REVIEW 4 major objections 5 minor 5 references
Superconductivity in MgHCu3 perovskite revisited
T0 review · 4 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash
Pith's one-line read A reexamination of the predicted perovskite hydride superconductor MgHCu3 finds that its previously reported 42 K critical temperature is not reproduced; with the same theoretical approach, computed values fall between about 10 and 31 K, al
desk verdict A useful, honest critical check: MgHCu3's 42 K claim doesn't survive routine k-mesh/functional variation, though the paper's own 40 K upper bound is overconfident because μ* and phonon convergence are unstated. 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 key object is the hypothetical cubic perovskite MgHCu3, studied with the standard DFT-based pipeline for phonon-mediated superconductivity: the PBEsol and PBE functionals, k-point meshes of 12×12×12 through 24×24×24, and the McMillan–Allen–Dynes formula applied to the Eliashberg function. The decisive physical quantity is the phonon dispersion, since the derived spectral function α²F(ω) determines both the electron-phonon coupling constant λ and the logarithmic average frequency ω_log, which together set Tc. The paper's central observation is that this phonon-derived quantity is highly unstable with respect to numerical settings, even though the underlying crystal and electronic structur
What would settle it
A controlled convergence study that varies the phonon q-grid, energy cutoff, and electronic smearing for MgHCu3 — something the authors did not perform — would settle the issue: if a single, robust Tc above 40 K emerges under such a stress test, the paper's central conclusion is wrong. Alternatively, synthesizing MgHCu3 and measuring its resistivity or magnetic susceptibility would provide a direct experimental test of whether superconductivity occurs at 42 K, in the 10–31 K window, or not at all.
Extended reading notes
Core claim
The central claim is a negative replication of the 42 K superconducting critical temperature predicted for MgHCu3 under ambient pressure. The lattice constant and Fermi-level density of states computed here are nearly identical to the earlier report, but the phonon dispersions are extremely sensitive to the choice of functional and k-point grid: some intermediate grids give imaginary (unstable) phonon modes, the densest grid restores stability, and the PBEsol and PBE functionals produce qualitatively different phonon spectra. As a result, the logarithmic average phonon frequency varies from 158 K to 423 K and the electron-phonon coupling constant λ from 0.75 to 2.95, which translates into Tc
Load-bearing premise
The entire conclusion depends on the assumption that the rapid, non-monotonic variation of the computed phonons across k-meshes is a genuine physical response of MgHCu3 rather than an uncontrolled artifact of the numerical settings, since the paper does not vary the phonon q-grid, plane-wave cutoff, or smearing and gives no convergence criterion for the phonon results.
Editorial extensions
If this is right
- If the spread of computed values reflects genuine methodological uncertainty, MgHCu3's ambient-pressure phonon-mediated Tc lies at 10–31 K, not near 42 K.
- The phonons show 're-entrant stability' — imaginary modes at intermediate k-grids but stable at coarse and dense grids — indicating MgHCu3 is close to a phonon-driven instability, which could affect whether the material superconducts at all.
- For MgHCu3, the k-point mesh and functional choice are at least as important as the underlying physics in determining predicted Tc; this caution extends to other computed hydride superconductors.
- When λ exceeds about 1.5 (PBE, 24×24×24), the Migdal–Eliashberg approximation becomes unreliable, so the actual Tc in that regime is not well determined by this method.
- The authors' recommendation is that theoretical high-Tc hydride predictions be cross-checked with multiple functionals and benchmarked against experiment before being cited as evidence for surpassing the classical limit.
Reading between the lines
- If the phonon sensitivity seen in MgHCu3 is common, other published ambient-pressure hydride Tc predictions may carry error bars far larger than their reported precision, and headline 'record' values could shift below or above the classical limit under the same stress tests.
- The 're-entrant stability' pattern suggests a nearby structural phase transition; anharmonic phonon calculations could reveal whether the soft modes actually drive MgHCu3 into a lower-symmetry phase with a different — possibly zero — Tc.
- A direct experimental attempt to synthesize MgHCu3 — even as a metastable film — and measure its resistivity or magnetization would be the decisive benchmark: observing or excluding superconductivity near the predicted temperatures would settle which of the computed values, if any, is physically relevant.
- The paper's methodology (scanning k-meshes and functionals) could be adopted as a routine 'stress test' for any proposed hydride superconductor, catching numerically fragile predictions before they enter the literature.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper re-examines the hypothetical perovskite hydride MgHCu3, which was previously predicted to superconduct at TC = 42 K, just above the often-quoted 40 K 'classical limit' for phonon-mediated superconductivity. Using first-principles DFT with PBEsol and PBE functionals and k-point meshes from 12×12×12 to 24×24×24, the authors find that the crystal structure and electronic density of states are robust, but phonon dispersions, the electron–phonon coupling constant λ, and the resulting McMillan–Allen–Dynes TC are highly sensitive to the computational settings. Their computed TC values range from 9.5 K to 31.3 K, all below 40 K, and they conclude that the earlier 42 K prediction is not reproduced and that high-TC hydride predictions need more careful numerical scrutiny and experimental benchmarking.
Significance. If fully supported, the paper would be a useful cautionary counterexample to a published ambient-pressure hydride superconductivity prediction, showing that numerical sensitivity alone can place TC far below the claimed value. The work has clear strengths: it is an independent re-computation, it compares two functionals and several k-meshes, it makes the underlying data available in the table and SI, and it openly states the limitation of Migdal–Eliashberg theory at large λ. It does not fit parameters to force a desired TC, so the negative result is not circular. However, the central claim that 'TC does not exceed the classical limit' currently rests on an unreported Coulomb pseudopotential μ*, on unverified phonon convergence, and on at least one TC value obtained outside the stated validity regime of the method. These issues are load-bearing and must be addressed before the conclusion can be considered established.
major comments (4)
- [Methods / §2.3, Table 1] The McMillan–Allen–Dynes formula used to compute TC depends on the Coulomb pseudopotential μ*, but μ* is never stated in the text, in Table 1, or in the SI. Since TC is exponentially sensitive to μ*, particularly at λ = 2.95 (PBE, 24×24×24), the reported TC values such as 28.3 K are not reproducible and the conclusion that all values 'do not exceed the classical limit' is not fully supported. The authors should quote μ* for every row and show the TC(μ*) dependence over the standard 0.08–0.15 range.
- [§2.2] The paper documents 're-entrant stability' — PBE is stable at 12×12×12, unstable at 15×15×15 and 18×18×18, and stable again at 24×24×24 — but provides no convergence criterion and no variation of the phonon q-grid, plane-wave cutoff, or smearing. If the intermediate imaginary modes are numerical artifacts, the 24×24×24 'stable' phonons and the TC derived from them may be equally unreliable. A q-grid convergence study, or at least a clear convergence criterion, is necessary before the spread in Table 1 can be interpreted as a faithful representation of method uncertainty.
- [§2.3] The authors state that for λ > 1.5 the Migdal–Eliashberg equations 'do not work well, so the precise TC value in fact remains unknown.' The PBE 24×24×24 row has λ = 2.95, yet its TC = 28.3 K is used as one of the main results. A value obtained outside the claimed validity range cannot support the upper-bound conclusion that the computed TC values do not exceed 40 K. Either an alternative treatment (e.g., full Eliashberg or a stated extrapolation) should be provided, or this row should be excluded from the central claim.
- [Abstract / §2] The abstract states that TC values for 'different Gaussian broadenings' vary from ca. 10 to 31 K, but the range in Table 1 is obtained at a fixed Gaussian broadening of 0.03 Ry while varying k-point mesh and functional. This wording conflates two different numerical sensitivities. The broadening dependence is only reported in the SI and never quantified in the main text. The authors should rephrase the abstract and report the broadening-sensitivity range explicitly so that the source of the TC spread is clear.
minor comments (5)
- [Table 1 caption] Typo: 'k-point mash' should be 'k-point mesh'. The footnote 'PPa)' is also unclear; please spell out the abbreviations.
- [Throughout] The symbol L is used for the electron–phonon coupling constant; the standard notation λ would improve readability and avoid confusion with angular momentum or lattice parameters.
- [§2.1] The reference to 'S1 in SI' should be more precise (e.g., Fig. S1), and the SI figure numbering should be tied explicitly to the statements in the main text.
- [§2.2 / Figure 2] The phrase 'increasing counterclockwise' in the figure caption is ambiguous. Label each panel with the k-point mesh used.
- [Methods] The Gaussian broadening value is given as 0.03 Ry with a citation to previous work, but the choice is not justified here and no convergence test in the broadening is shown in the main text. At minimum, state the range of broadenings explored and the resulting TC spread.
Circularity Check
No significant circularity: the paper recomputes TC from first-principles DFT and compares with an earlier independent prediction; no step reduces to its inputs by construction.
full rationale
The paper's derivation chain is an independent first-principles recomputation: lattice constant, DOS(EF), phonons, Eliashberg function, and the McMillan–Allen–Dynes TC are obtained from DFT calculations with stated functionals and k-meshes. The central claim that TC values lie in the range ca. 10–31 K and 'do not exceed the classical limit' is a direct output of these calculations, not a fitted or assumed value. The comparison target (42 K from ref. [21]) is an external prior prediction, and the authors explicitly report a broad spread of TC values rather than forcing agreement with any target. The choice of Gaussian broadening 0.03 Ry is cited to the authors' own prior work (ref. [19]), but this is a numerical setting, not a load-bearing derived result; the paper states that full broadening dependence is in the SI, so the TC value is not defined by the self-citation. Self-citations to refs. [40–42] appear only in interpretive remarks about soft phonons and the maximum hardness principle, not as evidence for the computed TC. The admitted limitation that Migdal–Eliashberg theory 'does not work well' for λ > 1.5, and the absence of a stated μ*, are correctness/convergence concerns, not circularity: they weaken the certainty of the TC range but do not show that the derivation reduces to its own inputs. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported, and no ansatz is smuggled in via citation. Hence the honest finding is no significant circularity.
Assumptions & free parameters
free parameters (4)
- Coulomb pseudopotential μ* (McMillan–Allen–Dynes)
- Gaussian broadening for Fermi-surface integration =
0.03 Ry
- k-point mesh =
12×12×12, 15×15×15, 18×18×18, 24×24×24
- Phonon q-point mesh / supercell size
assumptions (3)
- standard math BCS/McMillan–Allen–Dynes theory is valid for this system.
- domain assumption GGA-DFT (PBEsol/PBE) accurately describes the electronic structure and electron-phonon coupling of MgHCu3.
- domain assumption The cubic perovskite phase is the phase of interest; imaginary phonons indicate numerical sensitivity or a nearby soft-mode transition rather than a different ground state.
Cite this review
Pith. "Pith review of Superconductivity in MgHCu3 perovskite revisited." pith.science (2026). https://pith.science/paper/UNFITWBE
@misc{pith2026260718117,
author = {Pith},
title = {Pith review of: Superconductivity in MgHCu3 perovskite revisited},
year = {2026},
howpublished = {\url{https://pith.science/paper/UNFITWBE}},
note = {Machine review of arXiv:2607.18117}
}
read the original abstract
We reexamine the crystal structure, electronic structure, lattice dynamics, phonon dispersion, electron-phonon coupling, and superconducting properties of MgHCu3 perovskite using the PBEsol and PBE functionals. This perovskite phase was recently proposed to exhibit superconductivity with the critical superconducting temperature, TC, of 42 K, which falls slightly over the classical 40 K limit for the phonon driven superconductivity. We show that although the crystal and electronic structure of this hypothetical compound are quite robust with respect to the k point mesh and functional used, yet the phonons and phonon related properties are extremely sensitive to the density of the grid chosen as well as functional used for calculations. Correspondingly, the values of the critical superconducting temperature calculated here for different Gaussian broadenings vary in a broad range of ca. 10 to 31 K and they do not exceed the classical limit. We suggest that the properties of this and many other high TC hydrides claimed should be thoroughly scrutinized using a variety of functionals, and benchmarked with experiment, to provide more reliable values of TC.
Figures
Reference graph
Works this paper leans on
-
[1]
re-entrant stability
Introduction The quest for the high-temperature (high-TC) superconductivity (SC) has long been focused on elemental hydrogen[1] and hydrogen-rich materials[2,3,4,5]. This is due to the fact that hydrogen is the lightest chemical element and thus it provides the largest preexponential factor (Debye frequency) in the expression for the superconducting criti...
2021
-
[6]
Microscopic Theory of Superconductivity
L. N. Cooper ”Bound Electron Pairs in a Degenerate Fermi Gas”. Physical Review, 104, no.4 (1956): 1189. https://doi.org/10.1103/PhysRev.104.1189 7. J. Bardeen, L. N. Cooper, J. R. Schrieffer, “Microscopic Theory of Superconductivity”. Physical Review, 106, no.1 (1957): 162. https://doi.org/10.1103/PhysRev.106.162 8. A. P. Drozdov, M. I. Eremets, I. A. Tro...
-
[17]
Superconductivity in the palladium-silver-deuterium system
W. Buckel, B. Stritzker “Superconductivity in the palladium-silver-deuterium system” Physical Letters A 43 (1973): 403. https://doi.org/10.1016/0375-9601(73)90393-9 18. W. Grochala, P. P. Edwards ” Thermal Decomposition of the Non-Interstitial Hydrides for the Storage and Production of Hydrogen” Chemical Reviews 104, no.3 (2004): 1283. https://doi.org/10....
arXiv 1973
-
[28]
High-TC superconductivity above 130 K in cubic MH4 compounds at ambient pressure
X. Li, W. Xu, Z. Zhou, J. Shi, H. Liu, Y. -W. Fang, W. Cui, Y. Li, M. A. L. Marques, “High-TC superconductivity above 130 K in cubic MH4 compounds at ambient pressure”, arXiv:2511.04222 (2025). https://doi.org/10.48550/arXiv.2511.04222 29. K. Dolui, L. J. Conway, C. Heil, T. A. Strobel, R. P. Prasankumar, C. J. Pickard “Feasible Route to High-Temperature ...
-
[39]
Interactions between electrons and lattice vibrations in a superconductor
G. M. Eliashberg “Interactions between electrons and lattice vibrations in a superconductor” Soviet Physics - JETP 11 (1960): 3. 40. D. Slocombe, V. Kuznetsov, W. Grochala, R. P. J. Williams, P. P. Edwards, “Superconductivity in transition metals” Philosophical Transactions A 373, no.3 (2015): 20140476. https://doi.org/10.1098/rsta.2014.0476 41. W. Grocha...
arXiv 1960
Reviewed August 1, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.