{"id":"0a5760ae-e5da-4876-adc4-daf54454273e","arxiv_id":"2411.18815","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Light-induced orbital magnetism dominates the spin response in most non-magnetic transition metals, and the full periodic trend is mapped from first principles.","lead":"Simulations predict that laser light induces small magnetic moments in 24 transition metals, with the orbital component usually larger than the spin component. The results form a reference map for experiments on ultrafast light-controlled magnetism and THz spintronics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The factor-of-2 prefactor for the orbital response in Eq. (1) is the most load-bearing unresolved point: it silently scales every reported δL value, and the Methods provides no derivation, citation, or operator-normalization statement.","rationale":"The reader's weakest-assumption analysis identified exactly the same point: an unexplained factor of 2 that multiplies every orbital moment. I agree that this is the most load-bearing concern because it is the only place in the manuscript where a global numerical prefactor is asserted without derivation, and it directly controls the quantitative map that is the paper's main deliverable. The concern is not that the factor is demonstrably wrong; rather, the paper's normalization convention is under-specified, so an independent reader cannot verify whether the reported δL values are physical without reconstructing the derivation. I considered whether the strong lifetime-broadening dependence in Figs. 4 and 5 is a more serious issue. It is significant, but the authors explicitly present Γ-dependence for representative materials and warn that special care is needed when comparing different disorder implementations; the factor of 2 is instead silent, global, and easy to test. I also weighed whether the central qualitative claim would collapse if the factor were removed. For most nonmagnetic metals, |δL| would still exceed |δS| after halving, so the orbital-dominance trend is likely robust; however, the promise of a complete quantitative reference map is directly affected, and a few entries with near-unity orbital-to-spin ratios would change their message. The spin-channel values are independently supported by exact reproduction of Ref. [21], and the comparison with Ref. [20] gives additional confidence that the overall scale is reasonable, which is why the verdict should not be escalated beyond conditional acceptance. The appropriate remedy is a short derivation or a reference stating the operator normalization, after which the quantitative map could be accepted.","tokens_in":21371,"tokens_out":7229,"duration_ms":74823,"concrete_test":"Independently derive the orbital version of Eq. (1) from the Keldysh response tensor φ_ijk in Eq. (14) of Ref. [21], using L_i = (r×p)_i and the standard spin operator S_i = σ_i/2, and check whether the orbital response acquires a factor of 2 relative to the printed spin expression when both are expressed in units of μ_B. If the analytic derivation is inconclusive, recompute δLz for fcc Pt at ħω = 1.55 eV with the code path that evaluates φ_ijk using Wannier matrix elements of L_z directly, omitting the extra factor of 2, and compare with Table III. A result exactly half of -7.4×10⁻³ μ_B per unit cell confirms that the factor is a consistent normalization; a different or material-dependent ratio would indicate that the prefactor statement is a real error.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing unresolved point is the factor-of-2 statement in Methods, immediately after Eq. (1): \"For the orbital response, the prefactor in Eq. (1) must be multiplied by an additional factor of 2.\" No derivation or citation accompanies this statement, and the paper does not specify what operator normalization Eq. (1) assumes for the spin channel. Since every reported δLz, δLx, and all figures and tables derived from them inherit this prefactor, the quantitative map that the abstract offers as a \"key reference point\" depends on this one unexplained constant. The factor is global, so it cannot be a material-dependent correction; however, if it is a convention mismatch, for example with the spin operator written as Pauli matrix/2 while the orbital operator is written as r×p and both are converted to μ_B, the reader cannot check the reported values without redoing the derivation. Removing the factor would halve all δL values; for most nonmagnetic entries the orbital-dominance conclusion would survive, but for near-unity-ratio entries such as hcp Os at 1.55 eV, where δLz/δSz ≈ 2.4, the qualitative statement becomes much less clean. The concern is therefore not that the factor is known to be wrong, but that the central quantitative claim is anchored to an unverified prefactor. Other identified issues, including the strong Γ-dependence shown in Figs. 4 and 5, are at least explicitly characterized in the text; this factor is not addressed at all.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports first-principles calculations of the inverse Faraday effect in 24 transition metals of groups IV-XI across the 3d, 4d, and 5d series, using the Keldysh formalism with Wannier-interpolated band structures. It gives spin and orbital contributions to the light-induced magnetic moment for circular polarization in the xy and yz planes at photon energies 0.25 eV and 1.55 eV, with a fixed lifetime broadening of 25 meV. The central claims are that orbital and spin responses can differ by one to two orders of magnitude in nonmagnetic metals, that crystal-field splitting and spin-orbit coupling control the relative size and sign of the two channels, and that the resulting material map provides a reference for optical magnetism.","tokens_in":21669,"tokens_out":7074,"duration_ms":62064,"significance":"If correct, this is a useful systematic map: the work covers a uniform set of elements, separates spin and orbital channels, and connects the response to k-space anatomy and broadening dependence. Strengths include the consistent Wannier construction, the reproduction of previous spin-channel values for magnetic metals, and explicit analysis of polarization and crystal-structure anisotropy. The quantitative value of the map is currently constrained by one unexplained prefactor in the orbital channel and by the lack of a systematic uncertainty assessment, so the significance as a 'key reference point' depends on resolving those points.","major_comments":[{"comment":"The sentence immediately after Eq. (1) in Methods — 'For the orbital response, the prefactor in Eq. (1) must be multiplied by an additional factor of 2' — is the single most load-bearing step in the paper, yet it comes with no derivation, no citation, and no statement of operator convention. Eq. (1) is written for either O_i = L_i or S_i, so a reader cannot tell whether the factor of 2 is a property of the orbital operator, a consequence of the spin operator being defined as σ/2, or a normalization choice in φ_{ijk} of Ref. [21]. This factor multiplies every δL value in Figs. 1-3, 6-7 and Tables II-III, including the conclusion that orbital IFE dominates spin IFE by one to two orders of magnitude in nonmagnetic metals. Please justify the factor by deriving it from the operator definitions or by citing the exact convention, and state what normalization is used for the spin operator. If the factor is a convention artifact, demonstrate that the same convention was applied to the comparison with Ref. [20] in Table I.","section":"Methods, after Eq. (1)"},{"comment":"Validation of the method is restricted to the spin channel. The text reports that δSz and δSx for Fe, Co, Ni with yz polarization reproduce Ref. [21], but no equivalent check is given for δL. Table I compares total moments (δLz + δSz) with Ref. [20] at 1.55 eV; this comparison cannot separate errors in the spin and orbital channels, and the two notable disagreements (Au, left-handed Co) are attributed to 'difference in the computational methods' without analysis. Given that the factor of 2 in the orbital channel is unresolved and the orbital response is the paper's principal new content, an independent material-by-material comparison of δL, or at least a decomposition of the Table I comparison into spin and orbital parts, is needed before the map can serve as a quantitative reference.","section":"Results, 'Light-induced magnetism in transition metals'; Table I"},{"comment":"All entries in the main map (Figs. 1-2 and Tables II-III) are computed at Γ = 25 meV, but the paper's own Figs. 4 and 5 demonstrate a strong, non-monotonic dependence on Γ — for example, δLz and δSz in Co reach about 40×10^-3 μB and 70×10^-3 μB at Γ = 1 meV, respectively. The choice Γ = 25 meV is motivated by room temperature, yet no error bars or sensitivity tests are reported for the full set of elements. Since the abstract presents the material map as a 'key reference point,' the quantitative ranking of materials may depend on the chosen broadening. Please either provide a Γ-sensitivity analysis for representative elements from each group or explicitly state the range of Γ over which the map entries are stable.","section":"Results, 'Anisotropy of light-induced magnetism'; Figs. 4-5"}],"minor_comments":[{"comment":"There are several typographical errors, including 'transitional metals' in the paragraph above Table I and 'spin-obit interaction' in the 'K-space anatomy' section; these should be corrected.","section":"Throughout"},{"comment":"The caption of Fig. 5 states '(a-b) Light-induced orbital δL (a) and spin δS (c)' and the caption of SM Fig. 11 similarly refers to a panel (c) that is not clearly present; the panel labels and caption text should be reconciled.","section":"Fig. 5 and SM Fig. 11"},{"comment":"The statement 'A 128×128×128 interpolation k-mesh is sufficient to obtain well-converged results' is not accompanied by any convergence data; adding a short convergence test would help substantiate this important numerical claim.","section":"Methods"},{"comment":"The sentence 'The data presented in this work can be available from the corresponding author upon reasonable request' should be rephrased as 'are available upon reasonable request.'","section":"Data availability"}],"recommendation":"major_revision","confidential_remarks":"For the editor: the unresolved factor-of-2 prefactor for the orbital response sits at the center of the paper's quantitative claims and is likely fixable with a careful derivation or an independent implementation check, but as written the orbital moments are not fully auditable. The paper's scope is appropriate for the journal, and I do not see a novelty or attribution concern. The main revision should focus on making the orbital prefactor and the Γ sensitivity explicit rather than on expanding the material coverage."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a useful paper, and the conditional verdict is about right. What's new is not the formalism—the Keldysh framework and the tensor phi are from the authors' own prior work—but the sweep: a systematic first-principles map of spin and orbital inverse Faraday effect across 24 transition metals (groups IV–XI, 3d/4d/5d) at two frequencies, with k-space anatomy for Hf and Pt, band-filling plateaus, and broadening/polarization dependence. That fills a real gap. The computed values reproduce earlier same-method results for Fe, Co, Ni and are in the same ballpark as Berritta et al. where they overlap. The orbital-over-spin dominance in nonmagnets and the fcc Rh anomaly are interesting and well discussed in the context of THz orbitronics.\n\nThe soft spot is the one sentence in Methods: \"For the orbital response, the prefactor in Eq. (1) must be multiplied by an additional factor of 2.\" No derivation, no citation, no statement of what operator normalization Eq. (1) assumes for the spin channel. Every orbital moment in every table and figure inherits that factor. If the factor is wrong, all reported delta-L values are off by two; for most nonmagnetic entries the qualitative conclusion would survive, but near-unity-ratio entries like hcp Os at 1.55 eV become much less clean. The stress-test note is fair: this is the one load-bearing constant the paper does not anchor. I would not block review over it, but I would require a clear derivation or reference before accepting the numbers as a reference map.\n\nSecondary issues are more minor. The single broadening Gamma = 25 meV is a choice, and they show the Gamma dependence explicitly and warn about comparing across methods. There are no error bars or a convergence table, but they do state a 128^3 mesh is converged. Validation is against the group's own previous calculations rather than independent experiments, which is expected for a survey of this type; no data or code is shipped, but the method is established. None of that changes my overall read.\n\nBottom line: the paper is solid, useful, and mostly holds together on its own terms. It deserves a serious referee. I would not cite the absolute numbers in my own work until the factor-of-2 is nailed down, but I would be glad to see this in the literature after revision. Bring it to group if anyone cares about ultrafast orbital dynamics.","headline":"A worthwhile systematic IFE reference map whose central numbers hang on one unexplained factor-of-2 in the orbital prefactor.","tokens_in":22219,"tokens_out":3751,"would_cite":false,"duration_ms":36714,"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":"By mapping the inverse Faraday effect across 3d, 4d, and 5d transition metals from first principles, this paper establishes that the laser-induced orbital magnetic moment is generally one to two orders of magnitude larger than the spin…","keywords":["inverse Faraday effect","orbital magnetism","spin magnetism","transition metals","first-principles calculation","spin-orbit coupling","ultrafast magnetism","orbitronics"],"falsifier":"Compute the orbital IFE for a representative metal (for instance fcc Pt or hcp Co) with an independent method that does not rely on the ad hoc factor of 2—such as a time-dependent or Kubo-derived implementation—and compare magnitudes at both $\\hbar\\omega = 0.25$ eV and $1.55$ eV. A direct experimental route would be to measure the helicity-dependent magnetization of a non-magnetic $5d$ metal under a circularly polarized pump with a probe that separates orbital from spin angular momentum, e.g., X-ray magnetic circular dichroism, and check whether the reported $\\delta L/\\delta S$ ratio holds. If the factor is not exactly 2 across materials and frequencies, the map must be re-scaled.","tokens_in":21167,"feed_emoji":"🧲","tokens_out":13802,"duration_ms":101420,"temperature":0.7,"pith_summary":"The paper sets out to provide the first complete first-principles account of the inverse Faraday effect—the magnetization that a laser pulse induces in a metal—separated into its orbital and spin parts, across the $3d$, $4d$, and $5d$ transition metals of groups IV–XI. Its central claim is that in non-magnetic metals the light-induced orbital moment $\\delta L$ is one to two orders of magnitude larger than the spin moment $\\delta S$, while in ferromagnets the two are comparable. The paper traces this difference to distinct mechanisms: the orbital response is an intrinsic, non-relativistic effect tied to crystal-field splitting, whereas the spin response is generated through spin-orbit coupling. It also shows that both moments depend strongly on light frequency and polarization, and that the crystal structure and the direction of ferromagnetic magnetization create pronounced anisotropies, including helicity-dependent signs. If the calculations are right, the result is a reference map of optical magnetism that directly informs ultrafast switching, THz emission, and orbitronics.","feed_headline":"Laser-induced magnetization in non-magnetic metals is mostly orbital","feed_subtitle":"First full map of light-induced magnetism in transition metals: orbital response exceeds spin by one to two orders.","key_machinery":"The machinery is a Keldysh linear-response formula, Eq. (1) of the paper, computing the second-order response of the density matrix to the electric field of the pulse: $\\delta O = -\\frac{\\hbar a_0^3 I}{2 c E_H (\\hbar\\omega)^2} \\operatorname{Im}\\sum_{jk}\\epsilon_j \\epsilon_k^* \\varphi_{ijk}$, with $O$ standing for either the orbital angular momentum operator $L$ or the spin operator $S$. The tensor $\\varphi_{ijk}$, whose full form is given in Ref. [21], carries the material-specific band-structure information and is evaluated on a $128^3$ k-mesh from Wannier-interpolated first-principles wavefunctions. The operator distinction is the essential point: the orbital response survives without spin-orbit coupling, while the spin response does not; correspondingly, the paper states that the orbital prefactor carries an additional factor of 2. The reciprocal-space anatomy of $\\varphi_{ijk}$ explains the material trends—flat bands near the A point in hcp Hf integrate to a large orbital moment, while the band edges of d-states near X and L in fcc Pt give strong, sign-correlated hotspots for both channels.","core_discovery":"On its own terms, the discovery is that the spin and orbital channels of the inverse Faraday effect in elemental transition metals are two different phenomena. In non-relativistic calculations, only an orbital moment appears, aligned with the light propagation axis and odd under reversal of the light helicity; the spin response emerges only when spin-orbit coupling is switched on, making it a relativistic correction to the orbital effect. Across the 21 non-magnetic metals studied here, $\\delta L$ exceeds $\\delta S$ by one to two orders of magnitude, and the orbital moment varies smoothly with band filling in the $5d$ series—showing plateaus for Hf, Ta, W, Ir, and Pt—while the spin moment is erratic. In ferromagnetic Fe, Co, and Ni the two moments are of the same order, and their signs and magnitudes can reverse when the light helicity or the polarization plane is changed. The paper also reports a strong lifetime dependence: at a scattering broadening of $\\Gamma = 1$ meV the induced moments in Co grow to about $40\\times 10^{-3}\\,\\mu_B$ for $\\delta L$ and $70\\times 10^{-3}\\,\\mu_B$ for $\\delta S$, an order of magnitude larger than at the room-temperature value $\\Gamma = 25$ meV used throughout the main map. The complete set of computed values is presented as a benchmark for experimental and theoretical studies of light-induced magnetism.","pith_inferences":["If the orbital response is the primary non-relativistic one, then light-element metals and oxides with weak spin-orbit coupling should still exhibit sizable orbital IFE; this suggests a route to generating orbital moments and orbital currents in light materials without heavy elements, extending the orbital-Hall logic to optics.","The smooth band-filling plateaus in $\\delta L$ resemble the plateaus seen in orbital Hall and orbital Rashba systems, hinting that the same orbital texture governs both; a direct test would be to compute the orbital IFE spectrum of a single material and compare it with its orbital Berry curvature distribution.","The underexplained factor of 2 in the orbital prefactor could be settled by an analytic non-relativistic two-level or free-electron calculation; if it holds, it is a universal constant of the Keldysh operator matrix elements, and if not, the reported magnitudes require revision.","Extending the calculations to alloys and multilayers (e.g., CoPt, FePt) could reveal whether orbital dominance survives interface hybridization; this matters because THz emission experiments often use such heterostructures and their orbital currents are already known to be long-ranged."],"forward_implications":["In non-magnetic transition metals, the leading laser-induced magnetic response is orbital, not spin; experiments on light-induced magnetization in such metals should target the orbital channel.","The spin response is a spin-orbit-generated second-order effect, so tuning SOC (e.g., by choosing $5d$ vs $3d$ metals or by alloying) changes $\\delta S$ and $\\delta L$ in different ways.","The induced moments grow by roughly an order of magnitude as the lifetime broadening drops from 25 meV to 1 meV, implying that sample disorder directly sets the observable size of the IFE and that calculations must state their broadening.","Crystal anisotropy and magnetization direction produce helicity-dependent asymmetries—including components even in helicity—that should appear as anisotropic optical torques and helicity-dependent THz emission.","The tabulated values for all 24 metals at two frequencies constitute a direct reference for interpreting all-optical switching, THz spintronics, and orbitronics experiments."],"supporting_citations":[{"why":"Establishes the ab initio IFE method and the earlier prediction that the orbital component is sizeable; the paper's total-moment comparison table is benchmarked against it.","marker":"[20]"},{"why":"Supplies the Keldysh response tensor $\\varphi_{ijk}$ used in Eq. (1) and the laser-induced torque formalism this work extends.","marker":"[21]"},{"why":"The same formalism applied to rutile altermagnets, providing the reciprocal-space anatomy picture and a comparison point for signs and magnitudes.","marker":"[39]"},{"why":"An earlier application of the same method to Mn$_{2}$Au, used to contrast linearly polarized light effects and PT-symmetric behavior.","marker":"[29]"},{"why":"Provides the calculation parameters and Wannier-construction scheme (lattice constants, muffin-tin radii, cutoffs, 18 MLWFs) used in the present study.","marker":"[81]"},{"why":"Source of the plateaus-in-orbital-effects observation invoked to interpret the smooth band-filling dependence of $\\delta L$ in $5d$ metals.","marker":"[60]"},{"why":"Experimental demonstration of helicity-dependent THz emission linked to transverse IFE, the experimental motivation for computing transverse components.","marker":"[33]"}],"fun_headline_variants":["Light-induced orbital magnetism dominates in transition metals","Inverse Faraday effect: orbital response dwarfs spin in metals","First full map of light-induced magnetism in 3d-5d metals","Orbital magnetism wins over spin in laser-driven metals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire orbital response is multiplied by an extra factor of 2 in Eq. (1), stated in the Methods without derivation or citation; if that factor is wrong or material-dependent, every reported $\\delta L$ value—and the conclusion that orbital IFE dominates in non-magnetic metals—would be off by that factor.","fun_headline_variants_meta":{"raw":{"variants":["Light-induced orbital magnetism dominates in transition metals","Inverse Faraday effect: orbital response dwarfs spin in metals","First full map of light-induced magnetism in 3d-5d metals","Orbital magnetism wins over spin in laser-driven metals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000838,"raw_usage":{"total_tokens":3714,"prompt_tokens":1069,"completion_tokens":2645,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":685,"completion_tokens_details":{"reasoning_tokens":2576}},"tokens_in":685,"tokens_out":2645,"duration_ms":18609,"temperature":1.0,"reasoning_tokens":2576,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:51:20.786298+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the orbital IFE for a representative metal (for instance fcc Pt or hcp Co) with an independent method that does not rely on the ad hoc factor of 2—such as a time-dependent or Kubo-derived implementation—and compare magnitudes at both $\\hbar\\omega = 0.25$ eV and $1.55$ eV. A direct experimental route would be to measure the helicity-dependent magnetization of a non-magnetic $5d$ metal under a circularly polarized pump with a probe that separates orbital from spin angular momentum, e.g., X-ray magnetic circular dichroism, and check whether the reported $\\delta L/\\delta S$ ratio holds. If the factor is not exactly 2 across materials and frequencies, the map must be re-scaled.","supporting_citations":[{"cited_title":"Freimuth, S","cited_arxiv_id":null,"evidence_quote":"Supplies the Keldysh response tensor $\\varphi_{ijk}$ used in Eq. (1) and the laser-induced torque formalism this work extends."},{"cited_title":"Berritta, R","cited_arxiv_id":null,"evidence_quote":"Establishes the ab initio IFE method and the earlier prediction that the orbital component is sizeable; the paper's total-moment comparison table is benchmarked against it."},{"cited_title":"Adamantopoulos, M","cited_arxiv_id":null,"evidence_quote":"The same formalism applied to rutile altermagnets, providing the reciprocal-space anatomy picture and a comparison point for signs and magnitudes."},{"cited_title":"Merte, F","cited_arxiv_id":null,"evidence_quote":"An earlier application of the same method to Mn$_{2}$Au, used to contrast linearly polarized light effects and PT-symmetric behavior."},{"cited_title":"Go, H.-W","cited_arxiv_id":null,"evidence_quote":"Provides the calculation parameters and Wannier-construction scheme (lattice constants, muffin-tin radii, cutoffs, 18 MLWFs) used in the present study."},{"cited_title":"Adamantopoulos, M","cited_arxiv_id":null,"evidence_quote":"Source of the plateaus-in-orbital-effects observation invoked to interpret the smooth band-filling dependence of $\\delta L$ in $5d$ metals."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Experimental demonstration of helicity-dependent THz emission linked to transverse IFE, the experimental motivation for computing transverse components."}],"review_version":1}