REVIEW 2 major objections 6 minor 43 references
Polarized electroluminescence with magnetic spectral tuning in van der Waals magnet CrSBr
T0 review · 2 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read CrSBr LED emits 94.3% linearly polarized light that a magnetic field continuously tunes by 8 meV, establishing electrically driven van der Waals magnets as reconfigurable polarized light sources.
desk verdict First electrically driven polarized, magnetically tunable emission from CrSBr — a clean device demo that needs an I(B) control to close the thermal-artifact loophole. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central mechanism is exciton-assisted inelastic tunneling in a graphene/h-BN/graphene tunnel junction placed adjacent to a CrSBr flake: tunneling electrons lose energy inelastically and transfer it to excitons in CrSBr, which then radiatively recombine. The magnetic tunability comes from spin-canting-modulated interlayer hybridization, where the interlayer hopping integral scales as $t_h \propto \cos(\theta/2)$ with $\theta$ the angle between adjacent-layer magnetizations, renormalizing the band gap and shifting the exciton energy. The intrinsically polarized emission arises from CrSBr's quasi-1D electronic structure, with a nearly flat conduction band along the a-axis and a dispersive band along the b-axis, making optical transitions strongly linear-polarized along the b-axis.
What would settle it
A concrete experiment that would settle the claim is to measure the EL spectrum, the tunneling current, and the device temperature simultaneously as a function of magnetic field at a fixed bias, and to compare the EL peak shift with a magnetization measurement of the same CrSBr flake. If the EL peak shift does not track the spin-canting curve, or if it occurs with the field along the easy axis where no canting occurs, or if the tunneling current changes substantially over the same field range, the spin-canting attribution would be falsified.
Extended reading notes
Core claim
The central claim is that an electroluminescent device based on exciton-assisted inelastic tunneling in the anisotropic van der Waals magnet CrSBr emits polarized light whose wavelength is continuously tunable by a magnetic field. The emitted light is dominated by the surface exciton peak at about 1.34 eV with a degree of linear polarization of 94.3% (reaching 99.0% in a thicker device), and applying a magnetic field along the hard c-axis produces a continuous 8 meV redshift of the electroluminescence peak as the interlayer spin configuration goes from antiferromagnetic to ferromagnetic through spin canting. The tuning follows a quadratic energy shift $\Delta E \propto B^2$ below the saturation field of about 2.3 T. The paper attributes this behavior to field-dependent interlayer exciton hybridization, with interlayer hopping scaling as $t_h \propto \cos(\theta/2)$, where $\theta$ is the angle between adjacent-layer magnetizations. The device operates at ultralow currents below 1 microampere because energy transfer via inelastic tunneling bypasses the need for direct carrier injection into the emission layer.
Load-bearing premise
The load-bearing premise is that the observed 8 meV electroluminescence redshift is caused by spin canting in CrSBr through interlayer hybridization, not by magnetic-field-induced changes in the tunneling current, device temperature, or other artifacts; the paper relies on prior photoluminescence studies for the saturation field of 2.3 T and does not report EL current versus field or direct magnetization measurements on the device.
Editorial extensions
If this is right
- If the central claim holds, CrSBr and similar van der Waals magnets can serve as electrically driven, polarization-preserving light sources whose emission wavelength is set by an external magnetic field.
- Combining near-unity linear polarization with continuous spectral tuning in a single device would remove the need for external polarizers and multiple fixed-wavelength emitters in on-chip photonics.
- Magnetic-field-controlled spectral tuning could enable reconfigurable optical interconnects and polarization-resolved spectroscopy without any moving parts.
- The exciton-assisted tunneling mechanism could generalize to other van der Waals magnets and magnetic semiconductors, offering a generic route to magnetically tunable electroluminescence.
- The preservation of polarization in electroluminescence relative to photoluminescence suggests tunneling-based energy transfer does not depolarize excitons, which may allow electrical probing of spin-exciton physics in magnetic van der Waals materials.
- These devices, if developed further, could act as compact sources for near-infrared applications in telecommunications, biological imaging, and quantum technologies that currently rely on bulkier or less tunable emitters.
Reading between the lines
- A direct test that is not reported in the paper would be to measure the tunneling current and device temperature against magnetic field at fixed bias; if the current changes appreciably with field, part of the 8 meV shift could come from Joule heating or field-dependent electrode band structure rather than from spin canting in CrSBr itself.
- The quadratic-in-field energy shift below saturation is consistent with the interlayer-hopping picture, but comparing the EL shift to a magnetization curve measured on the same device would establish whether the tuning tracks the interlayer spin correlation function itself or simply $B^2$.
- The paper's operation is limited to temperatures below about 110 K; extending magnetic spectral tuning to room temperature would require a different van der Waals magnet with a higher magnetic ordering temperature, a direction the paper leaves unexplored.
- The paper notes parallel findings posted by another group; independent confirmation of magnetically tunable electroluminescence in CrSBr would strengthen the case that exciton-assisted tunneling is a robust tool for electrically probing and controlling magnetic excitons.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports electrically driven polarized light emission from a CrSBr/h-BN/graphene vertical heterostructure, operating via exciton-assisted inelastic tunneling. The authors demonstrate EL spatially confined to the overlap region, a linear EL-intensity-versus-current relation, near-unity degree of linear polarization (DOLP ≈ 94.3% for Device 2, 99.0% for Device 1), and a continuous magnetic-field-induced redshift of the dominant P exciton peak of about 8 meV below the saturation field B_sat = 2.3 T. The magnetic tuning is attributed to field-induced spin canting and interlayer exciton hybridization, with an interlayer hopping parameter scaling as t_h ∝ cos(θ/2). The claim is that this constitutes the first electrically driven van der Waals magnet light source combining intrinsic polarization and continuous spectral tuning.
Significance. If the central claim holds, the work is significant: it demonstrates a new functionality for van der Waals magnets, merging polarized emission with magnetic spectral control in an electrically driven device. The measurement set is internally coherent: the linear EL-current relation (Fig. 2b), spatial mapping (Fig. 1d), PL/EL spectral comparison (Fig. 1e), polarization analysis (Fig. 3), and field-dependent peak shifts (Fig. 4) are mutually consistent. The authors also report on two devices and show in the Supplementary Information that Device 1 reproduces the magnetic-field-induced behavior. The main weakness is the absence of a direct electrical control for the field-dependent EL spectra, which is load-bearing for the magnetic-tuning attribution.
major comments (2)
- [§4, Fig. 4b,c] The central magnetic-tuning claim is supported by EL spectra taken at a fixed bias V_bias = 3.1 V, but the paper does not report the EL tunneling current I(B) or the device temperature during the magnetic-field sweep. Because the P-peak energy is strongly temperature dependent (Fig. 2c shows a pronounced redshift from 2 K to 90 K), and because a magnetic field can alter the tunneling conductance of the graphene/h-BN/FLG/CrSBr stack through band shifts or magnetoresistance, the observed 8 meV redshift could in principle be caused by a field-induced change in dissipated power or in the electrical bias point rather than by spin-canting-driven band renormalization. The PL comparison (Fig. 4a) is supportive but does not close this control, since PL is optically pumped and does not constrain the electrical state of the EL junction. I request that the authors provide an I(B) trace at the same fixed bias, or EL spectra acquired at constant current, to rule out this alternative explanation.
- [§4, Fig. 4c] The quantitative model for the field-induced shift is presented as a quadratic fit (ΔE ∝ B^2) with a single free coefficient, while the underlying parameter t_h ∝ cos(θ/2) is taken from prior literature. The fit is not a parameter-free prediction, and the manuscript does not report an independent measurement of the spin canting angle in the operating device (e.g., magnetization or magneto-transport data). Thus the mechanism is plausible but not uniquely constrained by the presented data. The missing I(B) control in the previous comment is the more direct way to solidify the attribution; providing a fit of the full angular dependence using the known CrSBr spin Hamiltonian would strengthen the quantitative claim.
minor comments (6)
- [§3, Fig. 3c,d] DOLP values are quoted without uncertainties or a statement of the number of independent angle settings; please provide error bars or confidence intervals for the polarization degree.
- [§4, Fig. 4c] The EL peak positions in Fig. 4c are shown without error bars; please state the spectral fitting procedure and the uncertainty of the peak-energy determination, especially because the 8 meV shift is not much larger than typical linewidth variations.
- [Introduction] The phrase "three breakthroughs" is promotional and would be better replaced by a neutral description of the device capabilities.
- [Introduction, paragraph 2] There is a typo: "linealy polarized excitons" should be "linearly polarized excitons".
- [References, [33]] Reference 33 is cited to support the nearly flat conduction band along Γ–X in CrSBr, but the cited paper (Funk et al., Phys. Rev. Research 3, L042019, 2021) is about WSe2 phonon sidebands. Please verify the citation or replace it with a CrSBr-specific band-structure reference.
- [Fig. 1d caption] "Spacial mapping" should be "Spatial mapping".
Circularity Check
No significant circularity: the reported EL polarization and magnetic spectral tuning are direct measurements, and the interpretive model is imported from independent prior work rather than derived from the same data.
full rationale
The paper's central claims are measured quantities, not outputs of a fitted model that has been renamed as a prediction. The DOLP of approximately 94.3% is obtained by angle-resolved EL intensity measurements (Fig. 3d), and the 8 meV magnetic-field-induced EL redshift is read directly from the EL spectra in Fig. 4b, with the quadratic dependence in Fig. 4c presented as a fit to those data. The spin-canting interpretation uses B_sat = 2.3 T and the interlayer hopping scaling t_h proportional to cos(theta/2), both taken from prior photoluminescence and magneto-optical studies (refs. 25, 39-42), which are external to this paper and independently established. No load-bearing self-citation chain appears; the cited tunneling mechanism (ref. 27) and CrSBr exciton properties (refs. 24-26, 37) are external. The main scientific weakness is the absence of an EL current-versus-field control during fixed-bias field sweeps, which could in principle allow thermal or bias-point artifacts to contribute to the redshift; however, this is a missing control or a risk to the mechanistic attribution, not a circularity in which the conclusion is equivalent to the input by construction. The polarization and tuning are measured, and the model is imported from independent literature, so the derivation chain is not circular.
Assumptions & free parameters
free parameters (1)
- quadratic tuning coefficient c in Delta E = c B^2 =
not stated in text
assumptions (3)
- domain assumption The observed EL arises from exciton-assisted inelastic tunneling energy transfer to CrSBr, not from direct carrier injection or impact ionization.
- domain assumption The magnetic-field-induced redshift of the EL P peak reflects CrSBr spin-canting band renormalization with B_sat = 2.3 T from previous studies.
- domain assumption The assignment of the P, X*, and X_d peaks to surface excitons, bulk excitons, and defect states follows the prior literature.
Cite this review
Pith. "Pith review of Polarized electroluminescence with magnetic spectral tuning in van der Waals magnet CrSBr." pith.science (2026). https://pith.science/paper/IAT6PADB
@misc{pith2026250606734,
author = {Pith},
title = {Pith review of: Polarized electroluminescence with magnetic spectral tuning in van der Waals magnet CrSBr},
year = {2026},
howpublished = {\url{https://pith.science/paper/IAT6PADB}},
note = {Machine review of arXiv:2506.06734}
}
abstract
Polarized wavelength-tunable electroluminescence (EL) represents a critical on-demand functionality for next-generation optoelectronics. While conventional van der Waals (vdW) EL devices offer discrete wavelength switching constrained by fixed emission states, we report a novel platform enabling continuous spectral tuning combined with intrinsically polarized emission. By leveraging exciton-assisted inelastic tunneling in the anisotropic magnet CrSBr, our devices achieve uniform EL with a near unity degree of linear polarization ($\approx$ 94.3$\%$). The strong magneto-electronic coupling in CrSBr facilitates continuous magnetic-field-controlled spectral tuning through spin canting-induced band renormalization. This work establishes vdW magnets as a versatile platform for developing reconfigurable polarized light sources with simultaneous spectral and polarization control.
Reference graph
Works this paper leans on
-
[43]
All-electrical near-field injection of excitons in a van der Waals antiferromagnet
Ziegler, J. D.et al.All-electrical near-field injection of ex- citons in a van der waals antiferromagnet.arXiv preprint arXiv:2505.15457(2025)
work page Pith review arXiv 2025
-
[1]
B¨ unzli, J.-C. G. & Eliseeva, S. V. Lanthanide nir lumi- nescence for telecommunications, bioanalyses and solar energy conversion.Journal of Rare Earths28, 824–842 (2010)
work page 2010
-
[2]
Hong, G., Antaris, A. L. & Dai, H. Near-infrared fluo- rophores for biomedical imaging.Nature biomedical en- gineering1, 0010 (2017)
work page 2017
-
[3]
Lee, C.et al.Indefinite and bidirectional near-infrared nanocrystal photoswitching.Nature618, 951–958 (2023)
work page 2023
-
[4]
Yuan, F.et al.Bright and stable near-infrared lead-free perovskite light-emitting diodes.Nature Photonics18, 9 (2024)
work page 2024
-
[5]
Pu, J.et al.Continuous color-tunable light-emitting de- vices based on compositionally graded monolayer transi- tion metal dichalcogenide alloys.Advanced Materials34, 2203250 (2022)
work page 2022
-
[6]
Duan, X.et al.Synthesis of ws2 x se2–2 x alloy nanosheets with composition-tunable electronic proper- ties.Nano letters16, 264–269 (2016)
work page 2016
-
[7]
Wang, S., Ukhtary, M. S. & Saito, R. Strain effect on cir- cularly polarized electroluminescence in transition metal dichalcogenides.Physical Review Research2, 033340 (2020)
work page 2020
Show all 43 references
-
[8]
J.et al.Tunable emission from localized excitons deterministically positioned in monolayer p–n junctions.ACS Photonics9, 3067–3074 (2022)
Lenferink, E. J.et al.Tunable emission from localized excitons deterministically positioned in monolayer p–n junctions.ACS Photonics9, 3067–3074 (2022)
2022
-
[9]
Qi, Y.et al.Recent progress in strain engineering on van der waals 2d materials: Tunable electrical, electrochemi- cal, magnetic, and optical properties.Advanced Materials 35, 2205714 (2023)
2023
-
[10]
Henck, H.et al.Light sources with bias tunable spec- trum based on van der waals interface transistors.Nature Communications13, 3917 (2022)
2022
-
[11]
Y.et al.Electrically controlled emission from singlet and triplet exciton species in atomically thin light- emitting diodes.Physical Review B103, L161411 (2021)
Joe, A. Y.et al.Electrically controlled emission from singlet and triplet exciton species in atomically thin light- emitting diodes.Physical Review B103, L161411 (2021)
2021
-
[12]
S.et al.Electrically tunable excitonic light- emitting diodes based on monolayer wse2 p–n junctions
Ross, J. S.et al.Electrically tunable excitonic light- emitting diodes based on monolayer wse2 p–n junctions. Nature nanotechnology9, 268–272 (2014)
2014
-
[13]
Shin, J.-C.et al.Electrically confined electrolumines- cence of neutral excitons in wse2 light-emitting transis- tors.Advanced Materials36, 2310498 (2024)
2024
-
[14]
Wang, J.et al.Polarity tunable trionic electrolumines- cence in monolayer wse2.Nano Letters19, 7470–7475 (2019)
2019
-
[15]
Advanced Materials33, 2101375 (2021)
Zhu, Y.et al.A high-efficiency wavelength-tunable monolayer led with hybrid continuous-pulsed injection. Advanced Materials33, 2101375 (2021)
2021
-
[16]
Paur, M.et al.Electroluminescence from multi- particle exciton complexes in transition metal dichalco- genide semiconductors.Nature communications10, 1709 (2019)
2019
-
[17]
Wang, J.et al.Polarized light-emitting diodes based on anisotropic excitons in few-layer res2.Advanced Materi- als32, 2001890 (2020)
2020
-
[18]
H.et al.The magnetic genome of two- dimensional van der waals materials.ACS nano16, 6960–7079 (2022)
Wang, Q. H.et al.The magnetic genome of two- dimensional van der waals materials.ACS nano16, 6960–7079 (2022)
2022
-
[19]
Lee, K.et al.Magnetic order and symmetry in the 2d semiconductor crsbr.Nano Letters21, 3511–3517 (2021)
2021
-
[20]
& Zhou, Y
Duan, J. & Zhou, Y. Magnetic order as a tuning knob for coulomb correlation.Nature Materials24(2025)
2025
-
[21]
Lin, Z.et al.Magnetic van der waals materials: Syn- thesis, structure, magnetism, and their potential appli- cations.Chinese Physics B31, 087506 (2022)
2022
-
[22]
Qian, T.-X., Zhou, J., Cai, T.-Y. & Ju, S. Anisotropic electron-hole excitation and large linear dichroism in the two-dimensional ferromagnet crsbr with in-plane magne- tization.Physical Review Research5, 033143 (2023)
2023
-
[23]
npj 2D Materials and Applications9(2025)
Mondal, P.et al.Raman polarization switching in crsbr. npj 2D Materials and Applications9(2025)
2025
-
[24]
& Tan, P.-H
Wu, J.-B., Wu, H. & Tan, P.-H. Magneto-optical interac- tions in layered magnets.Advanced Functional Materials 34, 2312214 (2024)
2024
-
[25]
P.et al.Interlayer electronic coupling on de- mand in a 2d magnetic semiconductor.Nature Materials 20, 1657–1662 (2021)
Wilson, N. P.et al.Interlayer electronic coupling on de- mand in a 2d magnetic semiconductor.Nature Materials 20, 1657–1662 (2021)
2021
-
[26]
Shao, Y.et al.Magnetically confined surface and bulk excitons in a layered antiferromagnet.Nature Materials 1–8 (2025)
2025
-
[27]
Wang, L.et al.Exciton-assisted electron tunnelling in van der waals heterostructures.Nature Materials22, 1094–1099 (2023)
2023
-
[28]
& Snow, E
Lenzlinger, M. & Snow, E. H. Fowler-nordheim tunneling into thermally grown sio2.Journal of Applied Physics40, 278–283 (1969)
1969
-
[29]
R.et al.Probing magnetism in 2d van der waals crystalline insulators via electron tunneling.Science360, 1218–1222 (2018)
Klein, D. R.et al.Probing magnetism in 2d van der waals crystalline insulators via electron tunneling.Science360, 1218–1222 (2018)
2018
-
[30]
The journal of physical chemistry letters15, 6010–6016 (2024)
Lin, K.et al.Probing the band splitting near theγ point in the van der waals magnetic semiconductor crsbr. The journal of physical chemistry letters15, 6010–6016 (2024)
2024
-
[31]
& Coronado, E
Marques-Moros, F., Boix-Constant, C., Ma˜ nas-Valero, S., Canet-Ferrer, J. & Coronado, E. Interplay between optical emission and magnetism in the van der waals magnetic semiconductor crsbr in the two-dimensional limit.ACS nano17, 13224–13231 (2023)
2023
-
[32]
Klein, J.et al.Sensing the local magnetic environment through optically active defects in a layered magnetic semiconductor.ACS nano17, 288–299 (2022)
2022
-
[33]
Funk, V.et al.Spectral asymmetry of phonon sideband luminescence in monolayer and bilayer wse 2.Physical Review Research3, L042019 (2021)
2021
-
[34]
Cheng, R.et al.Electroluminescence and photocurrent generation from atomically sharp wse2/mos2 heterojunc- tion p–n diodes.Nano letters14, 5590–5597 (2014). 7
2014
-
[35]
Wang, S.et al.Efficient carrier-to-exciton conversion in field emission tunnel diodes based on mis-type van der waals heterostack.Nano letters17, 5156–5162 (2017)
2017
-
[36]
Temperature dependence of the energy gap in semiconductors.Physica34, 149–154 (1967)
Varshni, Y. Temperature dependence of the energy gap in semiconductors.Physica34, 149–154 (1967)
1967
-
[37]
Klein, J.et al.The bulk van der waals layered magnet crsbr is a quasi-1d material.ACS nano17, 5316–5328 (2023)
2023
-
[38]
Bianchi, M.et al.Charge transfer induced lifshitz transi- tion and magnetic symmetry breaking in ultrathin crsbr crystals.Physical Review B108, 195410 (2023)
2023
-
[39]
Dirnberger, F.et al.Magneto-optics in a van der waals magnet tuned by self-hybridized polaritons.Nature620, 533–537 (2023)
2023
-
[40]
J., Noble, C
Brennan, N. J., Noble, C. A., Tang, J., Ziebel, M. E. & Bae, Y. J. Important elements of spin-exciton and magnon-exciton coupling.ACS Physical Chemistry Au 4, 322–327 (2024)
2024
-
[41]
M.et al.Tunable interaction between exci- tons and hybridized magnons in a layered semiconductor
Diederich, G. M.et al.Tunable interaction between exci- tons and hybridized magnons in a layered semiconductor. Nature Nanotechnology18, 23–28 (2023)
2023
-
[42]
J.et al.Exciton-coupled coherent magnons in a 2d semiconductor.Nature609, 282–286 (2022)
Bae, Y. J.et al.Exciton-coupled coherent magnons in a 2d semiconductor.Nature609, 282–286 (2022)
2022
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.