{"id":"b09f2218-67da-46eb-8079-e1524cd5abe0","arxiv_id":"2412.13927","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Coulomb drag in altermagnetic bilayers is predicted to produce Hall drag and spin Hall drag without spin-orbit coupling, with orientation-dependent signatures of altermagnetism.","lead":"This paper proposes using Coulomb drag, the mutual friction between two close but electrically separate layers, to detect altermagnetism, a newly identified type of magnetism with anisotropic electronic bands. The authors predict that altermagnets produce a transverse, Hall-like dragged voltage that no other system without spin-orbit coupling shows, giving a clear signature of the phase.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The q-independent diffusive polarization in Eq. (B8) is used outside its validity regime, and the q integral in Eq. (C6) samples q≈1/d where q l ≫ 1, so the Table I angle signatures are not yet established.","rationale":"The reader's weakest-assumption statement identifies the same load-bearing concern: the diffusive, q-independent polarization is used in a regime where q l is large. I checked the internal consistency directly. Eq. (B7) is a small-ω, small-q expansion, but Eq. (C6) integrates q with weight q²/sinh²(qd), which is peaked at qd ≈ 1.5. Thus the q→0 approximation is applied at the very momentum scale that dominates the integral. The paper's own phrase 'q ≪ 1/d in the Boltzmann regime' after Eq. (2) conflicts with this integration. The numbers make the failure quantitative: with τ = 10⁻¹⁰ s and v_F ≈ 6 × 10⁵ m/s the mean free path is tens of microns, while the relevant q is about 0.1 nm⁻¹, so q l ≈ 10³–10⁴. This is not a small correction; the imaginary part of the polarization in the collisionless regime is q-dependent and approximately ν ω/(v_F q), not ν ωτ/ħ. Since Table I is obtained by pulling the q-independent Π out of the angular integral, the reported angle dependences could be modified by the true q-dependent response. I do not claim the qualitative phenomenon is impossible: a symmetry-based Hall drag in an anisotropic altermagnetic bilayer may well survive a full calculation. But the central predictive content of the paper, including the specific cos(2α) signatures and magnitudes in Fig. 4, is conditional on a check with the full Lindhard response. That is why the reader's CONDITIONAL verdict is appropriate and should remain unchanged until the check is performed.","tokens_in":15116,"tokens_out":14973,"duration_ms":155305,"concrete_test":"Recompute the drag resistivity in Eq. (C4) using the full spin-resolved Lindhard polarization Π_s(q,ω) for the Hamiltonian in Eq. (1), at minimum in the collisionless limit Im Π_s ≈ ν_s ω/(v_F^s(φ) q), and evaluate the q integral numerically for the Fig. 4 parameters (d = 12 nm, T = 2 K, ε_r = 5, τ = 10⁻¹⁰ s, μ = 0.3 eV, t = 0.127 eV nm²). Then compare ρ_HD^↑ and V^s_2y with Table I and Eqs. (7)–(8). If the transverse entries keep the same J-dependence and cos(2α) periodicity, the signature is robust; if they change sign, magnitude, or angular form, the Table I expressions are artifacts of the diffusive approximation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the replacement of the retarded polarization by the q-independent diffusive form Π_R(q,ω) = (1 + iωτ/ħ)/(2π√(4t² − J²)) in Eq. (B8), used in Eq. (C3) to obtain the nonlinear susceptibility Γℓ and hence all entries of Table I. The derivation in Appendix B is a small-ω, small-q expansion: Eq. (B7) is explicitly of order ωτ, and the text takes the q→0 limit. However, the momentum integral in Eq. (C6) is not restricted to q ≪ 1/d: the factor q²/sinh²(qd) is dominated by qd ≈ 1.5, i.e. q ≈ 1/d. With the plotted parameters (d = 12 nm, τ = 10⁻¹⁰ s, v_F ≈ 6 × 10⁵ m/s), this gives q l ≈ q v_F τ ≈ 10³–10⁴, so the calculation is applied deep in the collisionless regime. There the imaginary part of the polarization is of order ν ω/(v_F q), not ν ωτ/ħ, and it carries q and angular dependence coming from the anisotropic Fermi surfaces. Because Eq. (B8) is factored out of the angular integral, the specific forms ρ_HD^↑ = 4tJ F_T cos(2α₁) and Eq. (8) depend on this uncontrolled approximation; a q- and angle-dependent Π_s(q,ω) can mix angular harmonics and change, weaken, or even remove the reported signatures. This does not disprove the qualitative Hall-drag idea, but it means the central quantitative claims of Table I, and the proposed experimental signatures based on them, are not yet established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes Coulomb drag as an experimental probe of altermagnetism. The authors study a bilayer of two-dimensional d-wave altermagnets, modeled by a minimal quadratic Hamiltonian with a momentum-dependent spin splitting (Eq. 1), and compute the drag resistivities using the Aslamazov–Larkin diagrammatic formalism (Eq. 2, Fig. 3). They derive analytic expressions for longitudinal, Hall, and spin Hall drag resistivities (Table I) and show that the transverse components are proportional to the altermagnetic exchange coupling J and vanish for J=0, implying Hall drag without spin-orbit coupling. They further propose a multiterminal setup (Fig. 4) and identify angle dependences in α1 and α2 as signatures of altermagnetic spin-split Fermi surfaces.","tokens_in":15518,"tokens_out":2961,"duration_ms":29538,"significance":"If the results hold, the paper offers a non-contact transport signature for altermagnetic band splitting, which is timely given the ongoing controversy over candidate materials such as RuO2. The analytical treatment is complete and the proposed measurement scheme is concrete and falsifiable. The central advance—transverse drag generated by the intrinsic anisotropy of altermagnetic Fermi surfaces rather than by spin-orbit coupling—is an interesting and plausible idea that would be a valuable contribution to the altermagnetism literature. However, the quantitative predictions in Table I and Eqs. (6)–(8) rest on an uncontrolled approximation in the momentum integral, as detailed below, so the specific angle dependences and magnitudes are not yet established.","major_comments":[{"comment":"The q-independent, diffusive polarization Π_R(q,ω) = (1 + iωτ/ħ)/(2π√(4t²−J²)) is derived in Appendix B under the long-wavelength (q→0) limit and with an expansion in small ωτ (Eq. B7). Yet the final drag integral in Eq. (C6) integrates q from 0 to infinity without any restriction to q ≪ 1/d. The factor q²/sinh²(qd) is maximized near qd ≈ 1.5, so the integral is dominated by q ≈ 1/d. For the plotted parameters (d = 12 nm, τ = 10⁻¹⁰ s, v_F ≈ 6×10⁵ m/s), this implies q l ≈ q v_F τ ≈ 10³–10⁴, far outside the diffusive regime q l ≪ 1. In this collisionless regime the imaginary part of the polarization is of order νω/(v_F q), not νωτ/ħ, and it acquires q and angular dependence from the anisotropic Fermi surface. Because Eq. (C3) factors out the q-independent Im Π_R, the angular integrals that produce Table I and Eqs. (6)–(8) rely on an uncontrolled approximation. The reported angle signatures and magnitudes may be modified, weakened, or even reversed when a proper q- and angle-dependent polarization is used. The authors should either restrict the calculation to experimentally achievable parameters satisfying both q l ≪ 1 and q ≪ 1/d, or redo the derivation with the full polarization and show that the qualitative conclusions survive.","section":"Appendix B, Eq. (B8); Eq. (C6)"},{"comment":"The manuscript states that the screened interaction U12(q) and the drag formula apply 'under the random phase approximation and q ≪ 1/d in the Boltzmann regime.' This condition is contradicted by the actual evaluation in Eq. (C6), which integrates over all q without any cutoff. The text should explicitly acknowledge that the q-integral samples the region qd ~ 1 and therefore the stated regime is violated. Either impose a momentum cutoff consistent with the diffusive approximation or justify the use of the full q range; otherwise the self-consistency of the derivation is not established.","section":"Main text, after Eq. (2); Appendix C"},{"comment":"The nonlinear susceptibility Γ_ℓ is written as proportional to v_q^(ℓ) Im[Π_R(q,ω)]. For anisotropic Fermi surfaces, v_q^(ℓ) depends on the transferred momentum direction but not on the angle of k, which is why the angular integration over φ in Eq. (C6) can be done analytically. However, the correct nonlinear susceptibility involves an integral over the Fermi surface that also contains the energy denominator and the distribution functions; the factorization in Eq. (C3) is only valid if the polarization is q-independent in the relevant q range. Since that condition fails, the factorization itself is a load-bearing step. The authors should verify the factorization against a direct calculation of Γ_ℓ with the full Lindhard-type response for the altermagnetic model.","section":"Eq. (C3) and Table I"}],"minor_comments":[{"comment":"The arXiv number in reference [26] appears as 'arXiv:408.00320', which is likely a typo for 'arXiv:2408.00320'.","section":"Reference [26]"},{"comment":"The phrase 'strongly dependent of the orientation' should be 'strongly dependent on the orientation'.","section":"Introduction, second paragraph"},{"comment":"The labels ρCD, ρ↑CD, ρ↓CD in panels (e) and (f) are not explicitly defined in the caption; adding the spin-resolved definitions would improve readability.","section":"Fig. 4 caption"},{"comment":"The notation ω± = ω ± i0⁺ is introduced in the main text, but in Appendix B the analytic continuation is written as iω_m → ω + i0⁺; please ensure the sign convention is consistent throughout.","section":"Eq. (2) and Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well-written and the central idea—Hall drag without spin-orbit coupling arising from altermagnetic Fermi-surface anisotropy—is attractive and should be of interest to the condensed-matter community. The main concern is not the novelty or the qualitative physics but the uncontrolled approximation in the momentum integration that underlies all quantitative predictions. I believe this is fixable by either restricting to genuine diffusive parameters or performing a more complete calculation, so I recommend major revision rather than rejection. I also note that the model and one coauthored reference [43] are used, but the drag calculation itself is new and not circular."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper applies Coulomb drag to altermagnetic bilayers and predicts transverse (Hall) drag and spin Hall drag in the absence of spin-orbit coupling. That is a genuinely new proposal, and the symmetry argument is plausible: the anisotropic spin-split Fermi surfaces give the transferred momentum a transverse component, so a current in the active layer can induce a transverse voltage in the passive layer. The concrete multiterminal measurement setup and the tabulated angle dependences are useful, and the authors are careful to note that real materials with weak SOC should not qualitatively alter their main conclusions.\n\nThe soft spot is the polarization. The authors use the q-independent diffusive form Π_R(q,ω) = (1 + iωτ/ħ)/(2π√(4t²−J²)) derived in Appendix B under q→0 and ωτ≪1. The derivation of the drag resistivity then integrates momentum to infinity, and the factor q²/sinh²(qd) peaks at qd ≈ 1.5. For the plotted parameters d=12 nm, τ=10⁻¹⁰ s, v_F≈6×10⁵ m/s, the relevant q gives q l ≈ 10³–10⁴, deep in the collisionless regime where the diffusive form is not valid. There the imaginary part of the polarization is of order νω/(v_F q) and, more importantly, carries q and angular dependence from the anisotropic Fermi surface. Because the q-independent Π is factored out of the angular integrals, the specific forms in Table I—e.g., ρ_HD^↑ = 4tJ F_T cos(2α₁)—depend on that uncontrolled approximation. A q- and angle-dependent polarization can mix angular harmonics and could change, weaken, or even remove some of the predicted signatures.\n\nThis does not disprove the Hall-drag idea. On symmetry grounds, a transverse drag is allowed in an altermagnetic bilayer, and the leading J-linear term likely has a cos(2α) form. But the quantitative predictions and the plots in Fig. 4 are not yet established. The manuscript needs a full Lindhard-function calculation (or at least a justification for the diffusive regime with parameters that satisfy q l ≪ 1), which would also settle whether the presented parameters are compatible with the Boltzmann assumption.\n\nWho is this for? Researchers working on altermagnetism and transport probes; the proposal is valuable even if the current numbers are not final. It deserves a serious referee, but the referee should require the collisionless calculation before publication. I would not trust the magnitudes in Table I until that is done.","headline":"A genuinely new proposal for probing altermagnetic spin splitting with Coulomb drag, but the quantitative drag calculation rests on a diffusive polarization that is unjustified in the plotted parameter regime.","tokens_in":16061,"tokens_out":4500,"would_cite":false,"duration_ms":39602,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A bilayer of d-wave altermagnets should show Hall drag and spin Hall drag without any spin-orbit coupling, with orientation-angle dependences that serve as fingerprints of the spin-split Fermi surfaces.","keywords":["Coulomb drag","altermagnetism","Hall drag","spin Hall drag","spin-split Fermi surface","angle dependence","RuO2 bilayer","d-wave altermagnet"],"falsifier":"Measure the transverse drag voltage in a RuO2-based bilayer with $d\\approx 12$ nm at $T\\approx 2$ K while rotating the active-layer crystal orientation $\\alpha_1$ relative to a spin-polarized drive current; the predicted Hall drag is proportional to $\\cos(2\\alpha_1)$ and changes sign at $\\alpha_1=\\pi/4$. A null result at all angles, or a modulation with the wrong period, would falsify the central claim.","tokens_in":14867,"feed_emoji":"🧲","tokens_out":11871,"duration_ms":101635,"temperature":0.7,"pith_summary":"Coulomb drag is a non-contact probe: a current in one layer induces a voltage in a nearby, electrically isolated second layer through interlayer Coulomb interactions. This paper argues that in a bilayer of d-wave altermagnets, the anisotropic spin-split Fermi surfaces convert the transferred momentum into both longitudinal and transverse responses, so a driving current produces Hall drag and spin Hall drag even with no spin-orbit coupling. The predicted drag voltages oscillate with the orientation of the Fermi-surface splitting, with period that reflects the altermagnet's symmetry, and vanish when the altermagnetic exchange is zero. A multiterminal measurement of these angle dependences would give a transport fingerprint of the altermagnetic state.","feed_headline":"Altermagnets drag currents sideways without spin-orbit coupling","feed_subtitle":"Transverse and spin voltages dragged across a bilayer carry angle-dependent fingerprints of altermagnetic Fermi surfaces.","key_machinery":"The argument is carried by the d-wave altermagnet model used for each layer, $H(k)=tk^2+J[\\cos(2\\alpha)k_xk_y+\\sin(2\\alpha)(k_x^2-k_y^2)/2]\\sigma_z$, where $J$ is the altermagnetic exchange and $\\alpha$ sets the orientation of the spin splitting. The velocity shift $v^{(\\ell)}_{q}=v^{(\\ell)}_{k+q}-v^{(\\ell)}_{k}$ in this model has longitudinal and transverse parts whose magnitudes depend on $\\alpha_\\ell$ and on spin. Inserting this velocity shift into the drag resistivity formula $\\rho^{ij}_D\\propto\\int d^2q\\,d\\omega\\,|U_{12}(q)|^2\\Gamma^i_1\\Gamma^j_2/\\sinh^2(\\beta\\omega/2)$, with nonlinear susceptibilities $\\Gamma_\\ell=-2\\tau\\,v^{(\\ell)}_q\\operatorname{Im}\\Pi^R_\\ell/\\hbar$ and polarization $\\Pi^R_\\ell=(1+i\\omega\\tau/\\hbar)/(2\\pi\\sqrt{4t^2-J^2})$, makes the transverse and spin components arise from the $J$-dependent part of $v_q$. The common prefactor $F_T\\propto (k_BT)^2\\varepsilon^2\\tau^2(4t^2-J^2)^2/(\\hbar e^6 d^6\\mu_1\\mu_2 t^2)$ collects the temperature, screening, interlayer distance, and chemical-potential dependence.","core_discovery":"The central claim is that in a bilayer of two d-wave altermagnets separated by a dielectric, the Coulomb drag resistivity acquires transverse and spin-dependent components controlled by the altermagnetic exchange strength $J$ and the orientation angles $\\alpha_1,\\alpha_2$ of the two layers' spin-split Fermi surfaces. Starting from the minimal model $H(k)=tk^2+J[\\cos(2\\alpha)k_xk_y+\\sin(2\\alpha)(k_x^2-k_y^2)/2]\\sigma_z$, the authors derive the analytic drag resistivities of Table I and show that the Hall drag of a fully spin-polarized drive is $\\rho^{\\uparrow}_{\\mathrm{HD}}=4tJF_T\\cos(2\\alpha_1)$, so it is nonzero only when $J\\neq 0$ and requires no spin-orbit coupling. The spin Hall drag voltage $V^s_{2y}\\propto J F_T[\\cos(2\\alpha_1)+\\eta J\\sin(2\\alpha_1+2\\alpha_2)]$, with $\\eta$ the spin polarization of the drive, survives even for an unpolarized driving current. All these signals are periodic in the orientation angles, which is what makes them usable as transport signatures of altermagnetism.","pith_inferences":["The paper does not emphasize that forming the ratio of Hall drag to longitudinal drag would cancel the common prefactor $F_T$; such a ratio would isolate $J/t$ and could make the signature robust to uncertainties in density, dielectric constant, and scattering time.","A natural follow-up device would rotate the passive layer's crystal orientation in situ and record the transverse voltage; the predicted sign change under a $90^\\circ$ rotation of $\\alpha_1$ could be verified in one multiterminal sample.","The authors' symmetry argument also implies that a normal-metal active layer with spin-polarized current should drag a transverse signal into an altermagnet passive layer, which would extend the probe to heterostructures without requiring altermagnet growth in both layers.","Measuring the drag signal at several temperatures would discriminate the predicted $T^2$ diffusive drag from phonon- or magnon-mediated drag mechanisms that scale differently with temperature."],"forward_implications":["A spin-polarized driving current produces a transverse (Hall) drag voltage proportional to $\\eta J F_T\\cos(2\\alpha_1)$, so the Hall drag is a direct test for nonzero altermagnetic exchange $J$ in the absence of spin-orbit coupling.","The longitudinal drag voltage $V_{2x}\\propto F_T[2t^2+\\eta\\, tJ\\sin(2\\alpha_1)]$ is $\\pi$-periodic in the crystal orientation $\\alpha_1$, which gives an orientation signature in a simple measurement.","An unpolarized driving current still induces a spin Hall drag voltage proportional to $J F_T\\cos(2\\alpha_1)$, so the effect does not need ferromagnetic injection contacts.","Since the angle period is $\\pi$ for d-wave, $\\pi/2$ for g-wave, and $\\pi/3$ for i-wave altermagnets, Coulomb drag could distinguish the symmetry class of the altermagnetic order, as the paper notes.","The dependence on both $\\alpha_1$ and $\\alpha_2$ through $\\alpha_1\\pm\\alpha_2$ means the relative crystallographic alignment of the two layers can be read from the dragging signal."],"supporting_citations":[{"why":"Establishes altermagnetism as a collinear magnetic phase with anisotropic spin-split bands without spin-orbit coupling, the object the drag measurement targets.","marker":"[3, 4]"},{"why":"Supplies the minimal d-wave altermagnet Hamiltonian $H(k)$ used for each layer.","marker":"[43]"},{"why":"Supplies the Coulomb drag formalism and screened interlayer interaction underlying the drag-resistivity calculation.","marker":"[52]"},{"why":"Provides the diagrammatic linear-response calculation of Coulomb drag from which the polarization and resistivity formulas are taken.","marker":"[74]"},{"why":"Gives the nonlinear susceptibility and velocity-shift expressions used in deriving Eq. (3) and Table I.","marker":"[82]"},{"why":"Baseline for drag in the presence of spin-orbit coupling, whose lack of angle dependence is contrasted with the altermagnet result.","marker":"[79]"},{"why":"Earlier spin Hall drag proposal in bilayers, showing the conventional requirement of spin-orbit coupling that the altermagnet mechanism bypasses.","marker":"[59]"},{"why":"Contributes the frequency-integral identity and the treatment of anisotropic Fermi surfaces used in the analytic derivation.","marker":"[78]"}],"fun_headline_variants":["Coulomb drag reveals altermagnet Fermi surface angles","Transverse drag currents fingerprint altermagnetism","Hall drag without spin-orbit coupling in altermagnets","Angle-dependent Coulomb drag signals altermagnetism","Altermagnet drag: sideways currents without spin-orbit"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes the electron layers respond in the diffusive regime, meaning the momentum transferred between layers is small compared with the inverse electron mean free path; outside that regime the simplified polarization used for the drag resistivities would need to be replaced by the full momentum-dependent response.","fun_headline_variants_meta":{"raw":{"variants":["Coulomb drag reveals altermagnet Fermi surface angles","Transverse drag currents fingerprint altermagnetism","Hall drag without spin-orbit coupling in altermagnets","Angle-dependent Coulomb drag signals altermagnetism","Altermagnet drag: sideways currents without spin-orbit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000188,"raw_usage":{"total_tokens":1349,"prompt_tokens":982,"completion_tokens":367,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":598,"completion_tokens_details":{"reasoning_tokens":292}},"tokens_in":598,"tokens_out":367,"duration_ms":3877,"temperature":1.0,"reasoning_tokens":292,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:41:01.074372+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the transverse drag voltage in a RuO2-based bilayer with $d\\approx 12$ nm at $T\\approx 2$ K while rotating the active-layer crystal orientation $\\alpha_1$ relative to a spin-polarized drive current; the predicted Hall drag is proportional to $\\cos(2\\alpha_1)$ and changes sign at $\\alpha_1=\\pi/4$. A null result at all angles, or a modulation with the wrong period, would falsify the central claim.","supporting_citations":[{"cited_title":"Finite- momentum Cooper pairing in proximitized altermag- nets","cited_arxiv_id":null,"evidence_quote":"Supplies the minimal d-wave altermagnet Hamiltonian $H(k)$ used for each layer."},{"cited_title":"Coulomb drag","cited_arxiv_id":null,"evidence_quote":"Supplies the Coulomb drag formalism and screened interlayer interaction underlying the drag-resistivity calculation."},{"cited_title":"Coulomb drag in normal metals and superconductors: Diagrammatic approach","cited_arxiv_id":null,"evidence_quote":"Provides the diagrammatic linear-response calculation of Coulomb drag from which the polarization and resistivity formulas are taken."},{"cited_title":"Linear-response theory of Coulomb drag in coupled electron systems","cited_arxiv_id":null,"evidence_quote":"Gives the nonlinear susceptibility and velocity-shift expressions used in deriving Eq. (3) and Table I."},{"cited_title":"Coulomb drag and spin drag in the presence of spin-orbit coupling","cited_arxiv_id":null,"evidence_quote":"Baseline for drag in the presence of spin-orbit coupling, whose lack of angle dependence is contrasted with the altermagnet result."},{"cited_title":"Spin Hall Drag in Elec- tronic Bilayers","cited_arxiv_id":null,"evidence_quote":"Earlier spin Hall drag proposal in bilayers, showing the conventional requirement of spin-orbit coupling that the altermagnet mechanism bypasses."},{"cited_title":"Coulomb drag as a measure of trigonal warping in doped graphene","cited_arxiv_id":null,"evidence_quote":"Contributes the frequency-integral identity and the treatment of anisotropic Fermi surfaces used in the analytic derivation."}],"review_version":1}