Pith. sign in

REVIEW 3 major objections 4 minor 11 references

Unveiling intrinsic bulk photovoltaic effect in atomically thin ReS2

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper reports that atomically thin ReS2 flakes with graphite contacts and hBN encapsulation produce a zero-bias, polarization-dependent photocurrent in the pristine channel, which the authors attribute to the intrinsic bulk…

desk verdict A clean lateral ReS2 platform that likely shows a bulk photovoltaic signal, but the paper's own numbers contradict its claimed theory-experiment agreement and the non-centrosymmetric stacking is never verified. read the letter →

arxiv 2412.13863 v1 pith:OEJFICLP submitted 2024-12-18 cond-mat.mes-hall cond-mat.mtrl-sci

Maria Ramos , Tanweer Ahmed , Bao Q. Tu , Eleni Chatzikyriakou , Lucía Olano-Vegas , Beatriz Martín-García , M. Reyes Calvo , Stepan S. Tsirkin
show 5 more authors
Ivo Souza Félix Casanova Fernando de Juan Marco Gobbi Luis E. Hueso
This is my paper · ORCID
classification cond-mat.mes-hallcond-mat.mtrl-sci
keywords bulkphotovoltaiceffectintrinsicphotocurrentsecond-orderconductivityshiftcurrentReS2brokeninversionsymmetryvanderWaalsheterostructuresscanningmicroscopy
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 reports that atomically thin flakes of the semiconductor ReS2, connected to graphite electrodes and encapsulated in hexagonal boron nitride, produce a photocurrent at zero applied voltage when illuminated with visible light. The signal is measured in the pristine middle of the channel, away from the contacts, and its magnitude depends on the linear polarization of the light, which is the characteristic fingerprint of the bulk photovoltaic effect. The authors extract the in-plane second-order conductivity tensor from the polarization dependence and compare it with ab initio shift-current calculations for a non-centrosymmetric bilayer stacking, finding the same order of magnitude. If the claim holds, a single unstrained few-layer van der Waals material can convert light into current through a nonlinear symmetry-driven mechanism, and the lateral device geometry provides a clean way to separate that intrinsic response from interface photocurrents.

What carries the argument

The load-bearing object is the second-order conductivity tensor $\sigma^{(2)}_{ijk}$, defined by $j_i=\sigma^{(2)}_{ijk}E_jE_k$, which is nonzero only when inversion symmetry is broken. In the experiment only the in-plane components along the b-axis matter, so the polarization-angle fit uses $I_x = t r [\sigma^{(2)}_{xxx}\cos^2\theta + \sigma^{(2)}_{xyy}\sin^2\theta + 2\sigma^{(2)}_{xxy}\cos\theta\sin\theta]$, with $t$ the flake thickness and $r$ the beam radius. The same tensor is computed from first principles for the non-centrosymmetric bilayer stacking in which the top monolayer is rotated 180 degrees relative to the bottom one. Microscopically, the calculation evaluates the shift-current contribution, where the coherent displacement of photoexcited carriers produces the DC current. The device architecture, graphite electrodes plus hBN encapsulation in a lateral geometry, is what lets the tensor be measured in the pristine channel with a roughly one-micron laser spot, separating it from the large Schottky photocurrents that dominate metal-contacted devices.

What would settle it

A scanning second-harmonic generation map of the exact devices from Figs. 2 and 3 that shows no signal in the channel would remove the symmetry basis for the intrinsic-bulk-photovoltaic assignment; conversely, measuring a comparable channel photocurrent in a flake verified to be centrosymmetric would indicate that the current has a different origin.

Watch

Extended reading notes

Core claim

The central claim is that a lateral ReS2 device with graphite contacts and hBN encapsulation shows an intrinsic bulk photovoltaic effect: at zero bias, focused illumination of the channel center produces a short-circuit current along the b-axis that grows linearly with optical power and is modulated by, but never extinguished as, the linear polarization is rotated. From the angular data the paper fits the second-order conductivity components $\sigma_{xxx}^{(2)}=5.51\,\mu\mathrm{A}/\mathrm{V}^2$, $\sigma_{xyy}^{(2)}=3.06\,\mu\mathrm{A}/\mathrm{V}^2$, and $\sigma_{xxy}^{(2)}=-0.05\,\mu\mathrm{A}/\mathrm{V}^2$, and reports intrinsic responsivities of about 1.3 mA/W for a bilayer device and 1.6 mA/W at 633 nm and 1.0 mA/W at 532 nm for a four-layer device. Ab initio shift-current calculations for the 180-degree-rotated non-centrosymmetric bilayer give $\sigma^{(2)}$ components of the same order of magnitude and predict the photocurrent maximum within $[-12^\circ,+2^\circ]$ of the b-axis. The paper concludes that the channel photocurrent is an intrinsic bulk photovoltaic response, not a Schottky-barrier or edge effect.

Load-bearing premise

The claim rests on the measured flakes having the non-centrosymmetric 180-degree-rotated bilayer stacking used in the calculation; the paper does not directly verify this stacking in the devices it measures, and a centrosymmetric flake would not produce the claimed intrinsic bulk photovoltaic effect.

Editorial extensions

If this is right

  • Zero-bias bulk photovoltaic currents can be observed in unstrained, few-layer ReS2 without vertical device stacks or external tuning knobs.
  • The order-of-magnitude agreement between measured and calculated $\sigma^{(2)}$ supports assigning the channel photocurrent to the non-centrosymmetric bilayer stacking rather than to interface effects.
  • Because the channel photocurrent is nonzero at every polarization angle, the response is a direct electrical signature of the polar character of few-layer ReS2, matching the earlier observation of nonzero second-harmonic generation for all angles.
  • Graphite contacts suppress the interface photocurrent relative to Ti/Au contacts, making the intrinsic channel response visible and cleanly separable in scanning photocurrent maps.
  • The same fabrication and measurement protocol should apply to other non-centrosymmetric van der Waals materials without a perpendicular two-fold rotation axis, where in-plane bulk photovoltaic currents are generically allowed at normal incidence.

Reading between the lines

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

  • A decisive check would be to verify the stacking polarity of the measured devices by second-harmonic generation or ferroelectric switching, since the paper does not directly confirm the non-centrosymmetric stacking in the devices it reports.
  • The fitted tensor components give a polarization-dependent angular profile that could serve as an in-situ crystallographic alignment tool, because the photocurrent maximum encodes the b-axis orientation.
  • The same measurement on flakes of different thickness and stacking polytypes would map how the bulk photovoltaic response evolves with symmetry, connecting the bilayer calculation to the four-layer result.
  • If the observed signal instead came from a buried interface or strain at the hBN boundaries, devices with different encapsulation materials should show a different magnitude, a testable extension the paper does not report.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The manuscript reports fabrication of lateral, hBN-encapsulated ReS2 devices with graphite contacts, in which a zero-bias photocurrent is observed in the pristine channel under 633 nm and 532 nm illumination. The authors attribute this signal to the intrinsic bulk photovoltaic effect (BPVE) on the basis of linear power scaling, polarization dependence described by a second-order conductivity tensor, and scanning photocurrent microscopy showing channel-separated photocurrent. Ab initio shift-current calculations for a non-centrosymmetric (A', 180°-rotated) bilayer ReS2 give nonlinear conductivities of the same order of magnitude as the experimental fit, and the paper claims the calculated maximum-photocurrent angle lies within [-12°, +2°] of the b-axis.

Significance. If the interpretation holds, the work would be a valuable demonstration of an intrinsic BPVE in a lateral vdW device without strain or vertical geometry, with a clean contact scheme and an independent first-principles calculation. The paper includes detailed fabrication and characterization, and the shift-current calculation is independent of the measured data, which is a strength. However, the central claim depends on the measured flakes adopting the non-centrosymmetric bilayer stacking assumed in the calculation, and on the theory reproducing the polarization dependence; both assumptions currently need attention.

major comments (3)
  1. [Main text, 'While bulk ReS2 is centrosymmetric' paragraph; SI Methods §1] The non-centrosymmetric A′ stacking is assumed for the measured flakes, but no SHG, ferroelectric switching, or structural measurement is provided for the actual devices. Since the bulk parent crystal is centrosymmetric, exfoliation need not yield the A′ polytype, and the cited Refs. 38 and 39 characterized material that was specifically selected for those properties. Without direct verification that the measured bilayer and four-layer flakes are the A′ stacking, the assignment of the channel photocurrent to the intrinsic BPVE of that structure is not fully justified.
  2. [Main text, Fig. 4 and surrounding paragraph] Using the reported 633 nm tensor components (σ_xxx = 2.68, σ_xyy = −1.03, σ_xxy = −3.71 µA/V²) in the paper's own Eq. (3) gives θ0 = 0.5 arctan(2σ_xxy/(σ_xxx − σ_xyy)) ≈ −32°, which is outside the stated range [-12°, +2°] for the angle of maximum photocurrent. Thus the ab initio calculation does not reproduce the measured polarization response, whose maximum lies near the b-axis. The internal inconsistency in the claimed theory–experiment agreement needs to be resolved, either by correcting the tensor components or by re-evaluating the comparison.
  3. [Main text, Fig. 3 and thickness determination (SI §2)] The ab initio calculation is performed for a bilayer, but the SPCM device (device B) is a four-layer flake. The text does not explain how the A′ stacking and the calculated shift-current tensors for the bilayer apply to a four-layer flake, which may have different stacking energetics. The comparison between the two devices and the theory should be made explicit.
minor comments (4)
  1. [Fig. 2d caption] The figure caption states an optical power of 120 mW, but the text and SI §4 consistently use 120 µW; the units should be corrected.
  2. [Main text near Eq. (3)] The fitted conductivity values are given in mA/V² in the main text, while the SI reports the same quantities in µA/V²; the units should be made consistent.
  3. [Main text, intrinsic responsivity definition] The definition κ = <I_SC>/(r×t) divided by P/(π r²) uses an effective area r×t rather than the beam area; the geometric model should be clarified, and the sensitivity of κ to the assumed beam radius and refractive index should be stated.
  4. [Equation (4)] Equation (4) introduces the interband dipole matrix and generalized derivative notation without explicit definitions; a brief explanation of the symbols would improve readability for non-specialists.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the ab initio shift-current calculation is independent of the measured photocurrent, and the only notable weakness is an unverified sample-structure assumption, which is not a circularity.

full rationale

The paper's central comparison is between (i) experimental nonlinear-conductivity components obtained by fitting the polarization-dependent photocurrent in Fig. 2d to Eq. 3 and (ii) ab initio shift-current components computed from a Wannier-interpolated band structure of a non-centrosymmetric bilayer. The theoretical calculation is not fitted to the experiment: no experimental conductivity values, responsivities, or polarization angles enter Eq. 4 or the Wannier calculation. The two sets of sigma values are derived from distinct inputs (measured photocurrent vs. first-principles electronic structure) and are compared after the fact, so the 'same order of magnitude' claim is a genuine test rather than a tautology. The A' stacking assumption for the measured flakes is not directly verified by SHG or structural measurements, but it is introduced from prior external literature (Refs. 38, 39, 48) and is an assumption about the sample's structure, not a definition that makes the theoretical output equal to the experimental input. No load-bearing self-citation chain is used; Ref. 49 is a methodological citation to the authors' Wannier-interpolation code, which is independent and parameter-free with respect to the measured data. A possible internal inconsistency in the reported theoretical theta0 at 633 nm is a correctness concern, not a circularity: plugging the reported components into the paper's own Eq. 3 gives theta0 ~ -32 degrees, outside the stated [-12, +2] degree range, but this does not make the derivation equivalent to its inputs.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central quantitative claims rest on two fitted experimental conductivity components, an assumed beam radius and refractive index, and an unverified structural assumption about the ReS2 stacking. No new physical entities are introduced. The ab initio calculation uses standard software and parameters, but the structural input is not experimentally confirmed in the measured devices.

free parameters (5)
  • sigma_xxx experimental fit = 5.51 +/- 0.07 microA/V2 (main text says mA/V2)
    Fitted from polarization-dependent photocurrent using Eq. 3; the main quantitative output of the experiment.
  • sigma_xyy experimental fit = 3.06 +/- 0.07 microA/V2
    Fitted from the same polarization data; used in the reported nonlinear conductivity tensor.
  • sigma_xxy experimental fit = -0.05 +/- 0.06 microA/V2
    Fitted from the same polarization data; near zero in the experiment, in contrast to the ab initio value.
  • beam radius r = 0.5 micrometers
    Assumed for converting current to conductivity and responsivity; uncertainty is not propagated.
  • refractive index n = 4.6 at 633 nm
    Taken from a refractive index database; appears in Eq. 3 and the SI fitting equation, with no uncertainty estimate.
assumptions (4)
  • domain assumption Few-layer ReS2 flakes in the measured devices adopt a non-centrosymmetric A' (180-degree rotated) stacking.
    Invoked in the main text and SI Methods; not directly verified in the measured devices by SHG or ferroelectric switching, and it is the structural basis for the intrinsic BPVE interpretation.
  • domain assumption The channel photocurrent is purely second-order BPVE with no photothermoelectric or bolometric contribution.
    Used to interpret the linear power dependence and polarization dependence; no direct thermal-gradient or time-constant check is reported, and the SPCM spatial uniformity is indirect evidence.
  • domain assumption Graphite contacts do not contribute to the central channel signal, and interface signals are cleanly separated in space.
    Used to classify the central signal as intrinsic; SPCM shows opposite-sign interface currents at the contacts, but potential tails of the interface fields could extend into the channel.
  • standard math Standard DFT and Wannier interpolation accurately capture the shift-current conductivity in ReS2.
    The calculation uses Quantum ESPRESSO, Wannier90, and WannierBerri with PBE, DFT-D3, eta = 0.04 eV, and 0.1 eV broadening; these are standard methods but approximate exchange-correlation can affect band gap and conductivity magnitude.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Unveiling intrinsic bulk photovoltaic effect in atomically thin ReS2." pith.science (2026). https://pith.science/paper/OEJFICLP

@misc{pith2026241213863,
  author       = {Pith},
  title        = {Pith review of: Unveiling intrinsic bulk photovoltaic effect in atomically thin ReS2},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OEJFICLP}},
  note         = {Machine review of arXiv:2412.13863}
}
read the original abstract

The bulk photovoltaic effect (BPVE) offers a promising avenue to surpass the efficiency limitations of current solar cell technology. However, disentangling intrinsic and extrinsic contributions to photocurrent remains a significant challenge. Here, we fabricate high-quality, lateral devices based on atomically thin ReS2 with minimal contact resistance, providing an optimal platform for distinguishing intrinsic bulk photovoltaic signals from other extrinsic photocurrent contributions originating from interfacial effects. Our devices exhibit large bulk photovoltaic performance with intrinsic responsivities of 1 mA/W in the visible range, without the need for external tuning knobs such as strain engineering. Our experimental findings are supported by theoretical calculations. Furthermore, our approach can be extrapolated to investigate the intrinsic BPVE in other non-centrosymmetric van der Waals materials, paving the way for a new generation of efficient light-harvesting devices.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

  1. [1]

    Commercially available ReS2 and hBN crystals were obtained from HQ Graphene, while the graphite crystals were sourced from NGS Naturgraphit GmbH

    Methods Device fabrication. Commercially available ReS2 and hBN crystals were obtained from HQ Graphene, while the graphite crystals were sourced from NGS Naturgraphit GmbH. Each type of crystal was separately exfoliated on freshly cleaned SiO2/Si substrates using the mechanical cleavage technique. ReS2 flakes with a thickness of 2-10 layers and graphite ...

    work page

  2. [2]

    Thickness estimation of ReS2 active layers µ-Raman spectroscopy µ-Raman spectroscopy was employed to accurately characterize the thickness of the atomically thin ReS2 flakes composing our two devices exhibiting intrinsic BPVE. Previous studies have established that Raman spectroscopy can effectively distinguish the thickness of ReS2, particularly for very...

    work page

  3. [3]

    Output curve of a representative ReS2 device with graphite contacts under dark conditions

    ISD – VSD characteristics up to 1 V Figure S6. Output curve of a representative ReS2 device with graphite contacts under dark conditions. -1-0.50 0.51 VSD (V) -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 ISD (µA) 31

    work page

  4. [4]

    )𝐸)𝐸* (S1) Here, 𝜎$)*(

    Estimation of nonlinear conductivity from polarization-dependent photocurrent measurements Non-centrosymmetric materials can exhibit a second-order photocurrent response. Under linearly polarized light, a photocurrent density 𝑗+ along the i direction can be generated: 𝑗$%&'(=𝜎$)*(")𝐸)𝐸* (S1) Here, 𝜎$)*(") is the nonlinear conductivity, and 𝐸), 𝐸* are the ...

    work page

  5. [5]

    Z.; Haider, Z.; Iqbal, M

    Aftab, S.; Iqbal, M. Z.; Haider, Z.; Iqbal, M. W.; Nazir, G.; Shehzad, M. A. Bulk Photovoltaic Effect in 2D Materials for Solar-Power Harvesting. Adv. Opt. Mater. 2022, 10, 2201288. 6. Morimoto, T.; Nagaosa, N. Topological nature of nonlinear optical effects in solids. Science Adv. 2016, 2, e1501524. 7. Cook, A. M.; Fregoso, B. M.; De Juan, F.; Coh, S.; M...

    work page 2022

  6. [10]

    (a) – (c) Scanning photocurrent maps at 532-nm wavelength of a ReS2 flake with different few-layer thicknesses and contacted to Ti/Au electrodes

    Scanning photocurrent microscopy of a ReS2 device with Ti/Au contacts Figure S7. (a) – (c) Scanning photocurrent maps at 532-nm wavelength of a ReS2 flake with different few-layer thicknesses and contacted to Ti/Au electrodes. The power used for all SPCM maps is 400 µW. The light polarization is indicated at the top-left corner of each map. (d) Longitudin...

    work page 2014

  7. [12]

    An updated version of wannier90: A tool for obtaining maximally-localised Wannier functions. A. A. Mostofi, et al. 8, 2014, Computer Physics Communications, Vol. 185, pp. 2309-2310. 13. High performance Wannier interpolation of Berry curvature and related quantities with WannierBerri code. Tsirkin, S.S. 1, 2021, npj Computational Materials, Vol. 7, p. 33....

    work page 2014

  8. [17]

    Increasing bulk photovoltaic current by strain tuning

    Nadupalli, S.; Kreisel, J.; Granzow, T. Increasing bulk photovoltaic current by strain tuning. Science Adv. 2019, 5, eaau9199. 18. Ai, H.; Kong, Y.; Liu, D.; Li, F.; Geng, J.; Wang, S.; Lo, K. H.; Pan, H. 1T‴ transition-metal dichalcogenides: strong bulk photovoltaic effect for enhanced solar-power harvesting. J. Phys. Chem. C 2020, 124, 11221. 19. Schank...

    work page 2019

Show all 11 references
  1. [28]

    Giant intrinsic photoresponse in pristine graphene

    Ma, Q.; Lui, C.H.; Song, J.C.; Lin, Y.; Kong, J.F.; Cao, Y.; Dinh, T.H.; Nair, N.L.; Fang, W.; Watanabe, K; Taniguchi, T. Giant intrinsic photoresponse in pristine graphene. Nat. Nanotechnol. 2019, 14, 145. 29. Yang, D.; Wu, J.; Zhou, B. T.; Liang, J.; Ideue, T.; Siu, T.; Awan...

    2019 arXiv

  2. [39]

    Room-temperature ferroelectricity in 1T′-ReS2 multilayers

    Wan, Y.; Hu, T.; Mao, X.; Fu, J.; Yuan, K.; Song, Y.; Gan, X.; Xu, X.; Xue, M.; Cheng, X.; Huang, C. Room-temperature ferroelectricity in 1T′-ReS2 multilayers. Phys. Rev. Lett. 2022, 128, 067601. 40. Zhou, Y.; Zhou, X.; Yu, X. L.; Liang, Z.; Zhao, X.; Wang, T.; Miao, J.; Chen,...

    2022

  3. [49]

    S.; Souza, I

    Ibañez-Azpiroz, J.; Tsirkin, S. S.; Souza, I. Ab initio calculation of the shift photocurrent by Wannier interpolation. Phys. Rev. B 2018, 97, 245143. 20 Supporting Information Contents 1. Methods 2. Thickness estimation of ReS2 active layers 3. ISD – VSD characteristics up to...

    2018

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.