{"id":"7abe1f0c-2c7b-41e3-8734-b840a927b6ec","arxiv_id":"1908.03906","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"An extended drift-diffusion model separates spin Hall and anisotropic magnetoresistance contributions in W/CoFeB, W/Co, CoFeB/Pt, and Co/Pt metallic bilayers.","lead":"This paper measures how the electrical resistance of thin metal sandwiches changes when the magnetic layer's magnetization rotates, and separates two effects: spin Hall magnetoresistance and anisotropic magnetoresistance. The separation matters for designing magnetic field sensors and for measuring spin transport properties in spintronic devices.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Extracted θSH and θAMR disagree by 2–3x between tH and tF series of same bilayers; without uncertainty estimates the claimed 'precise' SMR/AMR decomposition is unsupported","rationale":"The reader's weakest assumption matches the most load-bearing concern: the separation of SMR and AMR depends on θSH and θAMR being independent of layer thickness, and the paper explicitly limits itself by stating this assumption 'may result in overestimated parameters for very thin ferromagnetic layers.' The reader's CONDITIONAL verdict already demands uncertainty quantification and thickness-dependence tests, which are exactly the checks that would settle my concern. Table II provides internal evidence that the assumption is strained, since complementary thickness series for nominally the same systems yield parameters that differ by factors of 2–3. Whether these differences are real or within noise cannot be judged because no uncertainties are given; that absence is itself incompatible with the word 'precise' in the central claim. I do not find a more fundamental error in the drift-diffusion derivation, but the empirical validation of the SMR/AMR decomposition is missing. Therefore my read does not change the reader's verdict: the paper should remain CONDITIONAL pending the global-fit consistency check, uncertainty reporting, and/or independent AMR measurement.","tokens_in":12234,"tokens_out":7025,"duration_ms":71287,"concrete_test":"Re-analyze the original resistance-vs-angle data for each bilayer system in a single global least-squares fit in which θSH and θAMR are shared between the tH and tF series, with measurement uncertainties propagated via covariance or bootstrap. If the global parameter estimates deviate from Table II by more than 1σ, or the reduced χ² substantially exceeds 1, the thickness-independence assumption is falsified. As an independent cross-check, deposit single F films with a non-SHE capping layer over the same tF range and measure θAMR(tF) by conventional AMR; if it varies by more than ~10% across the measured range, the Section IV decomposition is unreliable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the 'precise determination' of SMR and AMR contributions (Abstract; Sec. IV). This relies on the model correctly separating the two effects, which in turn relies on the stated assumption (Sec. IV) that θAMR and θSH are independent of layer thickness. The evidence does not yet support that assumption. Table II lists separate fits to the tF-varying and tH-varying series of the same nominal bilayers. For W/Co, |θSH| = 0.34 in W3 vs 0.47 in W4 and θAMR = 1.1% vs 0.5%; for Co/Pt, |θSH| = 0.09 in P3 vs 0.28 in P4 and θAMR = 2.4% vs 0.7%; fitted Gr also differs by an order of magnitude between complementary series. If the physical parameters were truly thickness independent, the two series should yield consistent values. No error bars, confidence intervals, or goodness-of-fit are reported, so the discrepancies could be noise—but that is exactly the problem: without these, 'precise determination' is not established. The SMR/AMR decomposition may be an artifact of the assumed thickness dependence.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper revisits the theory of spin Hall magnetoresistance (SMR) in heavy-metal/ferromagnetic-metal bilayers by explicitly including anisotropic magnetoresistance (AMR) and the anomalous Hall effect (AHE) in the spin drift-diffusion equations for the ferromagnet. The authors derive closed-form expressions for the magnetoresistance, define SMR by setting the AMR and AHE angles to zero in Eq. (22), and fit the model to resistance measurements on W/Co20Fe60B20, W/Co, Co20Fe60B20/Pt, and Co/Pt bilayers with varied heavy-metal or ferromagnetic thickness. From these fits they extract spin Hall angles, AMR angles, spin-mixing conductances, and spin diffusion lengths, and conclude that the approach allows precise determination of the SMR and AMR contributions to the magnetoresistance.","tokens_in":12536,"tokens_out":2320,"duration_ms":25333,"significance":"If valid, the model would provide a useful framework for separating SMR from AMR in metallic ferromagnet/heavy-metal bilayers, with relevance for magnetoresistance-based spin-transport metrology and for optimizing magnetic stray-field sensors. The derivation in Section II is a genuine extension of earlier SMR theory and the paper compares its model with the simplified Kim et al. model, which is a strength. However, the central quantitative claim of 'precise determination' is currently not supported because the fitted parameters are not accompanied by uncertainties and, more seriously, fits to complementary thickness series of the same nominal bilayers yield mutually inconsistent values. The paper would need a substantially more rigorous fitting and validation procedure before the claimed precision can be accepted.","major_comments":[{"comment":"The fitted parameters are reported without uncertainties, confidence intervals, or goodness-of-fit measures, yet the abstract and Section IV claim 'precise determination' of the SMR and AMR contributions. This is especially problematic because the two complementary series for the same nominal bilayers give inconsistent values: for W/Co, |θSH| = 0.34 in W3 versus 0.47 in W4 and θAMR = 1.1% versus 0.5%; for Co/Pt, |θSH| = 0.09 in P3 versus 0.28 in P4 and θAMR = 2.4% versus 0.7%; the fitted Gr also differs by an order of magnitude between the two series. Without uncertainty estimates or a demonstration that these differences are within expected scatter, the claim of precise determination is not established.","section":"Sec. IV, Table II"},{"comment":"The decomposition into SMR and AMR is obtained by setting θAMR = 0 and θAH = 0 in the fitted model, so the separated contributions are computed from the same fitted parameters that reproduce the total magnetoresistance. There is no independent measurement of θAMR or θSH against which the separation is validated. Because the functional forms of the SMR and AMR terms are not orthogonal in general, the decomposition could be an artifact of the model rather than a physical separation. The authors should validate the procedure, for example by comparing extracted θAMR with values from single-layer AMR measurements or by testing whether the separation is stable under small perturbations of the fitting constraints.","section":"Sec. II, Eq. (22); Sec. IV"},{"comment":"The paper states that 'we assume θAMR and θSH to be independent of layer thickness, which may result in overestimated parameters for very thin ferromagnetic layers.' This assumption is load-bearing: if θAMR varies with ferromagnet thickness, as is known for very thin layers, the fitted separation between SMR and AMR will be systematically wrong, and the differences between the tF and tH series in Table II may reflect exactly such thickness dependence. The authors should either justify the assumption quantitatively (e.g., by comparing fits with and without thickness-dependent angles) or restrict the conclusions to thickness ranges where the assumption is supported by the data.","section":"Sec. IV, thickness independence assumption"},{"comment":"The model uses several simplifying assumptions whose quantitative impact on the fitted parameters is not assessed: transparent interfaces (GF → ∞), negligible imaginary spin-mixing conductance (Gi = 0), fixed spin polarization β = 0.3, and neglect of AHE. While each assumption is plausible, the fitted values of θSH, θAMR, and Gr may be sensitive to them, especially the transparent-interface and β values. A sensitivity analysis, at least for the most influential parameters, would substantially strengthen the paper's central claim of precise determination.","section":"Sec. II and Sec. IV, model simplifications"}],"minor_comments":[{"comment":"The text contains several typographical errors, including 'in-p lane' in the title header and 'meansurements' in the Introduction; a careful proofreading pass is needed.","section":"Abstract and Introduction"},{"comment":"The figure captions are difficult to parse because they use abbreviations like 'H' and 'F' without fully defining them in the caption; consider spelling out 'heavy metal' and 'ferromagnet' at first use.","section":"Fig. 2 and Fig. 3 captions"},{"comment":"Reference [35] has a typographical extra comma after 'Ralph,' and the Supplemental Material reference [39] is cited as 'URL will be inserted by publisher'; the authors should provide the actual URL or DOI.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and the theoretical derivation is a useful contribution. My main concern is that the quantitative claims in the abstract and Section IV go beyond what the fitting analysis supports, given the lack of uncertainty estimates and the inconsistency between complementary thickness series. I would recommend major revision rather than rejection because the issues are fixable: adding a rigorous uncertainty analysis, validating the SMR/AMR separation with independent measurements, and softening the 'precise determination' claim would address the central weakness. The paper would also benefit from a sensitivity analysis of the model assumptions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe paper's real contribution is on the theory side. The authors add AMR to the drift-diffusion spin-transport equations for the ferromagnetic layer and derive an SMR expression that contains AMR explicitly, following Taniguchi's approach. That is a clean extension of Kim et al.'s SMR model, and the equations in Sec. II are internally consistent. The four experimental systems (W/CoFeB, W/Co, CoFeB/Pt, Co/Pt) are sensible testbeds, and the paper shows the model can capture the qualitative thickness trends of the MR.\n\nWhere it fails to deliver is the 'precise determination' promised in the abstract. The fitting uses five free parameters for each series, and no uncertainties are reported. More importantly, the stress-test note is right: for the same nominal bilayer, fits from the tH and tF series give parameters that differ by factors of two to three (W/Co: |θSH| 0.34 vs 0.47, θAMR 1.1% vs 0.5%; Co/Pt: |θSH| 0.09 vs 0.28, θAMR 2.4% vs 0.7%), and Gr differs by orders of magnitude. Unless those discrepancies are shown to be within noise, the separation between SMR and AMR is not reliable. The decomposition is also entirely model-based: Eq. (22) defines SMR by zeroing θAMR in the fitted model, with no independent check from, e.g., separate AMR measurements or known spin Hall angles. That is a limitation the authors partly acknowledge, but the acknowledgment does not weaken the claim.\n\nThe paper is honest about its main assumption — θAMR and θSH independent of thickness — and flags possible dead layers and varying spin-mixing conductance. Those caveats are in the right spirit, but they sit in tension with the abstract's wording.\n\nWho should read this: experimentalists who work on SMR/AMR separation in metallic bilayers and want a formula for the MR that includes both effects. For that purpose the paper is worth citing. It is not yet a quantitative reference for the specific parameters listed in Table II. A serious revision would add error bars, residual plots, and a discussion (or at least a quantitative test) of the cross-series parameter discrepancies.\n\nRecommendation: yes, send to peer review. A good referee can press for the missing uncertainty analysis, and the theoretical part deserves to see the light. But don't publish as is.","headline":"Worthwhile model extension, but the 'precise SMR/AMR separation' claim needs error bars and a resolution of cross-series parameter discrepancies.","tokens_in":13091,"tokens_out":3039,"would_cite":true,"duration_ms":30594,"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":"An extended drift-diffusion model, with anisotropic magnetoresistance explicitly included in the ferromagnetic layer, separates spin Hall magnetoresistance from AMR in four metallic bilayers and yields separate spin Hall and AMR angles.","keywords":["spin Hall magnetoresistance","anisotropic magnetoresistance","drift-diffusion model","spin Hall angle","AMR angle","heavy-metal/ferromagnet bilayers","W/CoFeB","Pt/Co"],"falsifier":"Measure the AMR of single ferromagnetic layers (with no heavy metal) across the same thickness range used in the bilayer fits; if the single-layer $\\theta_{AMR}$ varies with thickness while the bilayer model treats it as constant, the SMR/AMR separation is misattributing the thickness trend. A second check would be to fit the bilayer data with $\\theta_{AMR}(t_F)$ as a free function and compare residuals.","tokens_in":12088,"feed_emoji":"🧲","tokens_out":7611,"duration_ms":79239,"temperature":0.7,"pith_summary":"Spin Hall magnetoresistance (SMR) and anisotropic magnetoresistance (AMR) both contribute when a current flows in a heavy-metal/ferromagnet bilayer, and standard SMR-only analyses risk misreading one as the other. This paper shows that putting AMR (and anomalous Hall) terms directly into the spin drift-diffusion equations for the ferromagnetic layer produces a magnetoresistance formula whose angle and thickness dependence separates the two effects. Fitted to resistance measurements on W/Co20Fe60B20, W/Co, Co20Fe60B20/Pt, and Co/Pt bilayers, the model yields separate estimates of the spin Hall angle and the AMR angle for each system. The separation matters for magnetic field sensors and for any experiment that extracts spin transport parameters from magnetoresistance data.","feed_headline":"One model separates spin Hall and anisotropic magnetoresistance","feed_subtitle":"Extended drift-diffusion fits give separate spin Hall and AMR angles for W/CoFeB, W/Co, CoFeB/Pt, and Co/Pt.","key_machinery":"The carrying object is an extended spin drift-diffusion model for the H/F bilayer. Charge and spin currents in the heavy metal are governed by the spin Hall effect, while in the ferromagnet the current equations include $\\theta_{AMR}$ and $\\theta_{AH}$ terms alongside spin polarization $\\beta$; the two layers are connected by an interface boundary condition written in terms of spin-mixing conductances. Solving these equations with open boundary conditions yields the conductivity tensor whose anisotropy, inserted into $MR = (\\rho_{xx}(\\hat m \\parallel \\hat e_x)-\\rho_{xx}(\\hat m \\parallel \\hat e_y))/\\rho_{xx}(\\hat m \\parallel \\hat e_x)$, is the measured quantity. The mechanism that does the work is the explicit separation $\\sigma_x = \\sigma_x^{SH}+\\sigma_x^{AMR}$ and $\\sigma_y = \\sigma_y^{SH}+\\sigma_y^{AH}$, which lets the angular and thickness dependence of resistance single out each contribution.","core_discovery":"On its own terms, the paper claims that the measured magnetoresistance of heavy-metal/ferromagnet bilayers can be decomposed into a spin Hall part and an AMR part by solving coupled spin and charge diffusion equations in both layers with AMR terms included in the ferromagnet. The central result is the decomposition $MR \\approx (\\sigma_y-\\sigma_x)/\\sigma_0$, where $\\sigma_x$ contains $\\sigma_x^{SH}+\\sigma_x^{AMR}$ and $\\sigma_y$ contains spin Hall and anomalous Hall contributions; setting $\\theta_{AMR}\\to 0$ and $\\theta_{AH}\\to 0$ recovers the earlier SMR-only model. For the four studied bilayers, the fits give $|\\theta_{SH}|$ roughly 0.09–0.47 and $\\theta_{AMR}$ roughly 0.13–2.4%, with W-based samples dominated by SMR and Pt-based samples dominated by AMR. The paper also reports negative SMR in Co-based systems, where large AMR would otherwise be mistaken for spin Hall physics.","pith_inferences":["If thin-film AMR is genuinely thickness dependent, as suggested by the paper's own caveat, then the constant-$\\theta_{AMR}$ fit likely underestimates AMR in thick layers and overestimates it in very thin ones; this could be tested by capping single ferromagnetic layers with an insulating spacer instead of a heavy metal.","The anomalous Hall terms, which are negligible for the in-plane magnetized systems studied here, should become visible in out-of-plane magnetized bilayers or in ferromagnets with larger anomalous Hall angles, where the same framework predicts an extra $\\sigma_y^{AH}$ contribution to the transverse conductivity.","Because the model gives closed-form expressions for the conductivity tensor, one could extract both angles from a single angle-resolved magnetoresistance measurement rather than a full thickness series, speeding up materials screening."],"forward_implications":["In Pt-based bilayers, where the spin Hall angle is small, the measured magnetoresistance is dominated by AMR, so SMR-only fits would overestimate spin Hall transport parameters.","In Co-based bilayers, large AMR can mask or reverse the apparent SMR signal; the model identifies a negative SMR contribution that only appears once AMR is removed.","The model provides separate $\\theta_{SH}$ and $\\theta_{AMR}$ values for W/CoFeB, W/Co, CoFeB/Pt, and Co/Pt, which can be used to interpret other magnetoresistance-based spin transport experiments.","The decomposed magnetoresistance is directly relevant to optimizing magnetic stray field sensors, where the AMR background must be separated from the SMR response."],"supporting_citations":[{"why":"Provides the baseline SMR-in-metallic-bilayers model that this paper extends by adding AMR and AHE.","marker":"[4]"},{"why":"Supplies the ferromagnet charge and spin current equations containing AMR, anomalous Hall, and spin polarization terms.","marker":"[26]"},{"why":"Extends the same formalism to magnetoresistance from anomalous Hall charge-spin conversion in ferromagnet/normal-metal bilayers.","marker":"[27]"},{"why":"Provides the interfacial spin current boundary condition written in terms of spin-mixing conductances.","marker":"[37]"},{"why":"Defines the spin Hall magnetoresistance theory for ferromagnet/normal-metal hybrids on which the decomposition builds.","marker":"[3]"},{"why":"Provides the method for thickness-dependent resistivity determination and the Pt/Co anomalous SMR comparison used as a reference.","marker":"[6]"}],"fun_headline_variants":["Separating spin Hall and AMR in metal bilayers","Model isolates spin Hall magnetoresistance from AMR","AMR included in diffusion equations to separate SMR","W/Co and Co/Pt reveal distinct SMR and AMR roles","Spin Hall vs anisotropic magnetoresistance decomposed"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The fitting assumes the spin Hall angle and the AMR angle do not change with layer thickness, although the paper itself notes this may overestimate parameters for very thin ferromagnetic layers.","fun_headline_variants_meta":{"raw":{"variants":["Separating spin Hall and AMR in metal bilayers","Model isolates spin Hall magnetoresistance from AMR","AMR included in diffusion equations to separate SMR","W/Co and Co/Pt reveal distinct SMR and AMR roles","Spin Hall vs anisotropic magnetoresistance decomposed"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000191,"raw_usage":{"total_tokens":1346,"prompt_tokens":948,"completion_tokens":398,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":564,"completion_tokens_details":{"reasoning_tokens":318}},"tokens_in":564,"tokens_out":398,"duration_ms":4458,"temperature":1.0,"reasoning_tokens":318,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:57:43.476316+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the AMR of single ferromagnetic layers (with no heavy metal) across the same thickness range used in the bilayer fits; if the single-layer $\\theta_{AMR}$ varies with thickness while the bilayer model treats it as constant, the SMR/AMR separation is misattributing the thickness trend. A second check would be to fit the bilayer data with $\\theta_{AMR}(t_F)$ as a free function and compare residuals.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the baseline SMR-in-metallic-bilayers model that this paper extends by adding AMR and AHE."},{"cited_title":"Taniguchi, J","cited_arxiv_id":null,"evidence_quote":"Supplies the ferromagnet charge and spin current equations containing AMR, anomalous Hall, and spin polarization terms."},{"cited_title":"Taniguchi, Magnetoresistance generated from charge-spin conversion by anomalous Hall effect in metallic ferromagnetic/nonmagnetic bilayers, Phys","cited_arxiv_id":null,"evidence_quote":"Extends the same formalism to magnetoresistance from anomalous Hall charge-spin conversion in ferromagnet/normal-metal bilayers."},{"cited_title":"Brataas, G","cited_arxiv_id":null,"evidence_quote":"Provides the interfacial spin current boundary condition written in terms of spin-mixing conductances."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the spin Hall magnetoresistance theory for ferromagnet/normal-metal hybrids on which the decomposition builds."},{"cited_title":"Kawaguchi, D","cited_arxiv_id":null,"evidence_quote":"Provides the method for thickness-dependent resistivity determination and the Pt/Co anomalous SMR comparison used as a reference."}],"review_version":1}