First principles study of chalcogen vacancy effect on the optoelectronic and photocatalytic properties of transition metal dichalcogenides monolayers

Antonio Cammarata; Tahir Wahab; Tomas Polcar

REVIEW 2 major objections 6 minor 18 references

First principles study of chalcogen vacancy effect on the optoelectronic and photocatalytic properties of transition metal dichalcogenides monolayers

T0 review · 2 major / 6 minor · reviewed 2026-07-12 · grok-4.5

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

Pith's one-line read Chalcogen vacancies turn TMDC monolayers into better photocatalysts that should not be discarded.

desk verdict Solid DFT survey of hull-minimum chalcogen vacancies in four MX2 monolayers; the electrostatic-asymmetry and band-edge numbers are useful, but the photocatalytic “capability” claim rests only on vacuum-level alignment. read the letter →

arxiv 2607.02776 v1 pith:KYHJOCSG submitted 2026-07-02 cond-mat.mtrl-sci

Tahir Wahab , Antonio Cammarata , Tomas Polcar This is my paper · ORCID

classification cond-mat.mtrl-sci

keywords transitionmetaldichalcogenideschalcogenvacanciesphotocatalysishydrogenevolutionCO2reductionHSE06bandalignmentworkfunctionasymmetry

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

Chalcogen vacancies form spontaneously in working TMDC monolayers. First-principles calculations on MoS2, MoSe2, WS2 and WSe2 show that the vacancy concentrations that sit at the energy convex hull leave the lattices dynamically and thermally stable, insert mid-gap states that shrink the band gap, and break electrostatic symmetry so that a built-in field appears. Every defective monolayer can drive the hydrogen-evolution half-reaction; WS2 and WSe2 can also reduce CO2 all the way to methane and methanol. The authors therefore argue that vacancy-containing sheets are not defective waste but useful, vacancy-engineered photocatalysts.

What carries the argument

Convex-hull vacancy selection combined with HSE06 band-edge alignment against tabulated water and CO2 redox potentials; the resulting mid-gap states and work-function asymmetry (δφ) are the objects that convert an otherwise inactive basal plane into a photocatalyst.

What would settle it

A controlled synthesis of WS2 or WSe2 monolayers at the predicted vacancy concentration that fails to produce measurable HER or CO2-reduction products under visible light, or a measurement showing that the actual band-edge positions lie below the relevant redox potentials once solvation is included.

Watch

Extended reading notes

Core claim

At the convex-hull vacancy concentrations (6.25 % for MoS2, MoSe2 and WSe2; 1.56 % for WS2), the monolayers remain stable, acquire mid-gap states that lower the gap, develop an electrostatic asymmetry that separates charge, and thereby become capable of HER (all four) and of multi-electron CO2 reduction (WS2 and WSe2), outperforming their pristine counterparts.

Load-bearing premise

That vacuum-level HSE06 band edges alone, without solvation, pH corrections or kinetic barriers, are enough to declare a material capable of HER or selective CO2 reduction.

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.

Desk Editor's Note

read the letter

The new pieces here are the ATAT convex-hull concentrations (6.25 % for MoS2/MoSe2/WSe2, 1.56 % for WS2), the HSE06 band edges and work-function asymmetries at those concentrations, and the side-by-side overlay of those edges against both water redox and the five CO2 half-reactions. That package is cleanly executed and was not already in the literature they cite.

What they do well: phonon dispersions plus 8 ps AIMD at 300 K confirm dynamical and thermal stability; the projected bands correctly show metal-d character at the gap edges; the planar-averaged potential and the 20–90 meV work-function difference between the vacancy and pristine faces give a concrete, quantitative argument for built-in charge separation that is absent in the pristine sheets. Optical absorption shows the expected red-shift. All of that is standard, carefully documented DFT and is reproducible from the geometries they supply.

The soft spot is exactly the one the stress-test flags, and it is real but not fatal. They declare the defective monolayers “capable” of HER (all four) and of full CO2 reduction (WS2/WSe2) solely because the HSE06 CBM/VBM straddle the tabulated vacuum-level potentials. No adsorption free energies, no solvation or pH corrections, no exciton binding (0.5–1 eV in these 2-D systems), and no kinetic barriers are computed. That is the conventional first filter, not a demonstration of photocatalytic activity. The stronger phrasing in the abstract and conclusion (“own photocatalytic capabilities… should not be regarded as flawed products”) therefore over-reaches the evidence they actually present. The rest of the paper does not depend on that over-claim; the numbers themselves remain useful design data.

Citation pattern is normal for the sub-field; self-cites are to their own prior pristine-monolayer work and are not load-bearing. No circular fitting.

This is for people already working on TMDC photocatalysis or vacancy engineering who need concrete concentration targets and electrostatic numbers. It is not a conceptual breakthrough, but it is a competent, self-contained computational survey that a serious referee can evaluate and improve. I would send it out.

Referee Report

2 major / 6 minor

Summary. The manuscript reports a first-principles (PBE + HSE06, DFT-D2) study of chalcogen vacancies in MX2 monolayers (M = Mo, W; X = S, Se). Vacancy concentrations that minimise the formation-energy convex hull (6.25 % for MoS2, MoSe2, WSe2; 1.56 % for WS2) are identified with ATAT, and the resulting defective cells are shown to be dynamically and thermally stable by phonon dispersions and 8 ps AIMD at 300 K. Vacancies introduce mid-gap states that reduce the HSE06 band gaps (to 1.25–1.85 eV), produce an electrostatic-potential asymmetry and a work-function difference δφ (19–89 meV) that is argued to aid charge separation, and red-shift the optical absorption onset of the Mo-based systems. Band-edge positions relative to vacuum-level water and CO2 redox potentials are then used to conclude that all four defective monolayers can drive HER (but require a co-catalyst for OER) and that WS2 and WSe2 can reduce CO2 to the full suite of C1 products. The authors therefore argue that vacancy-containing TMD monolayers possess photocatalytic utility and should not be discarded as defective material.

Significance. If the thermodynamic band-edge criterion is accepted as sufficient, the work supplies a concrete, concentration-selected set of defective MX2 models whose stability, electronic structure and optical response have been carefully mapped. The systematic comparison of four chemically related monolayers, the use of the convex-hull minimum rather than an arbitrary vacancy density, and the explicit demonstration of a built-in potential asymmetry are useful additions to the defect-engineering literature on 2D TMDCs. The results can serve as a computational starting point for experimental vacancy-engineering efforts aimed at photocatalysis or optoelectronics. The principal limitation is that the photocatalytic claims rest solely on vacuum-level alignment; no free-energy barriers, solvation or excitonic corrections are provided, so the practical significance of the “capable” designation remains provisional.

major comments (2)

Section 3 and Figure 11: the central claim that the defective monolayers “are capable of performing HER” and that WS2/WSe2 “can serve as an efficient photocatalytic material for reducing CO2 into useful chemical products” is drawn exclusively from HSE06 vacuum-level CBM/VBM positions relative to tabulated redox potentials (Eqs. 2–6). No adsorption free energies, solvation or pH corrections, excitonic binding energies, or kinetic overpotentials are computed. In 2D TMDCs exciton binding energies of 0.5–1 eV are typical and can reverse the thermodynamic driving force for multi-electron CO2 pathways. The language of “capability” and “efficient” therefore over-states a necessary but far from sufficient criterion; the conclusions should be re-phrased as thermodynamic alignment only, or the missing free-energy calculations should be supplied.
Section 3 (paragraph comparing to pristine monolayers) and the Abstract: the assertion that vacancy-containing sheets “own photocatalytic capabilities compared to the pristine counterparts” is not quantitatively supported. The authors’ own prior work (Ref. 61) already showed that the pristine monolayers straddle the water redox potentials. The present manuscript does not recompute the pristine band edges on the same HSE06 footing or provide a side-by-side table of driving forces, so the claimed improvement remains qualitative. A direct numerical comparison of pristine versus defective CBM/VBM positions (and of δφ) is required to substantiate the “compared to” claim.

minor comments (6)

Abstract and Introduction: “M-X bond lengths about a vacancy are different compared to the bond lengths in the pristine structure” is contradicted by Table 1, which shows changes of only 0.00–0.02 Å (within the reported standard deviations). Soften the wording to “essentially unchanged” or “only marginally altered”.
Section 2: the statement that the lattice parameters “are large enough … to justify the use of only the Γ-point” should be accompanied by a short k-point convergence test (or a reference to one) for the largest cell, especially for the HSE06 band edges that underpin the photocatalytic conclusions.
Figure 3 caption and text: the small imaginary pocket near Γ is correctly attributed to a known 2D numerical artefact, but a brief remark on the magnitude (or a zoomed inset) would help the reader judge that it is negligible.
Section 3, CO2 discussion: the potentials listed for the half-reactions at pH = 7 are given without an explicit reference to the absolute vacuum scale conversion used; a short sentence or citation clarifying the alignment convention would improve reproducibility.
Typographical inconsistencies: “absortion” (twice), “stuctures”, “assite”, “WSe2 it is 1.56 %” (should be WS2), and occasional missing spaces after commas. A careful proof-read is needed.
Supplementary Information: the POSCAR-style geometries are welcome, but a short table of lattice constants and vacancy formation energies relative to the convex hull would make the SI more self-contained.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: first-principles outputs (hull concentrations, HSE06 edges, electrostatic asymmetry) are independent of the photocatalytic conclusions drawn from them.

full rationale

The derivation chain is self-contained and non-circular. Vacancy concentrations are obtained from ATAT convex-hull enumeration of formation energies; geometries are relaxed under stated force/energy tolerances; dynamical and thermal stability are checked by phonon dispersions and AIMD; electronic structure (PBE/HSE06 band structures, planar-average electrostatic potential, work functions, optical absorption) is computed from those geometries; photocatalytic capability is then assessed by comparing the resulting HSE06 CBM/VBM positions to tabulated water and CO2 redox potentials. None of these quantities is fitted to the HER/CO2 performance later claimed, nor is any central premise justified solely by a self-citation that itself encodes the target result. The sole self-citation (Ref. 61) supplies only the pristine reference geometries and band-edge baselines for comparison; it does not force the defective-structure conclusions. The paper’s strongest claim therefore rests on independent first-principles outputs, not on a definitional or fitted reduction. (Whether band-edge alignment alone is a sufficient criterion for photocatalytic capability is a separate correctness/overclaim issue, not circularity.)

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

The central claims rest on standard DFT approximations and on the conventional vacuum-level alignment criterion for photocatalysis; no new free parameters are fitted to the target photocatalytic metrics and no new physical entities are postulated.

assumptions (4)

domain assumption PBE and HSE06 exchange-correlation functionals yield band gaps and band-edge positions accurate enough for qualitative photocatalytic ranking of TMDCs.
Invoked throughout Section 3 and for all band-structure and electrostatic-potential results; known systematic errors of both functionals for 2-D materials are not quantified.
domain assumption The vacancy concentration that minimizes the ATAT convex hull is the relevant spontaneous concentration under working conditions.
Used in Section 2 to select the 6.25 % / 1.56 % models that are subsequently analyzed; experimental vacancy densities can differ under non-equilibrium growth.
domain assumption Thermodynamic alignment of CBM/VBM with tabulated vacuum-scale redox potentials is sufficient to declare HER or CO2-reduction capability.
Explicitly applied in the final paragraphs of Section 3 and Figure 11; kinetics, solvation and surface chemistry are omitted.
domain assumption Γ-point sampling is adequate once the supercell lattice parameters exceed ~12 Å.
Stated in Section 2; residual Brillouin-zone sampling error is not estimated.

how reviews work

0 comments

Cite this review

Pith. "Pith review of First principles study of chalcogen vacancy effect on the optoelectronic and photocatalytic properties of transition metal dichalcogenides monolayers." pith.science (2026). https://pith.science/paper/KYHJOCSG

@misc{pith2026260702776,
  author       = {Pith},
  title        = {Pith review of: First principles study of chalcogen vacancy effect on the optoelectronic and photocatalytic properties of transition metal dichalcogenides monolayers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KYHJOCSG}},
  note         = {Machine review of arXiv:2607.02776}
}

abstract

In working conditions, chalcogen vacancies spontaneously occur in two-dimensional transition metal dichalcogenides (TMDCs) monolayers, affecting their optoelectronic and photocatalytic properties. To study how chalcogen vacancies affect such properties, we use quantum mechanical calculations considering prototypical MX$_2$ (M = Mo, W, X = S and Se) TMDCs monolayers. Structural optimisations show that M-X bond lengths about a vacancy are different compared to the bond lengths in the pristine structure. Band structure calculations reveal that the introduction of vacancies produce electronic states about the Fermi level, hence resulting in the reduction of the band gap. Work function and electrostatic potential calculations show that the introduction of vacancies induce an asymmetry in the electrostatic potential facilitating the charge separation; such feature is absent in a pristine monolayer. All the considered defective structures are capable of performing hydrogen evolution reaction, while co-catalyst is required to perform oxygen evolution reaction when used for water splitting. WS$_2$ and WSe$_2$ defective monolayers can serve as an efficient photocatalytic material for reducing CO$_2$ into useful chemical products. The presented results show that vacancy-containing TMDCs monolayers own photocatalytic capabilities compared to the pristine counterparts, thus showing that defective TMD monolayers have prospective applications and should not be regarded as flawed products to be discarded. Finally, the results might constitute guidelines for the experimental synthesis of vacancy-engineered MX$_2$ monolayers for optoelectronic devices and photocatalytic applications.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references

[1]

Robotics and advanced industrial production

We know that the minimum band gap requirement for a semiconductor to split water is 1.23 eV, which is the difference of potential between HER and OER. As in case of CO 2 reduction, we can have many products (from formic acid to methane), thus the band gap requirement is product dependent. Ac- cording to Equations (2)–(6), the thermodynamic potential diffe...

2025
[2]

5 H. L. Zhuang and R. G. Hennig, Chemistry of Materials, 2013, 25, 3232–3238. 6 A. Wu, Y. Xie, H. Ma, C. Tian, Y. Gu, H. Yan, X. Zhang, G. Yang and H. Fu, Nano Energy, 2018, 44, 353–363. 7 S. Cao and L. Piao, Angewandte Chemie International Edition, 2020, 59, 18312–18320. 8 C.-F. Fu, X. Wu and J. Yang, Advanced materials, 2018, 30, 1802106. 9 A. Fujishima...

2013
[3]

Velempini, E

11 T. Velempini, E. Prabakaran and K. Pillay ,Materials To- day Chemistry, 2021, 19, 100380. 12 L. Jia, X. Tan, T. Yu and J. Ye,Energy & Fuels, 2022, 36, 11308–11322. 13 Q. Zhu, Q. Xu, M. Du, X. Zeng, G. Zhong, B. Qiu and J. Zhang, Advanced Materials, 2022, 34, 2202929. 14 S. Chandrasekaran, L. Yao, L. Deng, C. Bowen, Y. Zhang, S. Chen, Z. Lin, F. Peng an...

2021
[4]

Gattrell, N

25 M. Gattrell, N. Gupta and A. Co, Journal of electroana- lytical Chemistry, 2006, 594, 1–19. 26 M. D. Porosoff, B. Yan and J. G. Chen, Energy & Envi- ronmental Science, 2016, 9, 62–73. 27 Z. Weng, J. Jiang, Y. Wu, Z. Wu, X. Guo, K. L. Materna, W. Liu, V. S. Batista, G. W. Brudvig and H. Wang,Jour- nal of the American Chemical Society, 2016, 138, 8076–

2006
[5]

Nea¸ tu, J

28 ¸ S. Nea¸ tu, J. A. Maciá-Agulló and H. Garcia, Interna- tional journal of molecular sciences , 2014, 15, 5246–

2014
[6]

Kovacic, B

29 Z. Kovacic, B. Likozar and M. Hus, ACS catalysis, 2020, 10, 14984–15007. 30 S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angewandte Chemie International Edition , 2013, 52, 7372–7408. 31 K. Teramura, T. Tanaka, H. Ishikawa, Y. Kohno and T. Funabiki, The Journal of Physical Chemistry B, 2004, 108, 346–354. 32 S. Navalón, A. Dhakshinamoor...

2020
[7]

36 S. Wang, W. Yao, J. Lin, Z. Ding and X. Wang, Ange- wandte Chemie, 2014, 126, 1052–1056. 37 D. Sun, Y. Fu, W. Liu, L. Ye, D. Wang, L. Yang, X. Fu and Z. Li, Chemistry–A European Journal, 2013, 19, 14279– 14285. 38 Z. Li, N. H. Attanayake, J. L. Blackburn and E. M. Miller, Energy & Environmental Science, 2021, 14, 6242–6286. 39 M. Asadi, K. Kim, C. Liu,...

2014
[8]

Ai, J.-W

45 M. Ai, J.-W. Zhang, Y.-W. Wu, L. Pan, C. Shi and J.- J. Zou, Chemistry–An Asian Journal , 2020, 15, 3599–

2020

Show all 18 references

[9]

46 J. Hong, Z. Hu, M. Probert, K. Li, D. Lv, X. Yang, L. Gu, N. Mao, Q. Feng, L. Xie et al., Nature communications, 2015, 6,

2015
[10]

47 H. Lin, J. Wang, Q. Luo, H. Peng, C. Luo, R. Qi, R. Huang, J. Travas-Sejdic and C.-G. Duan, Journal of Alloys and Compounds, 2017, 699, 222–229. 48 M. Piao, J. Chu, X. Wang, Y. Chi, H. Zhang, C. Li, H. Shi and M.-K. Joo, Nanotechnology, 2018, 29, 025705. 12 | 1–14 + P V S O...

2017
[11]

52 H. Qiu, T. Xu, Z. Wang, W. Ren, H. Nan, Z. Ni, Q. Chen, S. Yuan, F. Miao, F. Songet al., Nature communications, 2013, 4,

2013
[12]

53 Y. Ji, J. K. Nørskov and K. Chan, The Journal of Physical Chemistry C, 2019, 123, 4256–4261. 54 J. Lee, S. Kang, K. Yim, K. Y. Kim, H. W. Jang, Y. Kang and S. Han, The Journal of Physical Chemistry Letters , 2018, 9, 2049–2055. 55 P. Vancsó, G. Z. Magda, J. Pet ˝o, J.-Y. No...

2019
[13]

64 J. P. Perdew, K. Burke and M. Ernzerhof,Physical review letters, 1996, 77,

1996
[14]

Connolly and A

65 J. Connolly and A. Williams, Physical Review B , 1983, 27,

1983
[15]

66 P. E. Blöchl, Physical review B, 1994, 50, 17953. 67 S. Grimme, Journal of computational chemistry , 2006, 27, 1787–1799. 68 J. Hafner, Journal of computational chemistry, 2008, 29, 2044–2078. 69 D. J. Evans and B. Holian, A)’=(A Is), 1985, 1,

1994
[16]

70 A. Togo, L. Chaput, T. Tadano and I. Tanaka,Journal of Physics: Condensed Matter, 2023, 35, 353001. 71 A. Togo, Journal of the Physical Society of Japan , 2023, 92, 012001. 72 J. P. Perdew, K. Burke and M. Ernzerhof,Phys. Rev. Lett., 1996, 77, 3865–3868. 73 J. Heyd, G. E. S...

2023
[17]

79 J. S. Jang, H. G. Kim and J. S. Lee, Catalysis today , 2012, 185, 270–277. 80 E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja, Chemical Society Reviews, 2009, 38, 89–99. 81 A. K. Singh, J. H. Montoya, J. M. Gregoire and K. A. Persson, Nature communications, 2019, 10,

2012
[18]

Momma and F

82 K. Momma and F. Izumi, Journal of Applied Crystallog- raphy, 2011, 44, 1272–1276. + P V S O B M / B N F < Z F B S > < W P M > 1–14 | 13 Fig. 11 CBM and VBM of all MX2 structure in comparison (a) water redox potential and (b) CO2 reduction potential 14 | 1–14 + P V S O B M /...

2011

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.

[1] [1]

Robotics and advanced industrial production

We know that the minimum band gap requirement for a semiconductor to split water is 1.23 eV, which is the difference of potential between HER and OER. As in case of CO 2 reduction, we can have many products (from formic acid to methane), thus the band gap requirement is product dependent. Ac- cording to Equations (2)–(6), the thermodynamic potential diffe...

2025

[2] [2]

5 H. L. Zhuang and R. G. Hennig, Chemistry of Materials, 2013, 25, 3232–3238. 6 A. Wu, Y. Xie, H. Ma, C. Tian, Y. Gu, H. Yan, X. Zhang, G. Yang and H. Fu, Nano Energy, 2018, 44, 353–363. 7 S. Cao and L. Piao, Angewandte Chemie International Edition, 2020, 59, 18312–18320. 8 C.-F. Fu, X. Wu and J. Yang, Advanced materials, 2018, 30, 1802106. 9 A. Fujishima...

2013

[3] [3]

Velempini, E

11 T. Velempini, E. Prabakaran and K. Pillay ,Materials To- day Chemistry, 2021, 19, 100380. 12 L. Jia, X. Tan, T. Yu and J. Ye,Energy & Fuels, 2022, 36, 11308–11322. 13 Q. Zhu, Q. Xu, M. Du, X. Zeng, G. Zhong, B. Qiu and J. Zhang, Advanced Materials, 2022, 34, 2202929. 14 S. Chandrasekaran, L. Yao, L. Deng, C. Bowen, Y. Zhang, S. Chen, Z. Lin, F. Peng an...

2021

[4] [4]

Gattrell, N

25 M. Gattrell, N. Gupta and A. Co, Journal of electroana- lytical Chemistry, 2006, 594, 1–19. 26 M. D. Porosoff, B. Yan and J. G. Chen, Energy & Envi- ronmental Science, 2016, 9, 62–73. 27 Z. Weng, J. Jiang, Y. Wu, Z. Wu, X. Guo, K. L. Materna, W. Liu, V. S. Batista, G. W. Brudvig and H. Wang,Jour- nal of the American Chemical Society, 2016, 138, 8076–

2006

[5] [5]

Nea¸ tu, J

28 ¸ S. Nea¸ tu, J. A. Maciá-Agulló and H. Garcia, Interna- tional journal of molecular sciences , 2014, 15, 5246–

2014

[6] [6]

Kovacic, B

29 Z. Kovacic, B. Likozar and M. Hus, ACS catalysis, 2020, 10, 14984–15007. 30 S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angewandte Chemie International Edition , 2013, 52, 7372–7408. 31 K. Teramura, T. Tanaka, H. Ishikawa, Y. Kohno and T. Funabiki, The Journal of Physical Chemistry B, 2004, 108, 346–354. 32 S. Navalón, A. Dhakshinamoor...

2020

[7] [7]

36 S. Wang, W. Yao, J. Lin, Z. Ding and X. Wang, Ange- wandte Chemie, 2014, 126, 1052–1056. 37 D. Sun, Y. Fu, W. Liu, L. Ye, D. Wang, L. Yang, X. Fu and Z. Li, Chemistry–A European Journal, 2013, 19, 14279– 14285. 38 Z. Li, N. H. Attanayake, J. L. Blackburn and E. M. Miller, Energy & Environmental Science, 2021, 14, 6242–6286. 39 M. Asadi, K. Kim, C. Liu,...

2014

[8] [8]

Ai, J.-W

45 M. Ai, J.-W. Zhang, Y.-W. Wu, L. Pan, C. Shi and J.- J. Zou, Chemistry–An Asian Journal , 2020, 15, 3599–

2020