. (13) with αx ≠ αy or explicitly bound the resulting shift.","section":"Sec. III A (first paragraph, polarizability anisotropy)"},{"comment":"The statement that X_i'^- 'is unlikely to be observed in bilayer CrSBr' is categorical, but the calculation finds its binding energy dropping to negligible values for d≳0.25 a_2D^B ≈ 0.28 nm, well below the nominal d≈0.8 nm. Because this near-threshold behavior is controlled by the same short-range potential whose macroscopic description the authors disclaim, the prediction should be reported as an extrapolation with a stated uncertainty rather than as a definitive no-observation statement. A quantitative estimate of how the X_i'^- binding depends on the uncertainty in r0, d, and the anisotropic polarizability is needed to support the conclusion.","section":"Sec. III B and Conclusion"}],"minor_comments":[{"comment":"The phrase 'few-patricle Hamilonian' should read 'few-particle Hamiltonian'.","section":"Introduction"},{"comment":"The second kinetic term in the denominator is written with k_x^2 but should be k_y^2, consistent with the integration over k_y and with Eq. (B2).","section":"Eq. (B5)"},{"comment":"The caption refers to 'dashed vertical lines in panels (b) and (d)', but the printed figure contains only panels (a) and (b); please update the caption or the panel layout.","section":"Fig. 7 caption"},{"comment":"The condition 'K ⩽ qb' should use the magnitude of the two-dimensional wavevector, e.g., |K| ≤ qb, to avoid ambiguity with the component Ky used in the same equation.","section":"Sec. IV, Eq. (21)"},{"comment":"The legend appears to contain two identical entries for 'μ_x/μ_y = 1, numerical calc.'; please disambiguate which curve corresponds to the finite-element result.","section":"Fig. 5(a) legend"}],"recommendation":"major_revision","confidential_remarks":"The comparison with experiment is made against Ref. [15], which shares several co-authors with the present paper. This is not a correctness problem by itself, but it means the 'reasonable agreement' statement is not an independent test; the authors may wish to state the overlap explicitly. The manuscript is within the scope of a condensed-matter physics journal, and the central framework is sound; the requested revisions concern robustness and precision of the headline quantitative claims rather than the overall validity of the approach."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is a credible order-of-magnitude theory paper for CrSBr bilayers, not a precision calculation. The authors are upfront about that. The new piece is the application to spin-polarized CrSBr bilayers, including interlayer trion species that previous anisotropic-material studies (black phosphorus, TiS3) didn't treat. The main result is the assignment of the trion observed in Ref. [15] to the intralayer X_d^- state with 18–20 meV binding, while the split-layer trion X'_i^- is essentially unbound at realistic interlayer distances.\n\nWhat they do well: the Hamiltonian is standard, the variational and finite-element checks for isotropic masses agree, and they state the accuracy honestly—20–30% for trion binding energies. They do not fit the trion data; r0 is calibrated to the monolayer exciton, and the trion predictions follow from the model. So there is no circularity in the trion binding itself.\n\nThe soft spots are real but not disqualifying. The paper itself admits in Sec. III B that the effective-mass approximation and macroscopic screening description \"are not entirely applicable\" at the computed exciton radii (about 0.6 nm by 1.8 nm) and interlayer distance 0.8 nm. That is the regime that controls the 18–20 meV trion binding and the near-zero binding of X'_i^-. The Rytova–Keldysh short-range part is not derived from a microscopic dielectric function, and the in-plane anisotropy of the polarizability (factor ~2) is neglected with a hand-wavy \"does not change results much\" and no calculation. Since the trion binding is only ~5% of the exciton binding, a localized change in the potential along x could shift the window and could bind X'_i^-. So the central assignment is plausible but not fully secured. Also, the comparison experiment shares authors with this paper; that is a mild concern, not a fatal one, because the calculation is not tuned to the trion line. No code or data are shipped, so the numbers can't be independently re-run, but the method is standard enough that this is a minor issue.\n\nWho is this for: people working on CrSBr magneto-optics and 2D magnetic semiconductors. They will want this as a reference for what the intralayer vs interlayer trion assignment looks like. It deserves a serious referee; the methods are sound, the caveats are stated, and the predictions are testable. My recommendation: send it to review, but ask the authors to show the sensitivity of the trion binding to the polarizability anisotropy and to discuss the short-range screening limitation more quantitatively. That would strengthen the paper without changing its scope.","headline":"Plausible order-of-magnitude theory for CrSBr trions, with an honest but unresolved short-range screening caveat; deserves review.","tokens_in":21476,"tokens_out":3332,"would_cite":true,"duration_ms":31212,"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":"The trion observed in CrSBr bilayers is the direct intralayer state, with 18–20 meV binding energy.","keywords":["CrSBr","exciton binding energy","trion","Rytova–Keldysh potential","anisotropic effective mass","bilayer van der Waals heterostructure","interlayer exciton","Fermi polaron"],"falsifier":"A decisive observation would be a gated hBN-encapsulated CrSBr bilayer measurement of the exciton–trion splitting at low electron density: a trion binding energy outside the 14–20 meV range, or the appearance of a second charged-exciton line with a binding energy of several meV at an interlayer distance near 0.8 nm, would contradict the paper's assignment that the observed trion is the direct intralayer $X_d^-$ and that the split-layer trion is essentially unbound.","tokens_in":20397,"feed_emoji":"🧲","tokens_out":10359,"duration_ms":86495,"temperature":0.7,"pith_summary":"This paper asks what Coulomb-bound electron–hole and two-electron–hole complexes look like in a two-dimensional semiconductor whose electron and hole masses are very different along the two in-plane directions, taking CrSBr bilayers as the concrete material. Using an effective-mass Hamiltonian with a screened Rytova–Keldysh interaction generalized to two polarizable layers, it finds a direct intralayer exciton binding energy of about 200 meV and a direct intralayer trion binding energy of 18–20 meV, which it regards as consistent with recent experiments. It also finds that interlayer complexes lose binding energy quickly with separation: the same-layer indirect trion has only about 2–3 meV binding at realistic distances, and the split-layer trion is essentially unbound. The paper concludes that the trion seen in the experiment is the direct intralayer one, and it gives radiative decay rates, exchange-induced anisotropic dispersions, and Fermi-polaron energy shifts for the same model.","feed_headline":"CrSBr trion is intralayer, at 18–20 meV binding","feed_subtitle":"A screened Coulomb model with strongly anisotropic masses matches the measured charged exciton and rules out interlayer states.","key_machinery":"The central object is the bilayer Rytova–Keldysh potential, a screened Coulomb interaction in which each CrSBr layer is a polarizable dielectric sheet; its same-layer and cross-layer Fourier transforms, Eqs. (11a) and (11b), encode the dependence on interlayer distance $d$ and reduce to the monolayer potential as $d\\to\\infty$. Binding energies come from solving the exciton and trion Schrödinger equations with this potential and strongly anisotropic kinetic terms, using finite-element calculations as a numerical check and a variational envelope—an exponential with separate length parameters along $x$ and $y$, plus a correlation factor for the two identical carriers—for the trions. The mass anisotropy ratio $M=\\mu_x/\\mu_y\\approx20$ is what elongates the exciton wavefunction and controls how steeply indirect binding energies fall with $d$.","core_discovery":"Within a minimum three-band $\\mathbf{k}\\cdot\\mathbf{p}$ model, the paper claims that CrSBr has strongly anisotropic carrier masses ($m_e^x=7.31m_0$, $m_e^y=0.14m_0$; $m_h^x=2.84m_0$, $m_h^y=0.45m_0$), hence a reduced-mass ratio $\\mu_x/\\mu_y\\approx 20$. Solving the relative-motion Schrödinger equation for excitons and trions with the bilayer-screened Rytova–Keldysh potential yields a direct exciton binding energy around 200 meV and a direct trion binding energy of 14–20 meV (quoted as 18–20 meV in the conclusion) for hBN-encapsulated bilayers, in reasonable agreement with measured values. The spatially indirect exciton remains bound at all interlayer distances, but the split-layer trion $X_i^{\\prime-}$ loses binding already for $d\\gtrsim0.25\\,a_B$ and is unlikely to be observable, while the indirect trion with both electrons in one layer has about 2–3 meV binding for $d=0.7$–$0.9$ nm. The central assignment is therefore that the trion observed in Ref. [15] is the direct intralayer trion $X_d^-$.","pith_inferences":["Read as a general trend beyond CrSBr, the distance-dependence curves imply that in any strongly anisotropic van der Waals magnet the charged exciton most likely to be observed is the intralayer trion, because interlayer trions lose binding energy rapidly once carriers sit in different layers.","The predicted anisotropy of exciton propagation could be tested by time-resolved or spatially resolved photoluminescence of an expanding exciton cloud in monolayer or bilayer CrSBr, independently of the trion assignment.","Tuning the interlayer distance with pressure or twist angle should leave the direct trion binding nearly constant while moving the same-layer indirect trion through its few-meV binding window, offering a clean experimental knob to separate intra- and interlayer charged states."],"forward_implications":["For hBN-encapsulated CrSBr bilayers, the trion line seen in photoluminescence should be assigned to the direct intralayer trion with 18–20 meV binding energy, and no split-layer interlayer trion should appear at realistic interlayer distances.","Direct intralayer exciton and trion binding energies increase with interlayer distance because the second layer contributes less screening, while indirect (interlayer) binding energies decrease with distance.","The same-layer indirect trion, with both electrons in one layer and the hole in the other, remains weakly bound, about 2–3 meV at $d=0.7$–$0.9$ nm, and is the most plausible interlayer charged state.","The bright exciton emits only in $y$ polarization, and at finite momentum its dispersion is anisotropic, with group velocity larger along the magnetic easy $b$-axis, which should be visible in exciton propagation."],"supporting_citations":[{"why":"Supplies the three-band $\\mathbf{k}\\cdot\\mathbf{p}$ model and the anisotropic effective masses used throughout the CrSBr calculations.","marker":"[3]"},{"why":"Reports the measured doping-controlled trion spectra in few-layer CrSBr; the calculated direct trion binding of 18–20 meV is compared against this experiment.","marker":"[15]"},{"why":"Derives the screened Coulomb potentials for charge carriers in the same and in different layers of a bilayer, the basis of Eqs. (11a)–(11b).","marker":"[32]"},{"why":"Provides the variational trial functions for excitons and trions that this work extends to anisotropic effective masses.","marker":"[21]"},{"why":"Supplies experimental estimates of the large exciton binding energy and radii in CrSBr used to benchmark the ~200 meV and nanometer-scale results.","marker":"[6]"},{"why":"Establishes the Rytova–Keldysh treatment of neutral and charged excitons in monolayer transition metal dichalcogenides that this work adapts to bilayers.","marker":"[31]"}],"fun_headline_variants":["CrSBr trion is intralayer, 18–20 meV binding","Anisotropic masses keep CrSBr trion intralayer","Direct trion outbinds interlayer state in CrSBr","CrSBr trion binding 18–20 meV, intralayer origin","Screened Coulomb model assigns CrSBr trion to one layer"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that electrons and holes form weakly bound effective-mass particles moving in a smoothly screening dielectric, even though the calculated exciton sizes (about 0.6 nm and 1.8 nm) and the 0.8 nm interlayer distance are comparable to the unit-cell size.","fun_headline_variants_meta":{"raw":{"variants":["CrSBr trion is intralayer, 18–20 meV binding","Anisotropic masses keep CrSBr trion intralayer","Direct trion outbinds interlayer state in CrSBr","CrSBr trion binding 18–20 meV, intralayer origin","Screened Coulomb model assigns CrSBr trion to one layer"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000246,"raw_usage":{"total_tokens":1534,"prompt_tokens":935,"completion_tokens":599,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":551,"completion_tokens_details":{"reasoning_tokens":499}},"tokens_in":551,"tokens_out":599,"duration_ms":5027,"temperature":1.0,"reasoning_tokens":499,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:13:30.034741+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive observation would be a gated hBN-encapsulated CrSBr bilayer measurement of the exciton–trion splitting at low electron density: a trion binding energy outside the 14–20 meV range, or the appearance of a second charged-exciton line with a binding energy of several meV at an interlayer distance near 0.8 nm, would contradict the paper's assignment that the observed trion is the direct intralayer $X_d^-$ and that the split-layer trion is essentially unbound.","supporting_citations":[{"cited_title":"Klein, B","cited_arxiv_id":null,"evidence_quote":"Supplies the three-band $\\mathbf{k}\\cdot\\mathbf{p}$ model and the anisotropic effective masses used throughout the CrSBr calculations."},{"cited_title":"Tabataba-Vakili, H","cited_arxiv_id":null,"evidence_quote":"Reports the measured doping-controlled trion spectra in few-layer CrSBr; the calculated direct trion binding of 18–20 meV is compared against this experiment."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Derives the screened Coulomb potentials for charge carriers in the same and in different layers of a bilayer, the basis of Eqs. (11a)–(11b)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the variational trial functions for excitons and trions that this work extends to anisotropic effective masses."}],"review_version":1}