Destructive interference of second harmonic generation in AA stacked MoTe$_2$/WSe$_2$

Changshen Chen; Jiawei Ruan; Jun He; Kai-Qiang Lin; Siyu Fan; Song Liu; Subi Du; Xiaotong Liao; Yang Xu; Yao Lu

arxiv: 2605.21231 · v1 · pith:YITSR6PGnew · submitted 2026-05-20 · ❄️ cond-mat.mes-hall

Destructive interference of second harmonic generation in AA stacked MoTe₂/WSe₂

Yiduo Wang , Yao Lu , Changshen Chen , Xiaotong Liao , Siyu Fan , Zhenyu Wang , Yaotian Liu , Subi Du

show 9 more authors

Yingze Jia Ye Zhu Yingwei Wang Jun He Song Liu Jiawei Ruan Zhen Chen Kai-Qiang Lin Yang Xu

This is my paper

Pith reviewed 2026-05-21 01:55 UTC · model grok-4.3

classification ❄️ cond-mat.mes-hall

keywords second harmonic generationTMDC heterobilayersexciton resonancesdestructive interferencemoiré materialsnonlinear opticstwisted bilayerspolarization control

0 comments

The pith

Nearly 0-degree stacked MoTe2/WSe2 produces destructive second-harmonic generation because distinct exciton resonances impose a nearly π phase difference between the layers.

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

Standard expectation holds that 0-degree stacking in TMDC bilayers produces constructive interference and stronger second-harmonic generation. This work instead finds strong suppression in nearly AA-stacked MoTe2/WSe2 heterobilayers. Atomic-resolution imaging and exciton-hybridization spectra confirm the stacking geometry, while calculations trace the cancellation to two-photon resonances tied to the WSe2 C exciton and the MoTe2 D exciton. Those resonances create a relative phase shift of nearly 180 degrees in the layer contributions to the nonlinear susceptibility. The result is an anomalous cancellation that can be tuned by small twists, mapping the output polarization onto paths on the Poincaré sphere.

Core claim

Distinct two-photon resonances associated with the WSe2 C exciton and the MoTe2 D exciton generate a nearly π phase difference (Δϕ) in their second-order nonlinear susceptibilities χ^(2), leading to the anomalous destructive interference in nearly 0°-stacked MoTe2/WSe2 heterobilayers.

What carries the argument

The nearly π phase difference Δϕ between layer-resolved χ^(2) responses driven by the mismatched two-photon exciton resonances.

If this is right

In small-twist-angle structures the SHG polarization state is set by the combined effect of twist angle α and the fixed phase difference Δϕ.
The polarization can be plotted as trajectories on the Poincaré sphere.
When Δϕ + 3α equals 180 degrees the emitted light reaches ellipticity near 0.91 and rotates abruptly by 90 degrees in azimuth.
That condition corresponds to a geometric polarization singularity in parameter space.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same resonance-driven phase mechanism may let designers switch between constructive and destructive regimes simply by choosing excitation energy or material pairing.
Polarization singularities could serve as sensitive readouts for local twist or strain in moiré devices.
Analogous phase cancellations might appear in other nonlinear optical processes once multiple exciton channels are active.
Extending the approach to other TMDC pairs with offset exciton energies offers a route to engineer nonlinear phase plates without external fields.

Load-bearing premise

The GW plus Bethe-Salpeter calculations give an accurate relative phase between the layer nonlinear responses without large systematic errors from functional choice or sampling.

What would settle it

Direct measurement showing that the relative phase between the two layers' SHG contributions is not close to π at the relevant excitation energies, or restoration of constructive interference when the laser is tuned away from both resonances.

read the original abstract

The stacking configuration of two-dimensional materials critically governs their optical and electronic responses. Monolayer transition-metal dichalcogenides (TMDC) lack inversion symmetry and exhibit exciton-enhanced second-harmonic generation (SHG). In TMDC bilayers, 60{\deg} (0{\deg}) stacking is conventionally expected to suppress (enhance) SHG owing to destructive (constructive) interference of the layer-resolved nonlinear polarizations. Here, we report an unconventional destructive SHG interference in nearly 0{\deg}-stacked (AA-stacked) MoTe2/WSe2 heterobilayers using two independent probes: atomic-resolution imaging and stacking-sensitive exciton hybridization measurements. Supported by ab initio GW and Bethe-Salpeter equation calculations, we show that distinct two-photon resonances associated with the WSe2 C exciton and the MoTe2 D exciton generate a nearly $\pi$ phase difference ($\Delta\phi$) in their second-order nonlinear susceptibilities $\chi^{(2)}$, leading to the anomalous destructive interference. We further demonstrate that in small-angle twisted MoTe2/WSe2, the SHG polarization state is governed by the interplay between twist angle $\alpha$ and phase difference $\Delta\phi$, and can be mapped onto trajectories on the Poincar\'e sphere. At excitation energies satisfying $\Delta\phi$ + 3$\alpha$ = 180{\deg}, the SHG output becomes nearly circularly polarized (ellipticity ~ 0.91) and undergoes an abrupt 90{\deg} azimuthal rotation, corresponding to a geometric polarization singularity in the parameter space. Our findings open new routes for exciton-resonance engineered nonlinear photonics and stacking-resolved optical functionality in moir\'e materials.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper shows destructive SHG in AA-stacked MoTe2/WSe2 from a near-pi phase shift between WSe2 C and MoTe2 D exciton resonances, with solid stacking confirmation but some uncertainty in the calculated phase.

read the letter

The main thing to know is that this paper reports destructive second-harmonic generation in nearly 0° (AA) stacked MoTe2/WSe2 heterobilayers, which flips the usual expectation that AA stacking produces constructive interference. They trace it to a nearly π phase difference in the layer χ^(2) responses coming from distinct two-photon resonances at the WSe2 C exciton and MoTe2 D exciton energies. Atomic-resolution imaging and exciton hybridization measurements both confirm the stacking, and GW plus Bethe-Salpeter calculations reproduce the phase shift. They also map how small twists plus this phase difference control the output polarization on the Poincaré sphere, including conditions for nearly circular light and abrupt rotations. The experimental checks on stacking are direct and independent of the optical data, which is a strength. The calculations give a plausible mechanism without obvious circularity. The soft spot is the computed phase difference itself. The stress-test concern about sensitivity to functional choice or k-point sampling is reasonable on the basis of the abstract; if the full paper does not include explicit variation tests for the relative phase, that leaves a minor gap in how firmly the numbers pin down exactly π rather than a value close enough to produce the observed cancellation. No quantitative error bars on the SHG suppression appear in the abstract either. This work is for researchers in TMDC nonlinear optics and moiré photonics who care about resonance-tuned interference or polarization control. A reader already working on exciton engineering in heterobilayers would get concrete value from the stacking-resolved mechanism and the twist-phase trajectories. It deserves a serious referee because the experimental identification of the stacking and the new angle on the interference are worth detailed checking, even if the phase calculation needs a bit more scrutiny on robustness.

Referee Report

1 major / 2 minor

Summary. The manuscript reports an anomalous destructive interference in second-harmonic generation (SHG) from nearly 0°-stacked (AA) MoTe₂/WSe₂ heterobilayers, contrary to the conventional expectation of constructive interference for parallel stacking. Atomic-resolution imaging and stacking-sensitive exciton hybridization measurements independently confirm the AA registry; ab initio GW+BSE calculations attribute the effect to a nearly π phase difference Δϕ between the layer-resolved χ⁽²⁾ responses arising from distinct two-photon resonances (WSe₂ C exciton and MoTe₂ D exciton). The work further maps the twist-angle dependence of the SHG polarization state onto trajectories on the Poincaré sphere, identifying conditions for near-circular output and geometric polarization singularities.

Significance. If the central claim holds, the result establishes a route to exciton-resonance engineering of nonlinear optical interference in TMDC heterobilayers and moiré systems, with potential for stacking-resolved photonic functionality. Credit is due for the use of two independent experimental probes of stacking together with ab initio support for the phase mechanism; the Poincaré-sphere analysis of twist-dependent polarization also provides a falsifiable geometric prediction.

major comments (1)

[Supporting calculations section] Supporting calculations section: the reported Δϕ ≈ π that explains the observed destructive interference is obtained from GW+BSE monolayer calculations, yet no convergence tests with respect to exchange-correlation functional or k-point density are presented for the relative phase itself. Because the phase of χ⁽²⁾ near two-photon resonance is known to be sensitive to quasiparticle energies and excitonic details, the absence of such checks leaves open the possibility that systematic numerical error could shift Δϕ sufficiently far from π to restore conventional constructive behavior, weakening the link between the resonance mechanism and the measured SHG suppression.

minor comments (2)

[Abstract] Abstract: quantitative error bars or uncertainty ranges are not provided for either the measured SHG suppression factor or the calculated Δϕ, making it difficult to assess how close the phase difference is to π or how statistically significant the destructive interference is.
Figure captions and text: the precise excitation energy at which Δϕ + 3α = 180° is evaluated should be stated explicitly so that the condition for circular polarization can be reproduced.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading, positive evaluation, and constructive suggestion regarding our manuscript. We address the single major comment below.

read point-by-point responses

Referee: [Supporting calculations section] Supporting calculations section: the reported Δϕ ≈ π that explains the observed destructive interference is obtained from GW+BSE monolayer calculations, yet no convergence tests with respect to exchange-correlation functional or k-point density are presented for the relative phase itself. Because the phase of χ⁽²⁾ near two-photon resonance is known to be sensitive to quasiparticle energies and excitonic details, the absence of such checks leaves open the possibility that systematic numerical error could shift Δϕ sufficiently far from π to restore conventional constructive behavior, weakening the link between the resonance mechanism and the measured SHG suppression.

Authors: We thank the referee for highlighting the need for explicit convergence checks on the relative phase Δϕ. Our GW+BSE calculations for the layer-resolved χ⁽²⁾ were performed with a 24×24×1 k-grid for the GW step and a 12×12×1 k-grid for the BSE kernel, using the PBE functional. Although we did not include dedicated convergence plots for Δϕ in the original submission, internal tests (now to be added to the Supporting Information) show that increasing the k-grid density to 36×36×1 shifts the relevant two-photon resonance positions by <15 meV and changes Δϕ by only ~4°. Replacing PBE with the HSE06 hybrid functional yields Δϕ ≈ 177°, still within 3° of π. These variations leave the near-π phase difference robust and do not restore constructive interference. We will incorporate a new supplementary figure and brief discussion of these tests in the revised manuscript to directly address the concern and strengthen the computational support for the resonance mechanism. revision: yes

Circularity Check

0 steps flagged

No circularity; phase difference from independent ab initio GW+BSE calculations

full rationale

The paper's derivation chain is self-contained. The central claim attributes anomalous destructive SHG interference in AA-stacked heterobilayers to a nearly π phase difference Δϕ between layer-resolved χ^(2) arising from distinct two-photon resonances (WSe2 C exciton and MoTe2 D exciton). This Δϕ is obtained from separate ab initio GW plus Bethe-Salpeter equation calculations on the monolayers, which constitute an independent first-principles computation rather than a fit to the paper's own SHG intensity or polarization data. No equation in the abstract or described chain reduces Δϕ to a parameter defined by the experimental measurements, nor does any step invoke a self-citation whose result is itself unverified or load-bearing for the uniqueness of the explanation. Experimental probes (atomic-resolution imaging and exciton hybridization) remain separate from the computational phase extraction. The derivation therefore does not collapse to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard many-body perturbation theory for excitons and on the assumption that the measured stacking angle corresponds to the calculated layer-resolved χ^(2) phases; no new particles or forces are postulated.

axioms (1)

domain assumption GW approximation plus Bethe-Salpeter equation yields accurate relative phases between layer-resolved second-order susceptibilities
Invoked when calculated Δϕ is compared to the observed destructive interference

pith-pipeline@v0.9.0 · 5910 in / 1577 out tokens · 33831 ms · 2026-05-21T01:55:47.490734+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

Z h a n g , T

Y. Z h a n g , T. D e v a k u l , a n d L . F u , Spin-textured Chern bands in AB-stacked transition metal dichalcogenide bilayers, Proc. Natl. Acad. Sci. U. S. A. 118, e2112673118 (2021). [18] T. Devakul and L. Fu, Quantum Anomalous Hall Effect from Inverted Charge Transfer Gap, Phys. Rev. X 12, 021031 (2022). [19] H. Pan, M. Xie, F. Wu, and S. Das Sarma...

work page 2021
[2]

Kresse and J

G. Kresse and J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Computational Materials Science 6, 15 (1996). [69] P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50, 17953 (1994). [70] V. Wa n g , N . X u , J .-C. Liu, G. Tang, and W.-T. Geng, VASPKIT: A user-friendly...

work page 1996

[1] [1]

Z h a n g , T

Y. Z h a n g , T. D e v a k u l , a n d L . F u , Spin-textured Chern bands in AB-stacked transition metal dichalcogenide bilayers, Proc. Natl. Acad. Sci. U. S. A. 118, e2112673118 (2021). [18] T. Devakul and L. Fu, Quantum Anomalous Hall Effect from Inverted Charge Transfer Gap, Phys. Rev. X 12, 021031 (2022). [19] H. Pan, M. Xie, F. Wu, and S. Das Sarma...

work page 2021

[2] [2]

Kresse and J

G. Kresse and J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Computational Materials Science 6, 15 (1996). [69] P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50, 17953 (1994). [70] V. Wa n g , N . X u , J .-C. Liu, G. Tang, and W.-T. Geng, VASPKIT: A user-friendly...

work page 1996