Hydride superconductors: pressure, evidence and recovery

Scope

Hydrogen-rich superconductors connect light nuclei, strong electron–phonon coupling and pressure-dependent chemistry. Ashcroft's metallic-hydrogen proposal and later hydrogen-dominant-alloy proposal motivate this route; neither established a recoverable ambient-pressure superconductor. H₃S and LaH₁₀ are important high-pressure experimental systems. Their formulae, pressure scales and sample histories must accompany any quoted transition temperature. R1 R2 R4 R7

“Chemical precompression” describes a design idea, not a guarantee of low operating pressure. A predicted metallic structure, a synthesized hydride and a superconducting phase are different achievements. Distinguish the starting H₂S reagent from its proposed H₃S product, and a clathrate-like hydrogen framework from directly measured hydrogen positions. The sulfur-hydride diffraction study supports the sulfur sublattice and coexistence with sulfur; its measured patterns cannot by themselves distinguish hydrogen arrangements with the same sulfur periodicity. R2 R3 R5

Mathematical assumptions

For a specified structure, a commonly used modified-McMillan estimate is $T_c=(\omega_{\log}/1.2)\exp[-1.04(1+\lambda)/(\lambda-\mu^*(1+0.62\lambda))]$, with $\omega_{\log}$ expressed in kelvin here. Duan's harmonic calculations give $\lambda=2.07$, $\omega_{\log}=1125.1$ K for R3m at 130 GPa and $\lambda=2.19$, $\omega_{\log}=1334.6$ K for Im$\bar{3}$m at 200 GPa. Substitution with $\mu^*=0.13$ and 0.10 gives approximately 155–166 K and 191–204 K, respectively. This reproduces the printed estimate convention, without additional strong-coupling or spectral-shape factors; it does not reproduce the underlying density-functional perturbation calculation. R3

Hydrogen motion can invalidate a harmonic treatment. Errea and colleagues calculated a change from $\lambda=2.64$ to 1.84 for H₃S at 200 GPa when including anharmonicity. Their full isotropic Migdal–Eliashberg calculation with $\mu^*=0.16$ changed the calculated transition from 250 to 194 K. That table is not obtained by simply substituting parameters into the preceding approximate formula. Zero-point contributions also alter the calculated decomposition landscape: both $3\mathrm{H_2S}\rightarrow2\mathrm{H_3S}+\mathrm{S}$ and $5\mathrm{H_2S}\rightarrow3\mathrm{H_3S}+\mathrm{HS_2}$ conserve atoms, but atom balance alone does not establish stability or a reaction pathway. R8

An isotope comparison uses $\alpha=\ln(T_{c,\mathrm H}/T_{c,\mathrm D})/\ln2$ only after checking structure, pressure and transition criteria. The final LaH₁₀ paper compares approximately 249 K and 180 K for fcc hydride and deuteride near 150 GPa; its Fig. 4 specifies 151 and 152 GPa. These rounded values give 0.468, while the paper states 0.46. The early author's 250/168 K comparison gives 0.573 and must not be silently substituted for the final result. Phase mixtures and different deuteride structures further limit interpretation. R7

Experimental criteria

For sulfur hydride, the 2015 report combined resistance, field response, isotope substitution and magnetization. Its author-version Fig. 4a reports a zero-field-cooled magnetic onset of 203(1) K at 155 GPa, with a 20 Oe field-cooled comparison. The plotted transport comparison is a different pressure, 145 GPa, and must not be treated as the identical specimen condition. Background subtraction and diamond-anvil-cell contributions require scrutiny; this scoped reading does not independently reproduce raw magnetization. The later simultaneous diffraction/transport experiment on deuteride at 173 GPa provides structural context, not a direct measurement of every hydrogen coordinate. R4 R5

The two early lanthanum-superhydride studies have different evidence scopes. Somayazulu's author v3 describes a large resistance drop around 260 K, with room-temperature pressure changing from 188 to 196 GPa across a thermal cycle; pressure at the actual transition was not measured in that run. The lowest four-probe resistance was about 20 μΩ, while pseudo-four-probe measurements carried contact-resistance limitations. This should not be rewritten as an independently established zero-resistance or magnetic-shielding result. The final journal abstract was checked, but the selected detailed methods here come from the identified author version. R6

Drozdov's final 2019 LaH₁₀ paper reports zero resistance, isotope dependence and field suppression, with a maximum near 250–252 K around 170 GPa. Its transition criterion is the onset of deviation from normal metallic resistance. Measurements to 9 T support an extrapolated zero-temperature upper critical field of approximately 95–136 T; they are not direct measurements at 136 T. The authors explicitly state that their tiny samples were below the available SQUID sensitivity. Their H₂/D₂ vibron pressure estimates were often 10–30 GPa lower than diamond-edge estimates around 150 GPa, so pressure calibrations cannot be freely interchanged. R7

Limits and disputes

The 2020 carbonaceous-sulfur-hydride article was retracted on 26 September 2022. The corrected notice identifies concerns about non-standard, user-defined magnetic-background subtraction; all authors disagreed with the retraction. The 2023 nitrogen-doped lutetium-hydride article was retracted on 7 November 2023 following concerns including sample provenance, measurements and processing. These original papers remain linked as historical claims, accompanied by their notices. Their claimed temperatures are not usable experimental benchmarks in this catalogue. A retraction notice is not a substitute for an independent misconduct finding. R9 R10 R11 R12

Ming's independent LuH₂±xNᵧ study found no superconductivity in its transport measurements down to 2 K at sampled pressures spanning 0.4–40.1 GPa. Samples were synthesized at 2 GPa and 300–350 °C for 10 hours. This preparation and finite measurement window do not exclude every possible Lu–N–H composition or an unmeasured pressure interval.

The methods specify room-temperature ruby pressure calibration and four-probe van der Pauw resistance with platinum electrodes. EDS detects nitrogen but cannot detect hydrogen here; it does not fix both x and y. The pressure range is a sampled envelope, not continuous coverage. The arXiv v3 abstract, dated 2 June 2023, still describes 1–6 GPa and a 10 K floor; this catalogue uses the expanded final-paper transport window. R13

Discovery implications

A credible search must connect composition and structure to dynamical and thermodynamic stability, finite-temperature and nuclear-quantum effects, synthesis access and multiple superconductivity probes. A finite structure search can miss larger cells; absence of imaginary harmonic frequencies is a local dynamical check, not proof of global stability. Preserve decompression history, mixed phases, pressure standards and unsuccessful samples instead of ranking formulae by isolated temperature claims. R3 R5 R8

A June 2026 study recovered Y₃Fe₄H₂₀ at ambient pressure after synthesis above about 60 GPa, with survival on the scale of tens of hours. This is useful progress in hydride recovery, but its reported SQUID test found no superconductivity down to 3 K in the tested recovered sample. Hydrogen positions were supported by structure searches, and the sample was small and not pure. Recovery, long-term stability and ambient-pressure superconductivity remain distinct questions. Full raw-data, convergence and cross-laboratory assessment remain open for this article. R14

Related topics

Continue with isotope effects, Eliashberg theory, critical fields and elemental superconductors.

Material cases

Values describe individual reports, not a ranking. Unknown means the checked source did not supply the field at this review scope.