SciPost Submission Page
Superconductivity in RbH$_{12}$ at low pressures: an \emph{ab initio} study
by Đorđe Dangić, Yue-Wen Fang, Ion Errea
Submission summary
| Authors (as registered SciPost users): | Dorde Dangic |
| Submission information | |
|---|---|
| Preprint Link: | scipost_202507_00044v2 (pdf) |
| Date submitted: | Jan. 16, 2026, 12:01 p.m. |
| Submitted by: | Dorde Dangic |
| Submitted to: | SciPost Physics |
| Ontological classification | |
|---|---|
| Academic field: | Physics |
| Specialties: |
|
| Approaches: | Theoretical, Computational |
Abstract
High-pressure polyhydrides are leading contenders for room temperature superconductivity. The next frontier lies in stabilizing them at ambient pressure, which would allow their practical applications. In this first-principles computational study, we investigate the potential for record-low pressure stabilization of binary superhydrides within the RbH$_{12}$ system, including lattice quantum anharmonic effects in the calculations. We identify five competing phases for the pressure range between 0 and 100 GPa. Incorporating anharmonic and quantum effects on ion dynamics, we find the $Immm$ and $P6_3/mmc$ phases to be the most probable, potentially metastable even at pressures as low as 10 GPa. Notably, all phases exhibit metallic properties, with critical temperatures between 50 and 100 K, within the pressure range where they are dynamically stable. These findings have the potential to inspire future experimental exploration of high-temperature superconductivity at low pressures in Rb-H binary compounds.
Author indications on fulfilling journal expectations
- Provide a novel and synergetic link between different research areas.
- Open a new pathway in an existing or a new research direction, with clear potential for multi-pronged follow-up work
- Detail a groundbreaking theoretical/experimental/computational discovery
- Present a breakthrough on a previously-identified and long-standing research stumbling block
List of changes
- The sentence in the last paragraph of the introduction has been revised following the first referee's suggestion.
- A reference to the first-principles Coulomb-interaction calculation has been added in the Methods section.
- The first paragraph on pg.~3 now explicitly states that at $0$~K the Gibbs free energy equals the enthalpy.
- The second-to-last paragraph on pg.~3 now clarifies that the values shown in Fig.~1 are relative enthalpies and not distances from the convex hull.
- The first paragraph on pg.~5 has been updated to clarify that the negative phonon frequencies in Fig.~3 arise from interpolation artifacts.
- Additional information regarding the ab-initio treatment of Coulomb repulsion in the Migdal--Eliashberg calculations has been included in the second paragraph on pg.~6.
- Figure~5 now presents the electronic density of states calculated using the same Gaussian smearing employed in the electron-phonon calculations.
- The caption of Fig.~6 has been revised to better explain how the superconducting critical temperatures were obtained.
- All Migdal--Eliashberg calculations involving $\mu^*$ have been repeated using $\mu^* = 0.118$, consistent with the value obtained for the $Immm$ phase.
- A section describing additional calculations using larger SSCHA supercells has been added to the Supplementary Material.
- A section explaining our approach to solving the Migdal--Eliashberg equations with an ab-initio Coulomb interaction has been added to the Supplementary Material.
- Nine additional references have been included to cite the relevant work that enabled the additional calculations.
Current status:
Reports on this Submission
Strengths
Weaknesses
Concerning the Raman activity, I don't see where the authors picked up in my comment that I suggested using the harmonic frequencies. They could use their anharmonic peak centers. I also do not see how the intensities obtained from a random tensor convey information on the relative mode visibility. Anyway, the authors put a disclaimer so at least the readers will be able to decide for themselves how to interpret this result.
Report
Recommendation
Accept in alternative Journal (see Report)
Anonymous on 2026-01-21 [id 6255]
Comment on the ML validation of phonon instabilities: Potential Circular Logic in ML-Based Phonon Validation
I may be missing something, but there seems to be a logical issue in how the ML potentials are used to validate the phonon results.
In Supplementary Section 8, the ML potentials are described as being fine-tuned on the same DFT data obtained from smaller supercells. These ML potentials are then used to argue that the imaginary phonon freqns observed in those smaller supercells are merely interpoln. artifacts.
The difficulty is that this does not provide an independent verification. If the ML model is trained on the same DFT configurations that produced the questionable phonons in the first place, then the agreement between the ML and DFT results mainly shows internal consistency and it does not show that the original instabilities are non-physical.
Any systematic bias or problems in the original DFT phonons will be ingested & learned and be reproduced by the ML potential.
To really show that the imaginary modes are in fact the interpolation effects and not the genuine soft modes, ideally you should do some form of independent validation - such as training on a genuinely different configuration set, performing direct DFT phonon calculations on the larger supercells, or comparing against experimental constraints. Without that independence, the ML analysis does not fully resolve the underlying stability issue.