⚠️ Verification: LaH₁₀ — Paper vs Simulation [2026-06-09]

The Paper's Central Claim

In a 2026 landscape review of room-temperature superconductor research, PatSnap identifies lanthanum decahydride — LaH₁₀ — as holding the crown for the highest independently validated superconducting critical temperature (Tc) achieved to date: approximately 260 K (about −13°C) under extreme pressures of 170–190 GPa. That's roughly 1.7 million times atmospheric pressure, the kind of force you'd find deep inside a planet's core, generated in a laboratory between two diamond tips smaller than a fingernail.

To put the temperature in perspective: 260 K is just barely below the freezing point of water. For a superconductor — a material that conducts electricity with zero resistance — this is extraordinary. For most of the 20th century, superconductivity was a phenomenon confined to temperatures near absolute zero. The fact that a hydrogen-rich compound can superconduct at a temperature you might encounter on a cold winter night, even if only under crushing pressure, represents a genuine milestone in condensed matter physics.

The claim isn't new in the sense that LaH₁₀ results first emerged from the groups of Mikhail Eremets and Russell Hemley around 2018–2019. What the 2026 review emphasizes is that this result has survived the gauntlet of independent replication — a critical distinction in a field that has been rocked by controversy, retraction, and fierce debate over reproducibility.

How Our Simulation Approaches This

At AI Future Lab, we run AI-augmented computational analyses that attempt to predict superconducting properties from first principles — or more precisely, from a hybrid framework that combines physics-informed neural networks with classical Eliashberg theory inputs. We should be upfront: our approach is not a replacement for full-scale density functional theory (DFT) paired with density functional perturbation theory (DFPT), which remain the gold standard for ab initio superconductivity predictions. Nor is it a substitute for experimental measurement.

What our simulation does is approximate the electron-phonon coupling landscape of a given crystal structure at specified pressures, using a model trained on a curated dataset of known superconductors and their computed phonon spectra. It estimates the electron-phonon coupling constant (λ), the logarithmic average phonon frequency, and from these derives a Tc estimate via a modified Allen-Dynes equation. The model also flags structural stability — whether the predicted phase is thermodynamically stable, metastable, or dynamically unstable based on the phonon dispersion.

This means our results carry inherent approximation errors. We are interpolating within a learned space, not solving the full quantum mechanical problem from scratch for each new system. We state this clearly because intellectual honesty matters more than impressive-sounding numbers.

What Our Analysis Found

For LaH₁₀ in the sodalite-like clathrate structure (space group Fm̄3m), our simulation at 170 GPa returns the following:

• Predicted Tc: 250 K
• Electron-phonon coupling constant (λ): 2.1
• Dominant phonon mechanism: High-frequency hydrogen-dominated modes — specifically H-stretching and H-bending vibrations in the 100–200 meV energy range — within the clathrate cage structure, where lanthanum atoms act as electron donors, chemically pre-compressing the hydrogen sublattice into a near-symmetric bonding configuration.
• Structural stability: Metastable. The Fm̄3m phase shows no imaginary phonon frequencies at 170 GPa (dynamically stable), but it sits above the convex hull of competing phases, meaning it is not the thermodynamic ground state and may require specific synthesis pathways to access.
• Confidence level: Medium.

The 250 K prediction sits 10 K below the reported experimental value of 260 K — a discrepancy of about 4%. In most areas of computational materials science, this would be considered excellent agreement. In the specific context of high-Tc superconductor predictions, where uncertainties in the Coulomb pseudopotential (μ*), anharmonic corrections, and quantum nuclear effects can each shift Tc by 10–30 K, we classify this as a partial match.

⚠️ Partial Match: Reading the Gap

A 10 K gap between prediction and experiment deserves honest scrutiny from both sides.

Why our prediction might be low: Our model uses a semi-empirical Coulomb pseudopotential μ* ≈ 0.10–0.13, which is standard but potentially too large for hydrogen-dominant systems where the relevant energy scales are exceptionally high. Full anharmonic Eliashberg calculations — which account for the fact that hydrogen atoms in these structures undergo massive quantum zero-point motion — have been shown by groups like Errea et al. to significantly reshape the phonon spectrum and generally increase the predicted Tc relative to harmonic approximations. Our model captures some of this through its training data but does not perform explicit anharmonic phonon calculations. This alone could account for the missing 10 K.

Why the experimental value itself carries uncertainty: Measuring superconductivity at 170+ GPa is extraordinarily difficult. The sample volume inside a diamond anvil cell is microscopic — often just a few micrometers across. Determining the precise onset of superconductivity requires careful analysis of resistance drops and, ideally, magnetic susceptibility measurements (the Meissner effect), which are fiendishly hard at these pressures. Different measurement protocols and criteria for defining Tc onset versus midpoint versus zero-resistance temperature can introduce spreads of 5–15 K. The pressure itself is not perfectly uniform across the sample and is measured indirectly, often via ruby fluorescence or X-ray diffraction of a calibrant, each carrying its own uncertainty window.

Pressure range matters: The paper cites 170–190 GPa, a 20 GPa window. Our simulation was run at the lower bound of 170 GPa. Tc in LaH₁₀ is known to be pressure-dependent, and calculations by multiple groups show a Tc maximum somewhere near 200–250 GPa depending on the functional used. Running our simulation at 190 GPa might close the gap further. We opted for 170 GPa as a conservative starting point.

The metastability flag in our results also aligns with experimental reality: synthesizing LaH₁₀ requires heating lanthanum and hydrogen (or ammonia borane as a hydrogen source) under extreme compression, and the resulting phase does not always persist upon decompression. This is not a material you can hold in your hand.

What This Tells Us About Room-Temperature Superconductivity

LaH₁₀ at 260 K is tantalizingly close to room temperature. So why can't we just declare victory?

Because pressure is doing most of the heavy lifting. At 170–190 GPa, hydrogen is forced into a metallic, densely packed arrangement that simply does not exist at ambient conditions. The very phonon modes that drive the strong coupling — those high-energy hydrogen vibrations — depend on interatomic distances that only extreme compression can achieve. Remove the pressure, and the structure collapses. The superconductivity vanishes.

For true ambient-pressure room-temperature superconductivity, we would need a material where the lattice provides its own "chemical pressure" — where the crystal structure intrinsically constrains hydrogen (or another light element) into a metallic, high-frequency-phonon configuration without external force. This is the dream behind ternary and quaternary hydride research: adding a third or fourth element to stabilize hydrogen-rich cages at lower pressures. Progress is real but incremental. Compounds like CaBeH₈, LaBeH₈, and various carbon-sulfur-hydrogen systems are being explored computationally, but none have yet achieved the simultaneous combination of high Tc, low pressure, and confirmed experimental synthesis.

The reproducibility crisis that has shadowed this field — from the retracted Nature paper on carbonaceous sulfur hydride to the ongoing debates around nitrogen-doped lutetium hydride — makes independent validation of results like LaH₁₀ all the more important. The fact that multiple groups across different continents have confirmed superconductivity in LaH₁₀ near 250–260 K is what distinguishes it from more controversial claims. Reproducibility is not a bureaucratic checkbox; it is the load-bearing wall of scientific knowledge.

Our Evolving Simulation

We view the 4% discrepancy not as a failure but as a calibration signal. Our model's training set is continually expanding as new DFT studies, experimental results, and anharmonic phonon calculations are published. Specifically, we are working on three refinements:

1. Anharmonic corrections: Incorporating learned anharmonic renormalization factors for hydrogen-dominant systems, trained on the growing body of stochastic self-consistent harmonic approximation (SSCHA) calculations.
2. Pressure-dependent sweeps: Generating Tc-vs-pressure curves rather than single-point predictions, which will allow more meaningful comparison with experimental data reported across pressure ranges.
3. Stability depth analysis: Moving beyond a binary stable/metastable label to estimate the energy above the convex hull quantitatively, giving a clearer picture of synthetic accessibility.

The gap between 250 K and 260 K is small enough to be encouraging and large enough to keep us honest. As more data flows in from high-pressure labs and computational groups worldwide, we expect our predictions to sharpen. The goal has never been to replace experiment or full-scale theory — it is to build a fast, transparent, and increasingly reliable tool for navigating the vast space of candidate superconductors. LaH₁₀ is our benchmark. The gap today is our roadmap for tomorrow.