⚠️ Verification: lanthanum decahydride — Paper vs Simulation [2026-06-12]

The Paper's Central Claim

Lanthanum decahydride — LaH₁₀ — has held a remarkable distinction in the world of superconductivity. According to the claim sourced from Wikipedia's overview of room-temperature superconductors, this material achieves a superconducting transition temperature (T_c) of approximately 250 K (about −23°C) when squeezed under immense pressure: 150 gigapascals, or roughly 1.5 million times the atmospheric pressure at sea level.

To put that in perspective: 250 K is warmer than a winter night in many northern cities. For a superconductor — a material that conducts electricity with zero resistance — that's extraordinary. Before the hydride revolution that began around 2015, the highest confirmed T_c was around 164 K in mercury-based cuprates, also under pressure. LaH₁₀ didn't just break that record; it shattered it, pushing the frontier of superconductivity tantalizingly close to everyday temperatures.

The catch, of course, is the pressure. 150 GPa is the kind of force found deep inside planetary cores. You need diamond anvil cells — tiny devices that crush samples between the tips of two gem-quality diamonds — to reach these conditions. The material exists in a regime far removed from any practical application. But as a proof of concept, LaH₁₀ represents one of the most important validations of a theoretical prediction in modern condensed matter physics: that hydrogen-rich materials, under sufficient pressure, could superconduct near room temperature.

How Our Simulation Approaches This

At AI Future Lab, we use an AI-driven computational pipeline to estimate superconducting properties of materials from their structural and electronic characteristics. We want to be transparent about what this is — and what it isn't.

Our model is not a first-principles density functional theory (DFT) calculation. It does not solve the Kohn-Sham equations from scratch for every crystal structure. Nor is it a direct application of Migdal-Eliashberg theory, which remains the gold standard for computing electron-phonon coupling in conventional superconductors. Instead, our system is trained on a curated dataset of known superconductors — their structures, electronic densities of states, phonon spectra, and experimentally reported T_c values — and uses machine-learned surrogate models to predict properties for new or underexplored compositions and pressures.

This approach trades some quantitative precision for speed and breadth. We can screen materials rapidly, identify trends, and flag candidates for deeper investigation. But our predictions carry inherent uncertainty, particularly for extreme-pressure phases where training data is sparse and where small structural differences can produce large shifts in superconducting behavior. We report confidence levels honestly: for LaH₁₀, our confidence is medium, reflecting both the strength of the hydrogen-dominant phonon signal our model detects and the difficulty of accurately modeling behavior at megabar pressures.

What Our Analysis Found

Our simulation predicts a superconducting transition temperature of 235 K for LaH₁₀, at a pressure of 170 GPa. The predicted electron-phonon coupling constant λ is 2.1, indicating very strong coupling — consistent with what's expected for a high-T_c conventional superconductor. The material is classified as metastable in our stability analysis, meaning it sits in a local energy minimum that requires pressure to maintain but is not the thermodynamic ground state at ambient conditions.

The mechanism our model identifies aligns well with the established theoretical picture: superconductivity in LaH₁₀ is driven by strong electron-phonon coupling, specifically through high-frequency optical phonon modes dominated by hydrogen vibrations within the material's distinctive clathrate-like H₃₂ cage structure. Lanthanum's d-orbitals hybridize with the hydrogen sublattice near the Fermi level, enhancing the coupling and stabilizing the sodalite-like Fm3m structure that hosts these properties.

In short: our model sees the right physics. The numbers, however, don't land exactly on the claimed values.

⚠️ Partial Match: Reading the Gap

The discrepancy is twofold: our predicted T_c is 15 K lower (235 K vs. 250 K), and our predicted pressure is 20 GPa higher (170 GPa vs. 150 GPa). These gaps deserve careful interpretation rather than dismissal.

First, the T_c difference. A 15 K gap — roughly 6% — is actually within the range of disagreement seen between different DFT studies of LaH₁₀. Published first-principles calculations have reported T_c values ranging from about 215 K to 280 K depending on the functional used, the treatment of anharmonic effects, and whether Coulomb pseudopotential corrections (μ*) are set to 0.10, 0.13, or somewhere in between. Our surrogate model, trained on experimental data that itself carries measurement uncertainty, naturally absorbs some of this spread. A prediction of 235 K is not wrong — it's within the noise floor of the field.

Second, the pressure discrepancy. LaH₁₀ exhibits a complex pressure-dependent phase diagram. The high-symmetry Fm3m phase — the one responsible for the highest T_c — is predicted by some calculations to become stable only above ~130 GPa, with T_c peaking somewhere in the 150–200 GPa range depending on the study. Experimental measurements are complicated by pressure gradients within the diamond anvil cell, calibration uncertainties in the ruby fluorescence scale at extreme pressures, and the challenge of confirming crystal structure in situ. Our model predicting optimal coupling at 170 GPa rather than 150 GPa may simply reflect a slightly different point on the T_c-pressure curve, or a bias in our training data toward higher-pressure stability windows.

There's also a broader reproducibility question. Superconductivity measurements at these pressures are extraordinarily difficult. Samples are micron-scale. Electrical contacts are fragile. The transition can be broad, and defining T_c — onset versus midpoint versus zero resistance — introduces ambiguity. The original experimental reports on LaH₁₀ by Drozdov et al. (2019) and Somayazulu et al. (2019) showed some variation in reported values, and subsequent studies have continued to refine the picture. The "250 K at 150 GPa" figure is a widely cited benchmark, but it is not a single, perfectly defined measurement — it's a representative summary of a body of challenging experimental work.

What This Tells Us About Room-Temperature Superconductivity

LaH₁₀ matters because it confirmed a theoretical prediction decades in the making: that metallic hydrogen and hydrogen-rich compounds should be extraordinary superconductors. The BCS-Eliashberg framework — the theory of phonon-mediated superconductivity — works beautifully here. Hydrogen is light, its phonon frequencies are high, and when you pack enough of it into a metallic lattice under extreme pressure, coupling constants become enormous.

But the gap between 250 K at 150 GPa and a room-temperature superconductor you could use in a power grid is vast. Pressure is the enabler and the prison. For ambient-pressure room-temperature superconductivity to become reality through a conventional (phonon-mediated) mechanism, we would need a material that somehow maintains the high phonon frequencies, strong electron-phonon coupling, and metallic hydrogen-like electronic structure of LaH₁₀ — without the diamond anvil cell. That's a staggering materials design challenge. It may require entirely new structural motifs, metastable phases quenchable to ambient pressure, or unconventional mechanisms that go beyond what Eliashberg theory describes.

Claims of ambient-pressure room-temperature superconductivity — such as the LK-99 episode in 2023 or earlier claims involving carbonaceous sulfur hydride — have not survived independent verification. The history of this field teaches a hard lesson: extraordinary claims demand extraordinary evidence, and superconductivity is a property where measurement artifacts (flux trapping, filamentary paths, resistive transitions in non-superconducting materials) can mimic the real thing. LaH₁₀, by contrast, has been reproduced by multiple groups using complementary techniques. It is real. It is also, for now, confined to the crushing interior of a diamond press.

Our Evolving Simulation

The partial match we see with LaH₁₀ is instructive for us. A 6% T_c deviation and a 13% pressure offset tell us that our surrogate model captures the essential physics — the mechanism, the coupling regime, the stability character — while still lacking the quantitative precision of dedicated ab initio calculations. That's expected, and it's honest.

We are actively working to narrow this gap. Our next model iteration will incorporate anharmonic phonon corrections, which are known to be significant in hydrides where hydrogen atoms undergo large-amplitude vibrations even at zero temperature. We are also expanding our training set with recently published data on ternary hydrides — materials like LaBeH₈ and CaBeH₈ — which may help the model better interpolate across the hydride composition space.

More ambitiously, we are exploring hybrid approaches that use our AI model as a fast pre-screener, then hand promising candidates to automated DFT workflows for refinement. The goal is not to replace first-principles calculations but to direct them — to use machine learning as a compass rather than a map.

LaH₁₀ is, in many ways, the ideal benchmark for this effort. It is well-characterized, theoretically understood, and experimentally reproduced. If we can get LaH₁₀ right to within a few kelvin and a few gigapascals, we can have greater confidence when our model points to something unexpected — a new composition, a new structure, a new pressure window where superconductivity might emerge in a place no one has looked.

The 15 K gap today is a challenge. Tomorrow, it's a calibration point. That's how science — and honest simulation — works.