Pith. sign in

REVIEW 3 major objections 5 minor 6 references

Experimental observation of coupled valley and spin Hall effect in p-doped WSe2 devices

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read The paper reports an all-electrical WSe2/graphene device that produces a transverse out-of-plane spin current via the spin-locked valley Hall effect and detects it with a ferromagnetic nonlocal probe.

desk verdict Plausible first all-electrical spin-valley Hall effect in WSe2, but the evidence is thinner than the paper's wording—small signals, averaged without error bars, and linear background subtraction in some traces—so it deserves peer review with a demand for raw data. read the letter →

arxiv 1908.01396 v1 pith:BINQMPZ5 submitted 2019-08-04 cond-mat.mes-hall

classification cond-mat.mes-hall
keywords valleyHalleffectspin-valleylockingout-of-planespinpolarizationWSe2nonlocalvalvegraphenechannelcurrentgenerationtransitionmetaldichalcogenides
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that an ordinary charge current through p-doped monolayer WSe2 produces a transverse current of spins polarized out of the plane, through the spin-locked valley Hall effect in the valence band. The spins are injected into an overlapping graphene channel and detected as a nonlocal voltage across a ferromagnetic contact whose magnetization is tilted by an out-of-plane magnetic field. The measured resistance tracks the contact's out-of-plane magnetization component, disappears when the contact is nonmagnetic, and reverses sign when the current direction reverses. If correct, this is the first all-electrical demonstration of out-of-plane spin generation in a two-dimensional transition metal dichalcogenide, and it provides a spin source that could switch perpendicularly magnetized memories without the external-field assistance that conventional spin Hall materials require.

What carries the argument

The load-bearing mechanism is spin-valley locking in the WSe2 valence band: time-reversal symmetry enforces $\Omega(K)=-\Omega(-K)$ and $s(K)=-s(-K)$, and the ~450 meV spin splitting leaves only one spin sub-band at the Fermi level per valley. The anomalous velocity $\boldsymbol{v}=(e/\hbar)\,\boldsymbol{E}\times\boldsymbol{\Omega}(\boldsymbol{k})$ then sends opposite-spin holes to opposite transverse edges, converting a longitudinal charge current into a transverse out-of-plane spin current, the coupled valley and spin Hall effect. The detection harnesses a permalloy electrode with in-plane anisotropy: an external $B_z$ tilts its magnetization out of plane by a controllable fraction $m_z$, independently calibrated by anomalous Hall effect on a co-processed Hall bar, so the nonlocal voltage is a direct map of the out-of-plane spin chemical potential in the graphene channel.

What would settle it

Replace the in-plane-anisotropy probe with a perpendicularly magnetized electrode: the out-of-plane spin interpretation predicts a maximum nonlocal signal at zero field that reverses with current polarity, whereas an in-plane probe requires the field to cant the magnetization and so shows signal only for $|B_z|>0$. Observing the zero-field signal with a perpendicular detector would confirm the claim; failing to see it would refute the central interpretation.

Watch

Extended reading notes

Core claim

Under a bias applied to p-doped monolayer WSe2, holes in the valence band are spin-valley locked: the K valley carries one out-of-plane spin and the -K valley the opposite. The finite Berry curvature gives these valleys opposite anomalous velocities, so the electric field separates them transversely; the result is a spin current with polarization $\pm z$ flowing into the overlaying graphene. The nonlocal ferromagnetic detector measures the spin chemical potential only through its out-of-plane component, which is why the signal grows continuously as the applied field $B_z$ pulls the detector magnetization out of plane and saturates for $|B_z|>0.8$ T. Reversing the charge current reverses the spin polarization, and replacing the ferromagnetic probe with a nonmagnetic electrode removes the signal. A drift-diffusion model with valley chemical potentials and interface valley-memory loss reproduces the sign of the nonlocal voltage and puts the interfacial spin polarization at about 38%.

Load-bearing premise

The load-bearing assumption is that the magnetic probe in the nonlocal device tilts out of plane under $B_z$ exactly as the separate calibration Hall bar does; if the two rotate differently, the overlay between measured signal and $m_z(B_z)$ loses its meaning.

Editorial extensions

If this is right

  • A monolayer WSe2 device can act as a purely electrical source of out-of-plane spin current, with no optical excitation and no external magnetic field required for generation.
  • Because reversing the charge current reverses the spin polarization, the same device can serve as a switchable spin source for a perpendicularly magnetized electrode.
  • The nonlocal-voltage-versus-$B_z$ method, calibrated by anomalous Hall effect on a co-processed Hall bar, can be applied to other two-dimensional materials to test for out-of-plane spin accumulation.
  • The saturation value of the nonlocal signal, compared with the no-loss upper bound, gives a quantitative estimate of the spin/valley polarization surviving the WSe2/graphene interface, about 38% in these devices.
  • The sign of the nonlocal voltage is tied to the Berry curvature sign and the valley/spin index, so the measurement is a direct electrical readout of the valley polarization flowing into the channel.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the claim holds, other monolayer transition metal dichalcogenides with large valence-band spin splitting, such as MoSe2 and WS2, should show the same effect, and the relative nonlocal signal sizes could be compared with their predicted spin splittings.
  • Integrating the same WSe2 source with a perpendicularly magnetized electrode should enable an attempt at field-free spin-orbit torque switching; a successful switching experiment would be a direct device-level validation of the out-of-plane spin interpretation.
  • Because conduction-band spin splitting in these materials is small, switching the doping from p-type to n-type should suppress the out-of-plane spin signal, providing a clean test that separates valley Hall spin generation from ordinary spin Hall effects.
  • The interface valley-memory-loss model predicts that improving the WSe2/graphene interface, for example by a barrier layer or a cleaner transfer, should move the measured nonlocal voltage toward the calculated roughly three-times-larger upper bound, a quantitative target for materials engineering.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper reports an all-electrical nonlocal spin valve measurement in a monolayer WSe2/graphene hybrid device. A DC current through p-doped WSe2 is claimed to generate a transverse spin current with out-of-plane polarization via the spin-locked valley Hall effect; the spins diffuse into a graphene channel and are detected by a permalloy (Py) electrode whose magnetization is tilted out of plane by a perpendicular field. The authors support the claim with a nonmagnetic control, an independent AHE calibration of mz(Bz), a semiclassical sign analysis, and a drift-diffusion model.

Significance. If the observation is robust, this would be the first all-electrical generation and detection of out-of-plane spin polarization via the valley Hall effect in a TMD, with direct relevance for SOT-MRAM and valleytronics. The paper's strengths include the use of a nonmagnetic control, the sign analysis based on Berry curvature and spin-valley locking, and the observation that the nonlocal resistance saturation tracks the independently measured AHE curve. The device geometry and room-temperature operation are also notable. However, the experimental evidence currently rests on a small number of averaged traces without error bars, and the magnitude modelling is a consistency check rather than a parameter-free prediction.

major comments (3)
  1. [Main text Fig. 3a; Supplementary Section II] The central claim rests on the nonlocal resistance data in Fig. 3a, which are presented as an average of three datasets with no error bars or per-trace statistics. Supplementary Section II states that in two of the six traces (Fig. S7d,f) a linear background was subtracted. This is a load-bearing issue: if the individual traces do not each show a clear zero baseline at Bz=0, a monotonic mz-like increase, and a reversal with current polarity, the averaging and background subtraction could generate the apparent saturating signal from noise or drift. Please show the raw individual traces, quantify the signal-to-noise and the significance of the effect per trace, and specify the averaging procedure and the background subtraction criteria.
  2. [Main text Figs. 2-3a] The interpretation of Rnl(Bz) as tracking mz(Bz) relies on transferring the anomalous Hall curve measured on a standalone Py Hall bar to the actual Py probe on graphene. The calibration assumes that the small Py electrode on graphene has the same demagnetization field, anisotropy, and field history as the Hall bar. If the real probe rotates differently, the saturation field and zero-crossing of Rnl would not be expected to match the AHE curve. Please provide evidence that the calibration transfer is valid, for example a local magnetization measurement on a geometrically identical probe, or, failing that, soften the claim that Rnl mirrors the independently measured mz.
  3. [Supplementary Section III] The sign derivation contains an internal inconsistency in the definition of Vnl. The text defines Vnl = V_NM - V_FM for magnetization along +z, but for magnetization along -z it writes Vnl = V_FM - V_NM > 0, which is the opposite sign convention. Since the sign of the nonlocal voltage is a central claim that distinguishes the spin polarization direction, this inconsistency needs to be resolved and the signs of all four quadrants in Fig. 4 re-derived consistently.
minor comments (5)
  1. [Supplementary Section VII heading] The heading 'Upper Bond' should be 'Upper Bound'.
  2. [Fig. 4 caption] The caption is truncated (it ends with 'lch=1.2um, w1'); please complete the figure caption.
  3. [Main text and Fig. 4] The symbols sz and mz are used without formal definition; please define those quantities when they first appear.
  4. [Supplementary Sections VI-VII] The model contains free parameters (alpha, beta, lv, lv', and sigma_xy_VH), so the 'upper bound' estimate is not parameter-free. The factor-of-3 agreement with experiment should be presented as a consistency check, not as a parameter-free prediction.
  5. [Main text and Supplementary Section II] The control device in Fig. 3b consists of a single trace; please state whether the control experiment was reproduced in multiple devices or provide additional control datasets.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central observation is supported by an external AHE calibration, a nonmagnetic control, and sign/magnitude estimates from independent semiclassical theory.

full rationale

The paper's central claim is an experimental observation of out-of-plane spin accumulation from spin-locked valley Hall effect in WSe2. The derivation chain is not circular. The non-local signal is compared with an independently measured anomalous Hall effect curve on a separate Py Hall bar, and the non-magnetic control device rules out non-spin artifacts. The sign of the expected non-local voltage is derived in Supplementary Section III from the semiclassical Berry-curvature equation of motion and the spin-valley locking relation s(K) = -s(-K), not from the measured data; this is an external theoretical prediction. The magnitude estimate in Supplementary Sections VI-VII uses a drift-diffusion model from Shimazaki et al. and literature/assumed parameters, and is explicitly labelled an upper bound. The discrepancy between the upper bound and the data is attributed after the fact to interface valley loss; it is not a parameter fitted to force agreement. The self-citations (refs 17, 26, 28) are to prior fabrication, valley-Hall, and graphene spin-valve results; the PD efficiency value from ref 28 is from an independent graphene spin-valve experiment and is not a fitted input to the observation. No equation is defined in terms of the claimed result, and no fitted parameter is renamed as a prediction. Therefore no circular step is present.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

No new particles, fields, or conserved quantities are introduced. The only interpretive elements are assumed transport parameters, an interface opacity model, and an inferred interfacial spin polarization (38%), which is a loss factor, not a new entity. The central detection uses external calibration and control devices, so the experimental claim is not generatively defined by the model, though the magnitude estimate depends on the same theory it tests.

free parameters (3)
  • WSe2/graphene interface opacity parameters alpha and beta = not directly measured; set to 0 for the upper-bound estimate, then invoked to account for the roughly 3x discrepancy…
    Introduced in Supplementary Section VI to model valley chemical potential and valley current discontinuities at the WSe2-graphene interface; not measured independently.
  • Valley Hall conductivity sigma_xy_VH = about 2 e^2/h
    Taken from the same spin-valley Berry curvature theory that the experiment tests; used to set the magnitude of the expected nonlocal voltage in Supplementary Sections V and VII.
  • Valley diffusion lengths lv and lv' = 0.6 um and 1 um
    Assumed for WSe2 and graphene; the values are not tied to independent measurements in this paper and affect the upper-bound voltage estimate.
assumptions (5)
  • domain assumption Monolayer WSe2 valence band has spin-valley locking with out-of-plane spin polarization and large spin splitting (~450 meV).
    Relied on from refs 10 and 11; this is the physical mechanism the experiment claims to detect.
  • standard math Semiclassical Berry-curvature equation of motion r_dot = (1/hbar) dE/dk - k_dot x Omega describes carrier motion in WSe2.
    Used in Supplementary Section III to fix the sign of the nonlocal voltage; standard result from Sundaram and Niu.
  • domain assumption Py magnetization orientation under an out-of-plane field follows a macrospin free-energy model with demagnetization field 4*pi*Ms.
    Used in Supplementary Section IV to relate Vnl(Bz) to mz; assumes the separate Hall bar calibration transfers to the actual probe.
  • domain assumption Valley chemical potential difference in graphene obeys a one-dimensional diffusion equation with no valley Hall term.
    Foundation of the nonlocal voltage estimate in Supplementary Section VI; assumes graphene's Berry curvature is negligible and the WSe2/graphene interface can be parameterized by two scalar opacities.
  • domain assumption The O2 plasma treatment leaves a clean monolayer WSe2 under a WOx layer.
    Raman and transfer characteristics support it, but the microscopic structure and disorder are not independently characterized; all spin-valley physics assumes a high-quality monolayer.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Experimental observation of coupled valley and spin Hall effect in p-doped WSe2 devices." pith.science (2026). https://pith.science/paper/BINQMPZ5

@misc{pith2026190801396,
  author       = {Pith},
  title        = {Pith review of: Experimental observation of coupled valley and spin Hall effect in p-doped WSe2 devices},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BINQMPZ5}},
  note         = {Machine review of arXiv:1908.01396}
}
read the original abstract

Giant spin Hall effect (GSHE) has been observed in heavy metal materials such as Ta, Pt, and W, where spins are polarized in the surface plane and perpendicular to the charge current direction. Spins generated in these materials have successfully switched magnets with in-plane magnetic anisotropy (IMA) and perpendicular magnetic anisotropy (PMA) through spin orbit torque (SOT) mechanism. It is generally accepted that PMA magnets are preferred over IMA magnets in data storage applications owing to their large thermal stability even at ultra scaled dimensions. However, SOT switching of PMA magnets by conventional GSHE materials requires either a small external magnetic field, a local dipolar field, or introducing tilted anisotropy to break the symmetry with respect to the magnetization. To deterministically switch a PMA without any additional assistance, nonconventional GSHE materials that can generate spins with polarization perpendicular to the surfaces are needed. Several monolayer transition metal dichalcogenides (TMDs) have been predicted to generate such out of plane spins due to their 2D nature and unique band structures. Interestingly, opposite spins are locked to their respective sub-band in each K valley of the TMD valence band with substantially large energy splitting, which enables polarized spins to be accessible through electrical gating and spatially separated by electric field through the valley Hall effect (VHE). Therefore, spatial separation and accumulation of spins in these 2D TMDs are uniquely referred to as coupled valley and spin Hall effect. Here, we report an experiment of electrical generation of spin current with out of plane polarization in monolayer WSe2 and detection of spin signals through a nonlocal spin valve structure built on a lateral graphene spin diffusion channel that partially overlaps with WSe2.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages

  1. [1]

    Wave-packet dynamics in slowly perturbed crystals: Gradient corrections and Berry-phase effects,

    G. Sundaram and Q. Niu, “Wave-packet dynamics in slowly perturbed crystals: Gradient corrections and Berry-phase effects,” Phys. Rev. B, vol. 59, no. 23, p. 14915, 1999

  2. [2]

    Coupled Spin and Valley Physics in Monolayers of MoS2 and Other Group-VI Dichalcogenides,

    D. Xiao, G.-B. Liu, W. Feng, X. Xu, and W. Yao, “Coupled Spin and Valley Physics in Monolayers of MoS2 and Other Group-VI Dichalcogenides,” Phys. Rev. Lett., vol. 108, no. 19, p. 196802, 2012

  3. [3]

    Generation and detection of pure valley current by electrically induced Berry curvature in bilayer graphene,

    Y. Shimazaki, M. Yamamoto, I. V. Borzenets, K. Watanabe, T. Taniguchi, and S. Tarucha, “Generation and detection of pure valley current by electrically induced Berry curvature in bilayer graphene,” Nat. Phys., vol. 11, no. 12, pp. 1032–1036, 2015

  4. [4]

    Influence of interface spin-flip scattering on spin accumulation and spin currents in magnetic multilayers with collinear magnetizations,

    M. Wawrzyniak, M. Gmitra, and J. Barnaś, “Influence of interface spin-flip scattering on spin accumulation and spin currents in magnetic multilayers with collinear magnetizations,” J. Appl. Phys., vol. 99, no. 2, p. 23905, 2006

  5. [5]

    Electron spin injection and detection at a ferromagnetic- paramagnetic interface (invited),

    M. Johnson and R. H. Silsbee, “Electron spin injection and detection at a ferromagnetic- paramagnetic interface (invited),” J. Appl. Phys., vol. 63, no. 8, pp. 3934–3939, 1988

  6. [6]

    Improvement of spin transfer torque in asymmetric graphene devices,

    C. C. Lin, Y. Gao, A. V. Penumatcha, V. Q. Diep, J. Appenzeller, and Z. Chen, “Improvement of spin transfer torque in asymmetric graphene devices,” ACS Nano, vol. 8, no. 4, pp. 3807–3812, 2014

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.