Synthesis and stability of high-$T_c$ LaH$_{10\pmδ}$ films at high pressures

Abstract: High-pressure hydrides hold the record for the highest superconducting critical temperatures across all classes of superconductors. Currently lanthanum decahydride, LaH$<em>{10}$, exhibits the highest critical temperature among binaries, with $T_c \approx$ 250 K at pressures between 140-180 GPa. Here, we report the synthesis of LaH$</em>{10\pmδ}$ films in two DACs at pressures of 168 GPa and 176 GPa via in situ laser heating of elemental lanthanum films with ammonia borane (NH$<em>3$BH$_3$) as the hydrogen donor. The high-symmetry fcc lanthanum sublattice (space group $Fm\bar3m$) is resolved using synchrotron X-ray diffraction, with unit cell parameters in excellent agreement with previous studies on bulk samples. We provide confirmation of high-$T_c$ superconductivity in LaH$</em>{10\pmδ}$ with highest $T_c$ of 247 K at 176 GPa evidenced in electrical measurements. The characteristic suppression of superconductivity is observed in magnetic fields. Furthermore, combined diffraction and electrical measurements reveal remarkable temporal stability of both the crystal structure and the high-$T_c$ superconductivity over the full measurement period of about 300 days post laser heating. Our work establishes film precursors using physical vapour deposition (PVD) techniques as a practical route to hydride formation, opening a pathway toward the controlled synthesis of promising ternary hydrides and the integration of micro-fabricated device geometries in diamond anvil cells.

• The paper demonstrates a film-based synthesis of cubic Fm3̅m-LaH10±δ in diamond anvil cells, achieving a superconducting transition onset of 247 K at 176 GPa.
• High-resolution synchrotron XRD, transport, and magnetic-field measurements confirm the phase assignment and indicate zero resistance, upper critical fields of 116–167 T, and coherence lengths near 1.4–1.7 nm.
• The paper finds that LaH10±δ retained its crystal structure and superconductivity for approximately 300 days when excess molecular hydrogen remained present, challenging reports of rapid dehydrogenation to LaH3.

Overview

This paper reports the synthesis of superconducting lanthanum decahydride films, LaH$_{10\pm\delta}$, in diamond anvil cells (DACs) at 168 GPa and 176 GPa, using elemental lanthanum films deposited by physical vapour deposition directly onto the diamond anvil and ammonia borane as the hydrogen donor. Synchrotron X-ray diffraction confirms the high-symmetry $Fm\bar{3}m$ fcc lanthanum sublattice, and four-point electrical transport measurements show a superconducting transition with an onset critical temperature of 247 K at 176 GPa — among the highest values reported for any binary compound. Beyond establishing the synthesis route, the work addresses a live controversy: recent NMR measurements claimed diffusion-driven dehydrogenation of LaH$_{10}$ to LaH$_3$ over roughly 70 days (2608.18865), whereas combined diffraction and transport studies on bulk samples reported stability exceeding five years. The present study provides independent evidence for the latter position through a ~300-day combined XRD and transport monitoring campaign.

Film-based synthesis methodology

The samples were prepared in MP35N DACs with 50 µm culet anvils, five tungsten–gold bilayer electrodes patterned through a shadow mask onto one anvil, and evaporated lanthanum films of thickness 262 nm (DAC 1) and 213.5 nm (DAC 2). Purified ammonia borane served both as hydrogen source and pressure medium, and laser heating was performed with a 1070 nm Yb-fiber laser in short pulses while monitoring the four-point resistance; heating was halted upon an irreversible resistance increase signalling hydride formation.

The authors argue that film precursors offer several advantages over bulk synthesis routes: a high hydrogen-to-metal ratio favouring high stoichiometries, reliable electrical contact to pre-patterned electrodes, and controlled sample geometry. They report that the success rate in maintaining electrical contact during loading exceeds 90%, compared with lower rates for hand-cut foil electrodes used in traditional approaches. This claim is based on the authors' own experience rather than a systematic comparison, but it is consistent with prior successes using film precursors for La$_4$H$_{23}$ and H$_3$S synthesis.

Structural characterisation

Synchrotron XRD on three beamlines (I15 at Diamond Light Source; ID11 and ID27 at the ESRF) confirmed the $Fm\bar{3}m$ phase in both cells, with Pawley-refined lattice parameters of $a = 5.1254(4)$ Å at 168 GPa and $a = 5.0909(2)$ Å at 176 GPa, in good agreement with bulk studies and DFT equation-of-state calculations. Because only the lanthanum sublattice diffracts measurably, the hydrogen stoichiometry cannot be determined from unit cell volume alone; the authors accordingly designate the phase LaH$Fm\bar{3}m$0 and explicitly note they cannot exclude variations in hydrogen content relative to other studies.

High-resolution spatial XRD mapping over a 25 × 25 µm² grid with 0.7 µm beam diameter proved essential for assigning transport behaviour to specific phases. DAC 1 contained $Fm\bar{3}m$1-LaH$Fm\bar{3}m$2 in the laser-heated region, localised $Fm\bar{3}m$3-LaH$Fm\bar{3}m$4 at its boundary, and $Fm\bar{3}m$5-LaH$Fm\bar{3}m$6 further out — a stoichiometry gradient attributed to lateral thermal gradients during heating and the multistep hydrogen release from ammonia borane. Two heating cycles in DAC 2 produced near-complete transformation to the $Fm\bar{3}m$7 phase contacting all five electrodes. An unexplained peak at $Fm\bar{3}m$8, tentatively assigned to the (111) reflection of cubic boron nitride formed from decomposed ammonia borane, and two further unindexed reflections in DAC 1 remain unassigned — a limitation the authors acknowledge.

Notably, Raman mapping revealed free molecular H$Fm\bar{3}m$9 retained above the hydride phase for over 300 days in both cells, indicating that excess hydrogen does not escape the chamber on these timescales. This observation becomes central to the stability discussion below.

Electrical transport and evidence for superconductivity

Resistive transitions were observed with onsets at $_{10}$0 K (DAC 1) and $_{10}$1 K (DAC 2), consistent with bulk LaH$_{10}$2 literature values and theoretical predictions for the quantum-anharmonically stabilised $_{10}$3 structure. In DAC 1, the transition showed a sharp drop by a factor of 28 followed by a second transition near 70 K before reaching zero resistance; the authors tentatively assign this low-temperature feature to the percolating $_{10}$4-LaH$_{10}$5 phase, since the spatial maps show it forms a continuous pathway between voltage probes while the localised $_{10}$6 regions do not. After a second heating cycle, DAC 2 exhibited a sharp drop by a factor greater than $_{10}$7 with a transition width of about 12 K ($_{10}$8), reaching zero resistance within noise below 235 K in all measured electrode configurations.

The zero-resistance state was quantified carefully. The RMS noise floor of 64–74 µΩ matches the preamplifier input noise, and comparison against an equivalently dimensioned high-purity copper reference (RRR ≈ 5760) shows the measurement resolution is more than two orders of magnitude finer than copper's resistance (~9 mΩ) at 235 K. The authors candidly note that resolving resistances below the residual resistance of the purest metals at the lowest temperatures remains experimentally challenging, and that improving resolution via larger excitation currents risks common-mode leakage artefacts — indeed, a small positive offset observed below 60 K in DAC 1 field measurements was traced to imperfect common-mode rejection from the current leads rather than the sample.

Magnetic fields up to 14 T suppressed $_{10}$9 by about 15 K. Linear $_3$0 near $_3$1 yielded slopes of −0.95 and −0.92 T/K, and GL/WHH extrapolations gave $_3$2 ranges of 116–167 T, coherence lengths $_3$3–1.7 nm, and BCS Fermi velocities of $_3$4–$_3$5 m/s — all consistent with bulk LaH$_3$6. WHH estimates fall about a factor of three below the Pauli limit (~455 T), indicating orbital pair-breaking dominance, as found across other high-$_3$7 hydrides. The authors appropriately caution that extrapolations may miss strong-coupling or multiband effects at low temperature, that direct verification would require fields beyond current DAC capabilities, and that without a measured penetration depth the type-II classification cannot be rigorously established through $_3$8.

Regarding transition broadening in field, the authors show that the apparent non-monotonic width evolution under the conventional 90%/10% criteria arises from field suppression of an inhomogeneity-related step near 240 K; using 90%/50% criteria the width increases monotonically, as expected for vortex dissipation in a type-II superconductor. Since 14 T is under 10% of the extrapolated $_3$9, more pronounced broadening at higher fields is anticipated but untested here.

Temporal stability

The central stability result is that neither the crystal structure nor the superconductivity degraded over approximately 300 days post-synthesis. The $_4$0 phase persisted in XRD up to 243 days after heating, with unit cell volume changes of at most 1.1 ų (DAC 1) and 0.5 ų (DAC 2) attributable, via the DFT equation of state, to pressure decreases of ~5–6 GPa in the cells. Spatial phase distributions were essentially unchanged between mappings taken ~60 days apart. Transport measurements showed stable $_4$1 throughout, with the normal-state resistance actually decreasing over time — behaviour the authors attribute to room-temperature annealing and which argues against decomposition to semi-metallic LaH$_4$2, which would raise the normal-state resistance.

These findings directly contradict the NMR study reporting progressive dehydrogenation toward LaH$_4$3 within ~70 days. The authors identify a potentially decisive experimental difference: in their cells, and in the multi-year bulk study of Minkov et al., excess molecular hydrogen remained in the chamber after laser heating, whereas no H$_4$4 vibron was observed in the NMR study. They suggest that chemical equilibrium with an excess hydrogen reservoir may be required for stabilisation, and note that the NMR study lacked corroborating XRD confirmation of the $_4$5 phase. However, the authors concede several caveats: qualitative map comparisons cannot rule out formation of amorphous hydride phases; contributions from hydrogen diffusion into the sample to the volume increase cannot be excluded; and integrated patterns from different beamtimes may sample slightly different positions under pressure gradients. Resolving the discrepancy fully would require combined XRD, Raman, and NMR measurements on samples prepared exactly as in the NMR study (laser-heated LaH$_4$6 + ammonia borane mixtures) — an experiment this paper leaves open.

Limitations and open questions

Several limitations qualify the results. Hydrogen stoichiometry is inferred only indirectly, since the hydrogen sublattice is invisible to diffraction, motivating the LaH$_4$7 notation. The c-BN assignment and the identity of two unindexed reflections in DAC 1 are tentative. The low-temperature transition in DAC 1 assigned to $_4$8-LaH$_4$9 is percolative inference rather than direct demonstration, and targeted syntheses of that phase are called for. Laser-heating temperatures could not be measured by black-body pyrometry because the reaction occurred below the glowing threshold. Finally, the stability conclusion covers ~300 days — far shorter than the multi-year claims it supports, though comfortably beyond the 70-day decomposition window claimed in the NMR work.

Conclusion

This work demonstrates that PVD-deposited elemental films are a viable, reproducible precursor platform for synthesising high-$_{23}$0 hydrides in DACs, achieving $_{23}$1-LaH$_{23}$2 with $_{23}$3 K at 176 GPa and rigorously quantified zero-resistance behaviour. The combined diffraction and transport stability data reinforce the thermodynamic stability of LaH$_{23}$4 in an excess-H$_{23}$5 environment and sharpen the terms of the disagreement with the NMR dehydrogenation report around the presence or absence of a residual hydrogen reservoir. The demonstrated compatibility of film methods with micro-fabricated electrode geometries provides a concrete foundation for ternary hydride synthesis, where precise control of precursor stoichiometry will be critical.