{"id":"d369a5af-dae3-4966-a22d-373368224a22","arxiv_id":"2411.18437","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"In 2D spin-orbit coupled systems, circularly polarized light can drive pure spin photocurrents, which vanish in all k-linear models unless a Zeeman field splits the bands.","lead":"This theoretical paper calculates how circularly polarized light can generate spin currents in two-dimensional materials without producing any charge current. It classifies which spin-orbit coupled systems allow this effect and predicts a strong response peak when a magnetic field is applied.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted ℏω=2μ_z peak rests on adding only a constant Zeeman term to d_z, but the paper defines μ_z via an external perpendicular B and never includes orbital coupling; for the quoted parameters, Landau quantization is the dominant low-energy scale.","rationale":"Re-deriving the q = 1 vanishing argument, I find no internal error: for d_z = 0 and d linear in k, the radial factor (1 − k/d ∂d/∂k) vanishes identically, so the in-plane matrix elements in Eq. (B3) vanish pointwise and the out-of-plane ones are multiplied by d̂_z = 0. The symmetry-based longitudinal/transverse classification is likewise internally consistent. The remaining soft spot is the Zeeman peak. The paper explicitly ties μ_z to an external magnetic field, but the calculation is a zero-field band calculation with a constant d_z added by hand. The orbital motion in a perpendicular field changes the low-energy density of states completely; the estimated cyclotron energy is about 40 times μ_z for the parameters in Fig. 2, so the 2μ_z singularity would not survive a faithful B-field calculation. This does not refute the model if μ_z is understood as an exchange splitting, but the paper neither says this nor analyzes the broken-TRS consequences for the pure-spin-charge selection rule. Thus the reader's CONDITIONAL verdict is appropriate: the analytic two-band results stand, but the external-B realization and the purity of the spin current require a stated qualification.","tokens_in":18447,"tokens_out":15752,"duration_ms":160002,"concrete_test":"Compute the joint density of states for the k-linear Rashba-Dresselhaus model with the same parameters as Fig. 2 (α = β = 0.03 eV, μ = 0.005 eV, μ_z = 0.001 eV, m* = 0.05 m0) but with the perpendicular field included by exact Landau-level diagonalization of H = (p − eA)^2/2m + σ_x d_x + σ_y d_y + μ_z σ_z for B ≈ 17 T. If the JDOS at ℏω = 2μ_z is replaced by discrete inter-Landau-level transitions with no pronounced feature, the external-field interpretation of the prediction fails, and the paper must be recast as an exchange-field model.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sec. II introduces μ_z = (1/2)gμ_B B as the Zeeman term from an external field along z, and all results—Eqs. (23)–(24), Figs. 2–5—use only the k-space Hamiltonian H0 = ε_k + d·σ with d_z = μ_z. No vector potential or Landau-level structure is included. This is not a harmless omission for the central 2μ_z peak: the transition energy E_- − E_+ = 2√((γ k^q)^2 + μ_z^2) is independent of ε_k, so the van Hove singularity is a property of the zero-field parabolic bands. For the paper's own parameters (m* = 0.05 m0, μ_z = 0.001 eV), an external field of B ≈ 17 T (for g ≈ 2) gives ℏω_c ≈ 40 μ_z, so the Landau-level spacing far exceeds the claimed peak. With minimal coupling k → k − eA/ℏ, the interband spectrum becomes discrete and generically does not exhibit a singular 2μ_z feature. If μ_z is instead meant as an exchange splitting, that reinterpretation must be explicitly stated; in addition, μ_z ≠ 0 breaks time-reversal symmetry, so the pure-spin-current selection rule quoted in the Introduction may no longer apply unless the shift charge conductivity is also checked.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript studies the shift spin photocurrent induced by circularly polarized light in two-dimensional two-band spin-orbit coupled systems. Starting from the general second-order response formula of Ref. [21], the authors derive analytic expressions for the six components of Re[M^{Ic,xy}_{-1,1}] for Hamiltonians of the form H0 = ε_k + d(k)·σ, and apply them to k-linear Rashba, Dresselhaus, Rashba-Dresselhaus, Weyl, and persistent-spin-texture models; to k-cubic Rashba, Dresselhaus, Rashba-Dresselhaus, and wurtzite models; and to Dirac surface states with a hexagonal warping term. The central claims are: (i) without Zeeman coupling, the shift spin conductivity vanishes for all k-linear systems, while k-cubic systems and warped Dirac surface states can give nonzero response; (ii) adding a constant Zeeman term μ_z to d_z produces a peak at ℏω=2μ_z, attributed to a van Hove singularity in the joint density of states; and (iii) mirror symmetry enforces longitudinal shift spin photocurrents for Rashba-type systems and transverse ones for Dresselhaus-type systems.","tokens_in":18708,"tokens_out":9264,"duration_ms":92780,"significance":"If the main claims hold, the paper provides a useful symmetry-based classification of shift spin photocurrents and a sharp selection rule: k-linear spin-orbit systems cannot generate shift spin photocurrents under circular light unless a Zeeman term is present. The analytic expressions in Appendix B are a strength: they are derived transparently, are internally consistent for the specific Hamiltonians in Tables I and II, and the k-linear vanishing follows cleanly from the homogeneity argument in Eq. (31). The paper also correctly identifies longitudinal versus transverse responses for Rashba- and Dresselhaus-type systems and supports the classification with numerical evaluation of the conductivity. However, the headline prediction of a peak at ℏω=2μ_z is not derived for a realistic perpendicular magnetic field, because orbital coupling is omitted, and the universal 'any k-linear system' statement is broader than the derivation actually proves. These issues are load-bearing for the paper's main conclusions and require substantive revision.","major_comments":[{"comment":"The central prediction of a shift-spin-conductivity peak at ℏω=2μ_z is obtained by adding only a constant Zeeman term d_z=μ_z to the Hamiltonian, with μ_z explicitly defined as (1/2)gμ_BB from a perpendicular magnetic field. The vector potential and Landau quantization are never introduced. For the parameters used in Fig. 1 (m*=0.05m0, μ_z=0.001 eV), the required field is B≈17 T for g≈2, and the cyclotron energy is ℏω_c≈40μ_z, so Landau quantization dominates the low-energy spectrum. The van Hove singularity in the zero-field joint density of states is therefore not a reliable prediction for a real perpendicular-field experiment. The authors must either include orbital coupling and compute the interband response in Landau levels, or explicitly restate μ_z as an exchange splitting rather than a Zeeman field. In the latter case they must also verify that the shift charge conductivity remains zero, because the time-reversal-symmetry argument quoted in the Introduction no longer applies when μ_z≠0.","section":"Sec. II; Eq. (29); Figs. 2 and 4"},{"comment":"The claim that the shift spin conductivity vanishes in 'any k-linear system' without Zeeman coupling is broader than the derivation. The vanishing is proven for models with d_z=0, where the surviving term in Eq. (B3) is proportional to [1-(k/d)∂d/∂k]=1-q and hence vanishes for q=1. A generic k-linear Hamiltonian can have d_z(k) linear in k; such a term is not excluded by time-reversal symmetry, and then the terms proportional to d_z^2 in Eq. (B3) can contribute even when q=1. The universal statement in the abstract and Sec. V should be restricted to the d_z=0 case or accompanied by a proof for general k-linear d(k).","section":"Abstract; Sec. V; Sec. VI.A; Eq. (31)"}],"minor_comments":[{"comment":"There are several typos that should be corrected: 'photoccurrent' in the abstract, 'cosely' in Sec. I, 'theoreticaly' in Sec. VII, 'finte' in Sec. VI.A, and 'circurlar' in Sec. VI.B.","section":"Abstract; Sec. I; Sec. VII"},{"comment":"The basis set in Eq. (A1) is written as (|3/2,3/2⟩, |3/2,1/2⟩, |3/2,−1/2⟩, |3/2,1/2⟩), which repeats the |3/2,1/2⟩ state; the last entry should be |3/2,−3/2⟩.","section":"Appendix A"},{"comment":"The caption contains repeated '(a)(a)(a)(a)' formatting artifacts; these should be cleaned up.","section":"Fig. 2 caption"},{"comment":"The equality σyy,xy=σxx,xy for Dirac surface states is asserted but not derived; a brief symmetry argument or explicit formula would make the statement easier to verify.","section":"Sec. VI.B"},{"comment":"The caption refers to 'top' and 'bottom' panels, but the panels are not labeled; adding (a) and (b) labels would improve readability.","section":"Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's algebra appears internally consistent for the specific Hamiltonians considered, and there is no fitting-to-target circularity: the derivation starts from the published formula of Ref. [21] and evaluates it analytically. The main risk is that the headline Zeeman-induced peak is physically fragile because orbital effects are omitted while μ_z is defined from an external perpendicular field. If the authors can reframe μ_z as an exchange splitting or include Landau quantization, the paper could be made publishable; as it stands, the central prediction is not yet supported for the stated physical setup."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a serious referee. The paper does two concrete things well: it lays out a symmetry rule that separates longitudinal (Rashba-type) from transverse (Dresselhaus-type) shift spin photocurrents, and it proves that k-linear SOC alone gives zero shift spin photocurrent under circular light, with a Zeeman term producing a peak at 2μ_z from a van Hove singularity. The derivations are in the appendices and are traceable; the tables of nonvanishing components are explicit and consistent with the symmetry argument. That is a genuinely useful catalog for people designing spin photocurrent experiments.\n\nThe soft spots are real but not fatal. The most serious is the Zeeman peak. The paper defines μ_z from an external perpendicular B, but never includes the vector potential; in a real 2DEG the Landau level spacing at the quoted parameters (m* = 0.05 m0, μ_z = 1 meV) is far larger than 2μ_z, so the predicted van Hove singularity would be cut off by orbital quantization. This is not a knock-down problem if μ_z is reinterpreted as an exchange splitting or a proximity-induced field, and the algebra for the k-space model is fine. The authors should either state that reinterpretation or add the orbital coupling. Related: once TRS is broken by μ_z, the 'pure spin current' selection rule from the TRS part of the paper no longer automatically applies; the paper does not check the shift charge conductivity for the Zeeman cases. That is a minor omission but should be addressed. The Dirac surface-state section is close to earlier work by Kim, Morimoto, and Nagaosa, but the warping and the TRS-preserving framing give it enough distance; the comparison in the text is honest.\n\nI also want to note that the 'any k-linear system' vanishing claim, which the reader flagged as possibly overreaching, is stated for the case without Zeeman coupling, i.e., d_z = 0. Under that condition the homogeneity argument in Eq. (31) is general for arbitrary k-linear d-vectors, so I do not see an overreach there.\n\nBottom line: the specific model results are internally consistent, the symmetry rule is the kind of thing people will cite, and the issues are fixable in revision. I would send this to a competent referee.","headline":"Solid analytic catalog with a useful symmetry rule; the Zeeman peak needs a caveat about orbital effects, but the core results hold.","tokens_in":19283,"tokens_out":3269,"would_cite":true,"duration_ms":29896,"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":"Circularly polarized light can drive a pure spin current in two-dimensional systems with higher-order spin-orbit coupling or Zeeman splitting, while $k$-linear systems remain dark; mirror symmetry fixes the current's direction.","keywords":["shift spin photocurrent","spin-orbit coupling","Rashba-Dresselhaus","Zeeman coupling","van Hove singularity","Dirac surface states","mirror symmetry"],"falsifier":"Measure the circular-photogalvanic spin current in a clean two-dimensional electron gas with purely linear Rashba coupling and no magnetic field, isolating the relaxation-time-independent shift contribution: the central claim predicts exactly zero shift spin photocurrent at every photon frequency, so any nonzero shift signal would falsify the vanishing theorem.","tokens_in":18195,"feed_emoji":"🧲","tokens_out":21158,"duration_ms":149342,"temperature":0.7,"pith_summary":"The paper sets out to show that in two-dimensional electron systems, circularly polarized light can produce a pure spin photocurrent (a flow of spin without a net charge current) through the shift-current mechanism, even when time-reversal symmetry is intact. The authors find an exact vanishing rule: in any spin-orbit-coupled system whose spin-orbit field is linear in momentum, the shift spin photocurrent is zero unless a Zeeman coupling is present. The effect therefore requires either higher-order ($k$-cubic) spin-orbit terms, as in wurtzite or $k$-cubic Rashba-Dresselhaus systems, or a Zeeman-split band structure. Mirror symmetry then dictates the geometry: Rashba-type systems give longitudinal spin currents (spin parallel to flow), Dresselhaus-type systems give transverse ones (spin perpendicular to flow). With Zeeman splitting, the joint density of states develops a van Hove singularity that produces a sharp peak in the shift spin conductivity at photon energy $2\\mu_z$, offering a tunable resonance.","feed_headline":"Circular light can generate pure spin currents in 2D crystals","feed_subtitle":"Theory predicts pure spin currents from circular light, with a tunable resonance at twice the Zeeman energy.","key_machinery":"The central object is the shift spin conductivity tensor $\\sigma^{Ic;ab}$ (Eq. (8)) and its matrix element $\\mathrm{Re}[M^{Ic;xy}_{-1,1}]$ for a two-band Hamiltonian $H_0 = \\epsilon_k + \\mathbf{d}(k)\\cdot\\boldsymbol{\\sigma}$. The calculation reduces to a few algebraic factors: the band-geometric factor $(1 - \\frac{k}{d}\\frac{\\partial d}{\\partial k})$, which is exactly zero for $k$-linear dispersion ($q=1$) and nonzero for $q\\neq 1$; the Zeeman energy $\\mu_z$ entering the dispersion $d = \\sqrt{(k^q \\gamma)^2 + \\mu_z^2}$; and the momentum-space symmetry operations (mirror $M_x$, $M_y$ and parity-mirror $PM_x$, $PM_y$) that constrain the allowed conductivity components. The paper evaluates these analytically and then confirms by numerical integration, identifying the longitudinal ($xx$, $yy$) and transverse ($xy$, $yx$) components and the out-of-plane ($zx$, $zy$) components.","core_discovery":"The paper's central claim is that the shift spin photocurrent induced by circularly polarized light in a two-band system is governed by a geometric matrix element $\\mathrm{Re}[M^{Ic;xy}_{-1,1}]$ whose value is fixed by the momentum dependence of the spin-orbit field $\\mathbf{d}(k)$. For any $k$-linear spin-orbit Hamiltonian with no Zeeman coupling, the factor $(1 - \\frac{k}{d}\\frac{\\partial d}{\\partial k})$ vanishes identically, making all shift spin conductivities zero; this is stated in Sec. V as 'for $k$-linear system $q=1$, we always have $\\mathrm{Re}[M]=0$'. The vanishing is lifted either by higher-order ($k$-cubic) terms or by a Zeeman term $\\mu_z$, which changes the dispersion to $d = \\sqrt{(k^q \\gamma)^2 + \\mu_z^2}$ and makes the factor nonzero. The paper further shows that mirror (Rashba-type) or parity-mirror (Dresselhaus-type) symmetries select which tensor components survive, producing longitudinal or transverse spin currents respectively, and that Dirac surface states with hexagonal warping support longitudinal shift spin photocurrent. Finally, when Zeeman splitting is nonzero, the joint density of states has a van Hove singularity at the band bottom, giving a sharp peak in the shift spin conductivity at $\\hbar\\omega = 2\\mu_z$.","pith_inferences":["An immediate extension would be to include orbital (Landau) effects beyond the Zeeman term; the $2\\mu_z$ peak would likely split into Landau-level resonances, changing the predicted line shape in real magnetic fields.","The same mirror-symmetry selection rules should apply to other second-order spin responses, such as injection spin photocurrents, so the longitudinal/transverse classification could be tested in complementary measurements.","The exact zero for $k$-linear two-band models appears to be a purely algebraic consequence of the factor $(1 - \\frac{k}{d}\\frac{\\partial d}{\\partial k}) = 0$, suggesting a general geometric selection rule that may protect other response tensors in linear-in-$k$ systems."],"forward_implications":["In a two-dimensional electron gas with purely linear Rashba or Dresselhaus spin-orbit coupling, circularly polarized light will produce no shift spin photocurrent unless a magnetic field is applied; experimental searches should focus on higher-order spin-orbit systems or field-tuned samples.","Rashba-type systems will always convert circular light into a spin current whose spin polarization is parallel to the current direction (longitudinal), while Dresselhaus-type systems produce a transverse geometry; this gives a symmetry-based design rule for generating spin currents with a chosen polarization direction.","Applying a perpendicular Zeeman field $\\mu_z$ introduces a sharp, tunable peak in the shift spin conductivity at photon energy $\\hbar\\omega = 2\\mu_z$, so sweeping the magnetic field tunes the optical frequency at which pure spin current is resonantly generated.","Dirac surface states with hexagonal warping support longitudinal shift spin photocurrent with no need to break time-reversal symmetry, and the response strengthens with warping, making topological insulator surfaces a candidate platform for all-optical spin injection."],"supporting_citations":[{"why":"Supplies the general second-order spin photocurrent theory and the shift spin conductivity formula (Eq. (8)) that the paper applies to two-band models.","marker":"[21]"},{"why":"Provides the symmetry transformation rules for spin and velocity matrix elements under mirror operations used in Sec. IV, plus the concept of pure spin photocurrent.","marker":"[5]"},{"why":"Used for the symmetry constraints on shift photocurrents (with [5, 21]) and as a basis for the Dirac surface state Hamiltonian (with [48]).","marker":"[7]"},{"why":"Prior calculation of shift charge and spin photocurrents in Dirac surface states; the paper extends it by including the hexagonal warping term and preserving time-reversal symmetry.","marker":"[19]"},{"why":"Provides the hexagonal warping term in the Dirac surface state Hamiltonian (Eq. (41)) that is essential for nonzero shift spin photocurrent in topological insulator surfaces.","marker":"[48]"}],"fun_headline_variants":["Circular light drives pure spin currents in 2D materials","Light-induced spin current without charge flow in 2D","Tunable spin photocurrent from circular light in 2D","Shift spin current: light makes spin flow, not charge","2D spin current from light: zero charge, tunable peak"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's results assume that an external magnetic field contributes only the Zeeman energy $\\mu_z$ to the spin-orbit field, with no orbital (Landau-quantization) effects; if orbital effects are significant, the predicted resonance at twice the Zeeman energy would not appear as calculated.","fun_headline_variants_meta":{"raw":{"variants":["Circular light drives pure spin currents in 2D materials","Light-induced spin current without charge flow in 2D","Tunable spin photocurrent from circular light in 2D","Shift spin current: light makes spin flow, not charge","2D spin current from light: zero charge, tunable peak"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000538,"raw_usage":{"total_tokens":2672,"prompt_tokens":1124,"completion_tokens":1548,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":740,"completion_tokens_details":{"reasoning_tokens":1471}},"tokens_in":740,"tokens_out":1548,"duration_ms":10859,"temperature":1.0,"reasoning_tokens":1471,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:13:54.905830+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the circular-photogalvanic spin current in a clean two-dimensional electron gas with purely linear Rashba coupling and no magnetic field, isolating the relaxation-time-independent shift contribution: the central claim predicts exactly zero shift spin photocurrent at every photon frequency, so any nonzero shift signal would falsify the vanishing theorem.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Used for the symmetry constraints on shift photocurrents (with [5, 21]) and as a basis for the Dirac surface state Hamiltonian (with [48])."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Prior calculation of shift charge and spin photocurrents in Dirac surface states; the paper extends it by including the hexagonal warping term and preserving time-reversal symmetry."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the hexagonal warping term in the Dirac surface state Hamiltonian (Eq. (41)) that is essential for nonzero shift spin photocurrent in topological insulator surfaces."}],"review_version":1}