REVIEW 5 major objections 5 minor 33 references
Effects of high-order Van Hove singularities on exciton and trion energy dispersions
T0 review · 5 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash
Pith's one-line read This paper claims that a higher-order van Hove singularity in a two-dimensional semiconductor's valence band is imprinted onto the exciton and trion densities of states, producing a power-law divergent DOS and new optical states, and that t
desk verdict The paper's real results are the Mexican-hat exciton/trion parameter maps; the abstract's trion HOVHS and materials claims are not in the body. 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 objects are the variational wavefunctions: a 1s hydrogenic form for the exciton (Eq. 4) and a symmetrized product of two 1s forms for the positive trion (Eq. 7), minimized against the Rytova-Keldysh Coulomb interaction. The Mexican-hat dispersion Ak^2 - Bk^4 provides a valence band with a VHS at the band edge and a divergent effective mass; adding a monkey-saddle term D k^3 cos 3φ creates a higher-order van Hove singularity (HOVHS) with a power-law DOS g(E) ~ |E|^{-1/3}. The comparison between the 'total' binding energies (exciton vs. trion) and the 'trion binding energy' (the difference between them) is the diagnostic that reveals width-and-depth-dependent favorability.
What would settle it
Recompute the exciton DOS for a monkey-saddle valence band using a variational basis larger than the single 1s hydrogenic state; if the log-log slope of the DOS near the singularity deviates from the -1/3 power law, the HOVHS mirroring is an artifact of the ansatz.
Extended reading notes
Core claim
The paper's central discovery is that the density of states of the valence band is directly imprinted on the density of states of the exciton and trion: when the valence band hosts a HOVHS of monkey-saddle type, the exciton dispersion retains the same three-fold symmetry and shows a power-law divergent DOS (log-log slope consistent with a power law), meaning the bound state itself hosts a HOVHS. Additionally, for a Mexican-hat valence band with spin-degenerate holes, the positive trion becomes more favorable relative to the exciton as the hat becomes narrower, despite the overall binding energy decreasing, because the trion binding energy—the difference between three- and two-particle bindin
Load-bearing premise
The quantitative predictions assume the fixed variational ansatz (a single 1s state for the exciton, a product of two 1s states for the trion) is close to the true ground state, so the trion binding energy—computed as a difference of two upper bounds—and the power-law DOS exponent could shift if the ansatz is too stiff; the artificial Brillouin-zone cutoff W=1.4 also shifts absolute binding energies by about 10 percent.
Editorial extensions
If this is right
- For a fixed Mexican-hat depth, increasing the hat width (k_max) raises the overall exciton and trion binding energies, but the positive trion becomes relatively more favorable when the hat is narrow.
- Decreasing the Mexican-hat depth at fixed width increases both binding energies, with the trion gaining more, so at small depths the trion can become more bound than the exciton plus a free hole.
- A monkey-saddle HOVHS in the valence band produces an exciton dispersion with the same three-fold symmetry and a power-law divergent DOS, i.e., an excitonic HOVHS, implying new states that affect optical properties.
- Adding a C3-symmetric HOVHS perturbation to a bright-exciton system splits the Γ-K and Γ-K' dispersions, creating an extra momentum-dark DOS peak that grows relative to the bright peak with increasing perturbation strength—potentially extending dark-state lifetime.
- Trion binding energy (the difference between three- and two-particle binding) is insensitive to the artificial Brillouin-zone cutoff, while absolute exciton binding shifts by about 10 percent with the cutoff W.
Reading between the lines
- If the HOVHS mirroring survives beyond the 1s ansatz, band-structure engineering (strain, twist, stacking) of a valence-band monkey saddle becomes a direct knob for creating power-law divergent exciton reservoirs, which could affect exciton transport and nonlinear optical response.
- The narrow-hat trion favorability rule is testable in real material families: among hexagonal chalcogenides with inverted Mexican-hat valence bands, the positive trion should dominate photoluminescence in the thinnest or most compressed members where k_max is smallest.
- The paper's arguments for valence-band HOVHS could be carried over to conduction-band HOVHS, where negatively charged trions would inherit the singularity; running the same calculation on the electron side would test whether the mirroring is universal.
- Since the paper's key diagnostic is the difference of two variational upper bounds, verifying the ~5 meV positive-trion binding with a more flexible wavefunction or a different method (e.g., quantum Monte Carlo) would either confirm or overturn the design rule that narrow hats favor trions.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This manuscript studies exciton and positive-trion formation in two-dimensional semiconductors whose valence bands contain either a conventional Mexican-hat van Hove singularity (VHS) or a higher-order van Hove singularity (HOVHS) of monkey-saddle type. The method is variational: the exciton uses a 1s hydrogenic momentum-space wavefunction (Eq. 4) and the trion uses a symmetrized product of two such states (Eq. 7), with a Rytova-Keldysh Coulomb interaction and monolayer-InSe-like parameters. In Sec. III the authors vary the Mexican-hat width and depth and conclude that narrower and shallower hats favor trion formation relative to the exciton. In Sec. IV they add a D k^3 cos(3φ) term to the valence band, compute the exciton dispersion and DOS, and claim that the HOVHS is "mirrored" in the exciton DOS. The abstract and title extend this claim to trions and to specific materials (InSe, GaSe, α-SnAs), but the body contains no trion HOVHS calculation and no materials-specific study beyond InSe parameters.
Significance. If correct, the qualitative result that a band-structure HOVHS imprints a power-law divergence on the bound-state exciton DOS is of interest for optical and lifetime engineering in 2D semiconductors. The paper also provides a systematic variational treatment of a Mexican-hat valence band, including useful Appendix checks of the dielectric environment, effective mass, and Brillouin-zone cutoff. However, the central advertised claims are broader than the evidence: the trion HOVHS half of the abstract is unsupported, and the quantitative reliability of the variational binding energies is not established. The paper would be strengthened by a clear separation of the structural (kinetic) contribution to the exciton dispersion from the interaction contribution, and by a report of the fitted power-law exponent. The manuscript is not written with reproducibility artifacts (no code or data deposition is mentioned), but the analytic framework is clearly stated.
major comments (5)
- [Title, Abstract, Sec. IV] The title and abstract promise that HOVHS effects are mirrored in both exciton and trion dispersions and DOS, and the abstract specifically states that "the DOS of excitons and trions containing such singularities is dramatically enhanced." In the body, Sec. IV.A announces that "we examine how the exciton and positive trion bound states are affected through their energy dispersion and DOS calculation," but only excitonic dispersions and DOS are shown (Figs. 5, 6, 7). No trion dispersion or DOS is computed for any HOVHS case. The trion half of the central claim is therefore unsupported. The abstract also lists InSe, GaSe, and α-SnAs as studied materials, but the body contains only InSe-derived parameters; no GaSe or α-SnAs calculation appears.
- [Sec. II, Eq. (4), Eq. (7), App. C] The variational ansatz is a fixed 1s hydrogenic state for the exciton and a symmetrized product of two 1s states for the trion, with no basis-convergence check. The reported positive-trion binding energy of about 5 meV (App. C) is the difference of two variational upper bounds of order 90–104 meV. A difference of upper bounds is not itself an upper bound, so the uncertainty in this difference is uncontrolled. The key Sec. III conclusion that trion formation becomes more favorable for narrower Mexican hats rests on this difference. A larger variational basis (e.g., multiple s-like or symmetry-adapted states) or an alternative benchmark is needed to establish that the 5 meV scale and its trends are robust.
- [Sec. IV.B, Fig. 7] The central HOVHS claim is supported by the log-log inset in Fig. 7, which the text says shows a linear relation consistent with a power-law divergence. However, no fitted exponent, fitting range, or convergence with grid density is reported. The expected exponent for a monkey-saddle DOS is |E|^{-1/3}. Reporting the fitted slope and the energy window over which it holds is essential, especially because the finite Monkhorst-Pack grid and the artificial Brillouin-zone cutoff (Eq. C1) will cut off any true divergence. Without this quantitative check, the statement that the exciton dispersion "hosts a HOVHS" is not established beyond the qualitative kinetic-structure argument.
- [Sec. II, Eq. (1), Sec. IV.B] The "mirroring" of the valence-band HOVHS in the exciton dispersion is, to leading order, a structural consequence of Eq. (1): the exciton kinetic term contains -E_v(k-Q), so a monkey-saddle in the valence band will appear directly in the exciton dispersion whenever the interaction (binding-energy) part varies weakly with Q. The paper presents this as a finding, but it is expected from the single-particle term. The nontrivial question is how the binding-energy variation modifies the dispersion and DOS. The manuscript does not isolate this contribution, e.g., by comparing the full result with a calculation that keeps the binding energy fixed at its Q=0 value. A quantitative decomposition would clarify what is genuinely due to many-body effects.
- [Sec. III, Fig. 2, Fig. 3] The Sec. III conclusion about trion favorability is presented as a general trend for spin-degenerate Mexican-hat valence bands, but the parameter sweep is narrow (one A/B ratio, one conduction-band mass, one screening length). The trend in Fig. 3(b) is plausible, but the claim is made without an error bar or sensitivity analysis. Given the variational-difference issue in App. C, at least a check at the extremes of the shown parameter range with a larger basis would be needed before this can be stated as a general result. The current evidence is suggestive rather than conclusive.
minor comments (5)
- [Sec. IV.B] There is an internal reference typo: "Inset of Fig. reffig:msDOSex" should refer to Fig. 7. Also, the sentence "The effects can be enhanced with large enough D, which causes these two states to be further apart in energy" is a fragment.
- [Sec. II, Eq. (6)] The symmetry factor (-1)^S is not defined clearly. Since the text says the ground state is a singlet, it would be clearer to write the explicit singlet/triplet symmetry and state that the singlet is used throughout.
- [Sec. III, Fig. 2] The caption of Fig. 2(b) says "continuous binding energies line of the exciton and trion," but the figure shows two curves; also the terms "total binding energy" and "trion binding energy" are used inconsistently in the text. Please define once whether "trion binding energy" refers to the energy difference between the trion and exciton plus free carrier, or to the total three-particle binding energy.
- [Sec. I, Fig. 1] The phrase "divergent effective mass" is potentially misleading: a Mexican-hat dispersion has a vanishing inverse effective mass at k=0, not an infinite effective mass. Please rephrase.
- [App. C] The Brillouin-zone cutoff W is a free parameter. The statement that the trion binding energy "does not depend on how far away from the VBM the defined grid edge is" is supported only at W=1.2, 1.4, 1.6 for one valence band. Please specify the grid size and convergence criteria used in the Monkhorst-Pack sums.
Circularity Check
Exciton HOVHS 'prediction' is the input valence-band singularity by construction; the abstract's trion HOVHS claim is not computed in the body.
-
renaming known result
[Sec. IV.B, Eqs. (1) and (16), Figs. 6-7]
"Eexϕex(k) = [Ec(k)−E v(k−Q)]ϕ ex(k) + 1/A Σ_q ϕex(k−q)V(q) ... We find that the exciton dispersion has the same monkey saddle form as that of the valence band retaining its same symmetry properties. ... In particular, we find that the exciton dispersion hosts a HOVHS."
The only Q-dependence of the exciton Hamiltonian enters through −E_v(k−Q). With E_v(k)=Dk^3 cos3φ (Eq. 16), the expectation of −E_v(k−Q) over a symmetric envelope reproduces the same cubic form in Q (the cubic term survives independently of the wavefunction, while the linear terms cancel for ⟨k_x^2⟩=⟨k_y^2⟩). Thus the 'exciton HOVHS' and its power-law DOS are essentially the input valence-band singularity rewritten in center-of-mass coordinates. The variational calculation supplies binding-energy magnitudes, but the headline singularity is structurally forced by the model, not independently derived.
full rationale
The paper is a variational model study, not a data-fitting exercise. The Mexican-hat width/depth trends (Sec. III), the effective-mass and dielectric dependencies (Apps. A and B), and the Brillouin-zone cutoff analysis (App. C) are genuine forward calculations with stated assumptions; they are not circular. Self-citations to [10] and [20] are used for the trion Hamiltonian/variational method and for the standard power-law DOS of a monkey saddle, respectively; they are not load-bearing uniqueness theorems and do not by themselves create circularity. The one reduce-by-construction step is the HOVHS mirroring claim in Sec. IV.B: Eq. (1) contains the valence band only through −E_v(k−Q), and with the monkey-saddle E_v of Eq. (16) the resulting exciton Q-dispersion is forced to have the same cubic saddle form. Consequently, the 'finding' that the exciton dispersion hosts a HOVHS, and the associated power-law DOS, restate the input valence-band singularity in new coordinates rather than providing an independent prediction. Separately, the abstract and title extend the HOVHS claim to trions ('the DOS of excitons and trions ... is dramatically enhanced'), but the body contains no trion dispersion or DOS calculation for any HOVHS case; this is an evidentiary gap rather than circularity. Overall, because the central HOVHS claim is partially forced by construction while the Mexican-hat competition and quantitative binding energies retain independent content, a score of 6 is appropriate.
Assumptions & free parameters
free parameters (6)
- Brillouin-zone cutoff factor W =
1.4 (scanned 1.2-1.6)
- Variational inverse-radii (beta, lambda; alpha, beta, gamma, kappa) =
minimized per configuration (unreported)
- Mexican-hat coefficients A, B =
e.g., 5 eV A^2, 300 eV A^4 (E_v2)
- Monkey-saddle amplitude D =
0, 0.5, 1, 2.5, 5 eV A^3
- Conduction-band effective mass m_e =
0.15 (scanned 0.05-0.35)
- Screening length r_0 (InSe parameters) =
InSe values from Ref. [9]
assumptions (5)
- domain assumption The 1s hydrogenic (Eq. 4) and symmetrized 1s x 1s (Eq. 7) trial states capture the true ground-state exciton and trion, with negligible wavefunction-overlap corrections.
- domain assumption Rytova-Keldysh form with InSe screening describes the effective Coulomb interaction.
- domain assumption Valence band is a low-order polynomial Ak^2 - Bk^4 (with optional +/- Dk^3 cos(3phi)); conduction band is parabolic with constant mass.
- ad hoc to paper Variational minimization over a Monkhorst-Pack grid with BZ cutoff W=1.4 yields converged dispersions.
- domain assumption Spin-degenerate valence bands and a singlet positive trion; results carry over to negative trions.
Cite this review
Pith. "Pith review of Effects of high-order Van Hove singularities on exciton and trion energy dispersions." pith.science (2026). https://pith.science/paper/7E2YOQL2
@misc{pith2026251115824,
author = {Pith},
title = {Pith review of: Effects of high-order Van Hove singularities on exciton and trion energy dispersions},
year = {2026},
howpublished = {\url{https://pith.science/paper/7E2YOQL2}},
note = {Machine review of arXiv:2511.15824}
}
abstract
We investigate the effects of Van Hove singularities in the electronic band structure of two-dimensional semiconductors on the energy dispersion of excitons and positive trions. In particular, we study valence band energy dispersions which possess (i) a typical logarithmic Van Hove singularity, (ii) a line high-order Van Hove singularity (HOVHS) from a Mexican-hat dispersion or (iii) a point HOVHS such as a monkey saddle. We find that the density of states (DOS) of excitons and trions containing such singularities is dramatically enhanced and shows, in general, how the HOVHS in the valence band can strongly affect and be mirrored in the DOS of excitons and trions. This leads to new states that govern the optical properties of the system. In addition, we study a set of materials, InSe, GaSe and $\alpha$-SnAs, from a class of materials in which the topmost valence band has an inverted Mexican-hat shape. The most favourable exciton occurs when the singularity is at the $\Gamma$-point, as in the example of monolayer $\alpha$-SnAs, which hosts a HOVHS. Our work thus provides a pathway to engineer specific bound states in two-dimensional materials that host such singularities, thereby opening new avenues for potential applications.
Figures
Figures from the paper (7 more)
Reference graph
Works this paper leans on
-
[1]
Frenkel, Physical Review37, 17 (1931)
J. Frenkel, Physical Review37, 17 (1931)
1931
-
[2]
M. A. Lampert, Physical Review Letters1, 450 (1958)
1958
-
[3]
Zhang, H
C. Zhang, H. Wang, W. Chan, C. Manolatou, and F. Rana, Phys. Rev. B89, 205436 (2014)
2014
-
[4]
H. Wang, C. Zhang, W. Chan, C. Manolatou, S. Tiwari, and F. Rana, Phys. Rev. B93, 045407 (2016)
2016
-
[5]
Vaclavkova, J
D. Vaclavkova, J. Wyzula, K. Nogajewski, M. Bartos, A. O. Slobodeniuk, C. Faugeras, M. Potemski, and M. R. Molas, Nanotechnology29, 325705 (2018)
2018
-
[6]
R. Tempelaar and T. C. Berkelbach, Nature Communi- cations10, 10.1038/s41467-019-11497-y (2019)
-
[7]
F. Rana, O. Koksal, and C. Manolatou, Phys. Rev. B 102, 085304 (2020)
2020
-
[8]
S. Shradha, R. Rosati, H. Lamsaadi, J. Picker, I. Par- adisanos, M. T. Hossain, L. Krelle, L. Oswald, N. En- gel, D. Markina,et al., arXiv preprint arXiv:2510.21422 (2025)
arXiv 2025
Show all 33 references
-
[9]
Ceferino, K
A. Ceferino, K. W. Song, S. J. Magorrian, V. Z´ olyomi, and V. I. Fal’ko, Phys. Rev. B101, 245432 (2020)
2020
-
[10]
L. J. Burke, M. T. Greenaway, and J. J. Betouras, npj 2D Materials and Applications9, 63 (2025)
2025
-
[11]
Brotons-Gisbert, H
M. Brotons-Gisbert, H. Baek, A. Campbell, K. Watan- abe, T. Taniguchi, and B. D. Gerardot, Physical Review X11, 031033 (2021)
2021
-
[12]
Brotons-Gisbert, B
M. Brotons-Gisbert, B. D. Gerardot, A. W. Holleitner, and U. Wurstbauer, MRS bulletin49, 914 (2024)
2024
-
[13]
C. Lui, A. Frenzel, D. Pilon, Y.-H. Lee, X. Ling, G. Ak- selrod, J. Kong, and N. Gedik, Physical review letters 113, 166801 (2014)
2014
-
[14]
E. Liu, J. van Baren, Z. Lu, M. M. Altaiary, T. Taniguchi, K. Watanabe, D. Smirnov, and C. H. Lui, Phys. Rev. Lett.123, 027401 (2019)
2019
-
[15]
M. R. Molas, C. Faugeras, A. O. Slobodeniuk, K. No- gajewski, M. Bartos, D. Basko, and M. Potemski, 2D Materials4, 021003 (2017)
2017
-
[16]
Malic, M
E. Malic, M. Selig, M. Feierabend, S. Brem, D. Chris- tiansen, F. Wendler, A. Knorr, and G. Bergh¨ auser, Phys- ical Review Materials2, 014002 (2018)
2018
-
[17]
Zinkiewicz, A
M. Zinkiewicz, A. Slobodeniuk, T. Kazimierczuk, P. Kapu´ sci´ nski, K. Oreszczuk, M. Grzeszczyk, M. Bar- tos, K. Nogajewski, K. Watanabe, T. Taniguchi,et al., Nanoscale12, 18153 (2020)
2020
-
[18]
Perea-Causin, S
R. Perea-Causin, S. Brem, O. Schmidt, and E. Malic, Physical Review Letters132, 036903 (2024)
2024
-
[19]
Van Hove, Phys
L. Van Hove, Phys. Rev.89, 1189 (1953)
1953
-
[20]
Chandrasekaran and J
A. Chandrasekaran and J. J. Betouras, Phys. Rev. B105, 075144 (2022)
2022
-
[21]
Classen and J
L. Classen and J. J. Betouras, Annual Review of Con- densed Matter16, 229 (2025)
2025
-
[22]
Chandrasekaran, L
A. Chandrasekaran, L. C. Rhodes, E. A. Morales, C. A. Marques, P. D. C. King, P. Wahl, and J. J. Betouras, Nature Communications15, 9521 (2024)
2024
-
[23]
Chandrasekaran and J
A. Chandrasekaran and J. J. Betouras, Advanced Physics Research2, 2200061 (2023)
2023
-
[24]
Z´ olyomi, N
V. Z´ olyomi, N. Drummond, and V. Fal’Ko, Physical Re- view B89, 205416 (2014)
2014
-
[25]
N. S. Rytova, arXiv preprint arXiv:1806.00976 (2018)
2018 arXiv
-
[26]
L. V. Keldysh, Soviet Journal of Experimental and The- oretical Physics Letters29, 658 (1979)
1979
-
[27]
H. J. Monkhorst and J. D. Pack, Phys. Rev. B13, 5188 (1976)
1976
-
[28]
J. D. Pack and H. J. Monkhorst, Phys. Rev. B16, 1748 (1977)
1977
-
[29]
S. J. Magorrian, V. Z´ olyomi, and V. I. Fal’ko, Physical Review B94, 245431 (2016)
2016
-
[30]
Shtyk, G
A. Shtyk, G. Goldstein, and C. Chamon, Phys. Rev. B 95, 035137 (2017)
2017
-
[31]
Chowdhury, S
T. Chowdhury, S. Chatterjee, D. Das, I. Timokhin, P. D. N´ u˜ nez, G. M. A, S. Chatterjee, K. Majumdar, P. Ghosh, A. Mishchenko,et al., Physical Review B110, L081405 (2024)
2024
-
[32]
Qu and Y
H. Qu and Y. Li, Physical Review B107, 235407 (2023)
2023
-
[33]
N. T. Paylaga, C.-T. Chou, C.-C. Lin, T. Taniguchi, K. Watanabe, R. Sankar, Y.-h. Chan, S.-Y. Chen, and W.-H. Wang, npj 2D Materials and Applications8, 12 (2024). 8 Appendix A: Dependence on the dielectric environment Here we investigate the effect of the product of the in-pla...
2024
Reviewed August 3, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.