WorksPhysics and Astronomy
Synthesis and stability of high- LaH 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, exhibits the highest critical temperature among binaries, with 250 K at pressures between 140-180 GPa. Here, we report the synthesis of LaH films in two DACs at pressures of 168 GPa and 176 GPa via in situ laser heating of elemental lanthanum films with ammonia borane (NHBH) as the hydrogen donor. The high-symmetry fcc lanthanum sublattice (space group ) 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- superconductivity in LaH with highest 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- 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.
(Dated: Thursday 20ᵗʰ August, 2026)
Introduction
Since the discovery of high-temperature superconductivity in sulfur hydride, , in 2015 [?], numerous compressed metal hydrides have been found to exhibit very high superconducting critical temperatures, [?, ?, ?, 4, 5, 6, 7, 8, 9]. The current record , among binaries, is held by lanthanum decahydride, , with at pressures above [?, ?]. In , lanthanum atoms form an fcc lattice and are surrounded by clathrate hydrogen networks that are stabilised by the charge transfer from electropositive atoms to promote the dissociation of molecular units and the formation of hydrogen cages [10, 11]. This yields a high hydrogen-derived density of states at the Fermi level coupled to high-frequency phonon modes, crucial for realising high- superconductivity in and many other hydride systems [?, ?, 5, 6, 7, 8, 9, 12].
Questions nevertheless remain regarding the long-term stability of - under sustained compression. Initially, first-principles calculations within the harmonic approximation predicted the clathrate phase to be dynamically stable above [13]. Later, quantum anharmonic effects were shown to dynamically stabilise - down to [14]. This was consistent with experimental observations of the - phase persisting to , below which a distortion to a lower-symmetry structure occurs, accompanied by a rapid reduction in upon further decompression [9]. Recent nuclear magnetic resonance (NMR) measurements have raised questions concerning the long-term stability of , reporting a highly diffusive state of the hydrogen sublattice at room temperature which was found to drive progressive dehydrogenation of the sample toward lanthanum trihydride, , over the course of about 70 days [15]. It should be noted that the NMR study did not conduct X-ray diffraction (XRD) measurements to confirm the presence of -. In contrast, diffraction and transport studies of bulk samples have shown stability of the crystal structure and superconducting critical temperature of over timescales exceeding four years [16]. Further investigations combining structural and transport probes are therefore desirable to resolve this controversy.
The synthesis and characterisation of hydrides are challenging owing to the extreme pressures required. Typically hydrides are synthesised via in situ laser heating of bulk elemental precursors with a hydrogen donor compound, such as pure , ammonia borane or paraffin oil, at high pressures inside a diamond anvil cell (DAC). While many experimental successes have been reported, the synthesis of clean single-phase hydride samples remains challenging owing to multiple competing phases that may coexist at similar pressures. Phase coexistence in the lanthanum hydride system has been evidenced in experimental studies [17] and corroborated by calculations that identify competing thermodynamically stable phases [11, 13]. Partial or inhomogeneous hydrogenation during laser heating coupled with pressure and thermal gradients can further promote the formation of multiple phases within the same sample. Nevertheless, targeting the synthesis of single-phase hydrides is important for accurate characterisation of intrinsic structural and superconducting properties, and highlights a pertinent challenge for the field.
One promising avenue toward this goal is the use of film precursors deposited directly onto the diamond anvil. Such methods have allowed successful synthesis of superconducting films at 100 GPa by laser heating a lanthanum film with ammonia borane as the hydrogen donor [5]. Film electrodes have been used to study the superconductivity in bulk [4] and in elemental yttrium films at high-pressures [18]. The use of films in hydride synthesis ensures a high ratio of hydrogen to metal, thus promoting the formation of hydrides with a high hydrogen stoichiometry. Furthermore, deposition directly onto the diamond culet simplifies sample preparation, ensures reliable contact to pre-patterned electrodes on compression, and enables better control over the initial sample geometry. While certain challenges remain, such as effective laser heating of a thin sample with strong thermal link to the anvil, and weaker XRD intensities relative to bulk samples, film techniques have greatly improved the reproducibility and success rate of our DAC preparation and synthesis attempts. Based on our experience, we estimate that the success rate in maintaining electrical contact to film samples during DAC loading and pressurising exceeds 90%, in contrast to the lower success rates typically achieved using traditional methods, namely positioning bulk samples onto hand-cut electrodes prepared from a metallic foil [19].
Here, we report the first synthesis of high- - films at pressures above 160 GPa, formed by laser heating elemental lanthanum films with purified ammonia borane as the hydrogen donor. The target phase was successfully obtained in two DACs prepared for this study. In section II we present a structural and electrical characterisation of the synthesised films, and provide strong evidence for superconductivity in - with detailed analysis of the zero-resistance state. In section III we present a stability study of our films by combining high-resolution X-ray diffraction and electrical transport measurements, addressing key questions regarding the long-term stability of LaH.
LaH10 films: Structural and electronic properties
Structural characterisation
Two DACs were prepared with elemental lanthanum films and ammonia borane at high-pressures as detailed in Section V. The successful synthesis of the high-symmetry -LaH phase in both DACs is evidenced using synchrotron X-ray diffraction measurements presented in Figure 1(a). Unit cell parameters were extracted for the phase from Pawley refinements [20] and were found to be Å ( Å) in DAC 1 at 168 GPa and Å ( Å) in DAC 2 at 176 GPa. The azimuthal projection of the detector image corresponding to the frame employed for the Pawley refinement of DAC 2 is shown in Figure 1(b). For our polycrystalline films, typically spotty diffraction patterns were observed as in Figure 1(b) similar to those observed in previous studies of laser-heated LaH samples [?, 9]. The beam diameter was 0.7 µm in full width at half maximum (FWHM), and hence only a small number of crystallites were illuminated, and relative peak intensities for the phase often deviated from those expected from a powder average. Hence, the Pawley refinement procedure was used to extract unit cell parameters [20]. The refined unit cell volumes are in good agreement with prior structural studies on bulk LaH samples in Refs. [?, ?, 9, 17, 21] as demonstrated in Figure 1(c). Since only the lanthanum sublattice is detected in diffraction, there is an uncertainty in inferring the hydrogen stoichiometry from the measured unit cell volume, and we cannot rule out that there is a variation in hydrogen content between our samples and those synthesised in different studies via different formation pathways. We therefore refer to the synthesised hydride as LaH, where represents the deviation from the ideal LaH stoichiometry.

Figure 1. Synchrotron X-ray diffraction measurements, measured 195 days and 103 days after laser heating of DACs 1 and 2 respectively. (a) Integrated diffraction patterns demonstrating the presence of -LaH in both DACs. Data are shown as open black circles, and red solid lines are Pawley refinements. Intensities have been normalised to the (111) reflection intensity for comparison. Reflections that could not be indexed are indicated with an asterisk (*). (b) Azimuthal projection of the detector image in frame 407 in DAC 2, corresponding to the integrated pattern in (a). Blue arrows indicate the reflections. (c) Volume per lanthanum atom as a function of pressure, including data for LaH samples synthesised in this work together with those in Refs. [?, ?, 9, 17, 21]. The DFT equation of state from Ref. [17] is shown as the solid blue line. Experimental data for -La from Ref. [26] are shown as the green inverted triangles. (d) and (e) Spatial XRD mappings and photomicrographs of both samples. White open squares on the spatial maps correspond to the positions of the frames of the respective integrated patterns in (a). Blue regions on the spatial maps correspond to intensity distribution of the -LaH (311) reflection, and the red and black regions correspond to the (101) reflections of -LaH and -LaH, respectively. The electrode maps are shown in yellow, corresponding to the distribution of the -Au (111) reflection intensity.
In order to assign the electrical properties of the hydride films to the crystal structures present within the sample, high-resolution spatial X-ray diffraction mapping was performed for both DACs on the ID27 beamline at the ESRF as shown in Figure 1(d) and (e). Spatial maps for both DACs were performed over a µm grid, with 1 µm step size and with a beam diameter of 0.7 µm at FWHM. Photomicrographs of each sample after laser heating are shown below each respective spatial X-ray mapping. In both DACs, the intense (111) reflection corresponding to elemental Au was observed, which traced clearly the five electrodes as shown in Figure 1(d) and (e). After one laser-heating cycle, the film in DAC 1 consisted of a phase mixture of cubic -LaH in the laser-heated region, as shown by the blue region in Figure 1(d), with highly localised regions of hexagonal -LaH on the boundary of the phase (black regions). Hexagonal LaH impurities have been observed in almost all prior LaH syntheses reported in the literature [?, 9, 17], and is consistent with theoretical structure searches that predict the stability of this phase under similar conditions [22, 23]. Further away from the laser-heated region we identified body-centred tetragonal -LaH, shown as the red region in Figure 1(d). We note that the spatial maps were constructed using the intensity of the LaH (311) reflection and the LaH (101) and LaH (101) reflections due to their separated positions, enabling unambiguous mapping of the phase distributions. The coexistence of , , and hydrides at the synthesis pressure of DAC 1 is in agreement with the findings of Laniel et al., in Ref. [17]. In both DACs, a diffraction peak was observed at which appeared to be localised to the dark laser-heated region on each sample, visible in the integrated pattern for DAC 1 in Figure 1(a). We cannot explain this peak with elemental lanthanum or any of the known lanthanum hydride phases [17], but we do find that the peak is consistent with the (111) reflection of cubic boron nitride (c-BN) at this pressure [24] which could explain the origin of the dark region on each sample. It is plausible that c-BN formed directly above the sample on laser heating via the complete decomposition of ammonia borane where the highest temperatures were achieved, although we acknowledge that this is a tentative assignment. Two further reflections observed in DAC 1 at and could not be indexed, and are indicated with an asterisk in Figure 1(a).
The decrease in hydrogen stoichiometry with increasing distance from the laser-heated region likely arises from a lateral thermal gradient across the La film during laser heating, and hence a variation of hydrogen excess across the sample. Ammonia borane (AB) undergoes a multistep decomposition and hydrogen release as temperature increases [25]. Consequently, hydrogen release from the surrounding AB during laser heating is likely to be spatially non-uniform due to these temperature gradients across the sample. Interestingly, Raman spectroscopy mapping of the molecular H vibron signal across the sample in DAC 1 (measured 146 days after laser heating, Figure S1) reveals that free molecular hydrogen, identified by its characteristic vibron, remained present in the region above the LaH phase likely where the highest temperature and greatest hydrogen excess were achieved during laser heating. The vibron remained present after 301 and 370 days post-synthesis in DAC 1 and 2 respectively (Figure S2 and Figure S3), suggesting that hydrogen is trapped in a region above the phase and surrounded by partially decomposed ammonia borane, and does not diffuse out of the sample chamber, e.g. through the diamond anvils or the gasket, over such timescales. Further laser heating at various positions over the film would likely have transformed to , although this was not done for DAC 1 to reduce the risk of anvil failure before electrical measurements could be carried out.

Figure S1. (a) Raman pressure map for DAC 1, with pressure estimated at various positions over the culet using the Akahama scale [37]. Larger outer circle marks the approximate culet edge, while the inner smaller circle marks the dark region observed on the sample after laser heating. (b) Raman map of the molecular hydrogen vibron intensity in DAC 1. Pressure and vibron mapping collected 146 days after laser heating. (c) Photomicrograph of the sample in DAC 1 in the same orientation and on the same scale as in (a) and (b). (d) X-ray diffraction mapping of the sample in DAC 1, with the three identified hydride phases labelled.

Figure S2. (a) Raman spectra for DAC 1 showing the first order diamond edge and hydrogen vibron immediately after laser heating (blue curve) and after 301 days post laser heating (red curve). Inset shows the sample and Raman laser spot ( nm). (b) First derivative minima used to measure the pressure according to the Akahama scale [37]. (c) Variation in the wavenumber of the hydrogen vibron over the measurement period. Estimates of the pressure from the hydrogen vibron, , were made according to Ref. [42].

Figure S3. Raman spectra of DAC 2 showing the diamond edge before laser heating, immediately after the second laser heating, and 370 days after the second laser heating. Curves have been offset for clarity. (b) The derivative of the spectra in (a) to determine the pressure from the derivative minima according to the Akahama calibration [37]. The pressure decreased by about 5 GPa after 370 days post laser heating. (c) Evidence of the hydrogen vibron, confirming the long-term retention of excess H in DAC 2. The pressure estimated from the hydrogen vibron, , using Ref. [42] is indicated.
Two laser-heating cycles were performed for the film in DAC 2, which resulted in an almost complete transformation to the - phase, which made contact with all five electrodes as shown in Figure 1(e). Localised regions of - on the outer boundary of the phase were also observed, as in DAC 1. No evidence of - was observed in DAC 2 confirming sufficient hydrogen excess around the sample during laser heating. Representative integrated patterns demonstrating the presence of each phase within the DACs are shown in supplementary Figure S4 and Figure S5.

Figure S4. Integrated XRD patterns at different positions on the sample in DAC 1 at 168 GPa, providing evidence for and phases in addition to the high- phase. (a) and (b) demonstrate the presence of -LaH from measurements on ESRF beamlines ID11 (134 days after laser-heating) and ID27 (195 days after laser-heating) respectively. The strong (101) reflection of the phase present between the first two reflections has been observed in other studies of LaH [?, 9]. Inset spatial map of the (101) reflection’s intensity shows the localised region of the phase, and red squares indicate the frame positions where integrated patterns were measured. From the ID11 measurement, we estimate the unit cell parameters to be Å, Å, Å, , but do not perform full structure refinement owing to poor resolution of the phase. (c) and (d) Evidence for the -LaH phase in DAC 1. Inset spatial maps show the distribution of the phase by mapping of the intense (101) reflection. From the ID11 measurements of the phase, we estimate Å, Å, Å, .

Figure S5. Evidence for the - phase in DAC 2 at 176 GPa. Measurements performed 43 days after the second laser heating. Inset spatial maps show the distribution of the phase by mapping of the intense (101) reflection at . We estimate Å, Å, Å, .
Electrical transport measurements
Prior to laser heating, the metallic character of the elemental lanthanum films was verified in measurements of resistance versus temperature (Figure S6). After laser heating, the normal state resistances increased for both DACs, and the onset of resistive transitions were observed at and in DACs 1 and 2 respectively as shown in Figure 2, as extracted from cooling cycles at rates of between -. A careful analysis of the extrinsic thermal hysteresis observed in our measurement systems is provided in Figure S7. The observed values of are in good agreement with those reported in the literature for bulk , shown in the inset of Figure 2, and with the value predicted for the - structure in theoretical studies [14]. In DAC 1, there is a sharp resistance drop by a factor of 28 at 245 K, with a residual normal state contribution attributed to - that was observed in the XRD measurements. There appears a second resistive transition in DAC 1 at before the resistance falls to zero within the noise. A complete percolation pathway between the voltage probes necessary for zero resistance in DAC 1 is realised for the phase but not for the localised regions of the phase, evidenced in the XRD spatial maps in Figure 1(d). We therefore tentatively assign this lower transition to the phase, further supported by a recent study reporting a higher for the phase, at around 160 K at 140 GPa [27]. Further targeted syntheses of -LaH would be desirable. The electrode configuration used to measure the curve for DAC 1 in Figure 2 was chosen such that the voltage probes were closest to the -LaH phase. Measurements in various contact configurations for DAC 1 are shown in Figure S8(a).

Figure 2. Four-point resistance versus temperature measured for both DACs 9 days after their respective laser heating. Measurements were taken in cooling cycles with cooling rate 0.5-1 K min, and with excitation current of 30 A. Lower panel shows the zero resistance region. Inset shows versus pressure to compare of the films with the reported values for bulk LaH in Ref. [?]. Spatial XRD mappings for each DAC to the left of the main panel indicate the electrode configuration used to measure each curve.

Figure S6. Resistance measurements of the elemental lanthanum films in both DACs prior to laser heating, showing metallic behaviour. Measurements taken with excitation current of in warming runs at rates between -. The residual resistance ratios (), a measure of the quality of the films, are indicated for the respective samples.

Figure S7. Hysteresis between cooling and warming runs measured at different temperature sweep rates for both DACs, using two different measurement probes (probes 1 and 2). Zero-field measurements were performed using probe 1, where the DAC is surrounded by He gas at 1 bar. A Cernox thermometer was mounted on the probe cold plate, to which the DAC is attached. A 3.5 K–4.5 K temperature hysteresis in is observed with this probe. Measurements in magnetic field were performed using probe 2, which is custom designed for a magnet cryostat, with the sample under vacuum. In this case, a Cernox thermometer was mounted directly on the DAC seat at the base of the diamond directly next to the sample, resulting in a much smaller temperature hysteresis between cooling and warming. Owing to the improved thermal coupling between the thermometer and the sample, the measured with probe 2 is expected to provide the most accurate estimate of the intrinsic superconducting transition temperature. This value is in excellent agreement with the measured during cooling using probe 1. For this reason, all measurements acquired using probe 1 and presented in the main text are shown for the cooling runs only. Arrows on each curve indicate the cooling and warming directions.

Figure S8. Resistive transitions in various electrode configurations. (a) Measurements for DAC 1, where the current/voltage contacts and days since laser heating are indicated in the legend. The inset spatial XRD map illustrates the contact numbering relative to the observed phases, where the blue region is -LaH, red is -LaH and black is -LaH. (b)–(d) Measurements for DAC 2 after first and second laser heating cycles in three different measurement configurations. The electrode configuration used for each measurement is shown on the XRD spatial map in the inset of each plot. The pressure after the first and second laser heating of DAC 2 was 171 GPa and 176 GPa respectively. Successive laser heating cycles caused the normal state resistance to increase, and the resistive transition to sharpen, with an onset at 247 K after the second laser heating. Zero resistance within the noise floor is observed in all measured configurations of DAC 2, consistent with the phase forming a continuous pathway between all electrodes.
Following the first laser heating of DAC 2, a broad transition to a zero-resistance state was observed with an onset at K and transition width () of about 30 K (red curve in Figure S8(b)). After the second laser heating of DAC 2, the normal state resistance increased further and a sharp resistance drop by a factor was observed at K as shown by the red curve in Figure 2. The transition width was about 12 K (). We attribute the further increase in the normal state resistance to the transformation of a larger volume fraction of the elemental lanthanum film to the -LaH phase on successive laser heating cycles. After the second laser heating of DAC 2, the resistance reaches zero, within the noise of the measurement, below 235 K. Zero resistance is observed in all measured electrode configurations of DAC 2 (Figure S8(b)–(d)), consistent with the phase forming a continuous pathway between all electrodes as evidenced by the XRD spatial map in Figure 1(e).
In the zero-resistance regime, the measured RMS resistance noise with a 30 A excitation current was found to be and for DAC 1 and DAC 2, respectively (Figure S9(a) and (b)). This is consistent with the dominant input noise of the SRS SIM921 preamplifier (JFET LSK489), which is 60 for a 30 A excitation and with the 1 s time constant used in our measurements. We compare this noise floor for DAC 2 with the expected resistance of a high-purity copper sample with residual resistivity ratio RRR [28], with the same dimensions, providing a stringent benchmark for the zero-resistance state. The copper resistance is estimated from its known resistivity, [28], and the measured dimensions of the sample in DAC 2 (length m, width m, thickness nm) according to . To obtain a conservative lower-bound estimate of the copper resistance, we use the deposited lanthanum thickness at ambient pressure and neglect the reduction in sample thickness under pressure. We find that at 235 K, just below where the measured resistance reaches zero within the noise floor, the expected resistance of copper is approximately 9 m. Our noise floor is therefore more than two orders of magnitude smaller than the resistance of copper at this temperature (Figure S9(c)), and demonstrates that our measurement resolution is sufficient to clearly distinguish the superconducting zero-resistance state from the lowest resistances achieved in highest purity copper. Below approximately 35 K however, the residual resistance of copper () falls below our noise floor (inset of Figure S9(c)). This illustrates that demonstrating a zero-resistance state at lowest temperature with a resistance resolution below the residual resistance of even the highest-purity metals remains experimentally challenging. In principle, the resistance resolution could be improved by applying an excitation current 50–100 times larger. However, while increasing the excitation current improves the voltage sensitivity, it could also introduce systematic effects associated with the high current density, including common-mode leakage due to the finite contact resistance of the current leads.

Figure S9. Quantitative confirmation of zero resistance in LaH. (a) and (b) Resistance curves as a function of temperature with the corresponding histograms of resistance residuals in the zero-resistance state for DAC 1 and DAC 2. The RMS noise is extracted as , where is the standard deviation obtained from a Gaussian fit to the distribution. The measured RMS noise is consistent with the input noise of the SRS SIM921 preamplifier (JFET LSK489), which we calculate to be with excitation and a time constant. The non-Gaussian tails observed in both distributions are not expected from instrumental noise and indicate additional contributions from the measurement environment. (c) Comparison of the resistance of DAC 2 and high-purity copper as a function of temperature. The copper resistance was estimated using the resistivity from [28] and the measured dimensions of the LaH region in DAC 2 (, , and ) according to . The residual resistivity ratio of the high-purity copper sample is .
Upper critical field
In order to provide further evidence for superconductivity in the LaH films, resistance measurements were performed in magnetic fields up to 14 T. Temperature sweeps in static field and isothermal field sweeps as measured in DAC 2 are shown in Figure 3(a) and (b) respectively. The transition temperature was suppressed by about 15 K at 14 T as shown in Figure 3(a). We observe a broad step at around 240 K in the zero-field measurement in Figure 3(a), which disappears above 5 T. Very similar behaviour was observed in a bulk sample by Drozdov et al., [?], and we attribute this to inhomogeneity of the superconducting phase at this temperature. Measurements performed in magnetic field for DAC 1 are shown in Figure S10. The magnetic phase diagram is shown in Figure 4, constructed by extracting and from the intersection of linear fits to the normal state and superconducting transitions for both samples as shown in Figure 3. Linear behaviour of is observed close to , with the slope equal to and in DACs 1 and 2 respectively, close to values reported in Refs. [?, 9]. Fits to the Ginzburg–Landau (GL) and simplified Werthamer–Helfand–Hohenberg (WHH) models [29] yield estimates of the zero-temperature limit of the upper critical field in the range 120–167 T for DAC 1 and 116–159 T for DAC 2 respectively, in good agreement with previous studies of bulk LaH samples [?, 9]. The best estimate of for -LaH is around 144 T as determined by Sun et al., from WHH fits to pulsed field measurements up to 60 T [9]. We note that extrapolations of GL and WHH models may not capture any strong coupling or multiband effects that could become prominent towards lower temperatures in , and that direct measurement of requiring magnetic fields in excess of 100 T is beyond current experimental capabilities for samples within a DAC.

Figure 3. Suppression of the superconducting transition of the LaH film in DAC 2 at 176 GPa. (a) Temperature sweeps in static magnetic fields up to 14 T, demonstrating the downward shift in critical temperature . The value of was extracted as the intersection between linear fits to the normal state and superconducting transition as indicated for the 14 T curve. The XRD mapping of the phase is shown in the inset with the electrode configuration for the measurements. (b) Isothermal field sweeps for the sample in DAC 2. The value of was extracted as indicated for the curve measured at K. Measurements of DAC 1 in magnetic field are shown in Figure S10.

Figure 4. The magnetic phase diagram near for the LaH films constructed from measurements of and in Figure 3 and Figure S10, demonstrating linear behaviour of close to . The slope near , , is indicated for each DAC. Inset shows extrapolated fits to the Ginzburg-Landau (GL) and simplified Werthamer–Helfand–Hohenberg (WHH) models to estimate , found to be in the range 120-167 T for DAC 1 and 116-159 T for DAC 2.

Figure S10. Suppression of the superconducting transition of the LaH film in DAC 1 at 168 GPa, measured using probe 2 (see Figure S7 caption). (a) Temperature sweeps in static magnetic fields up to 14 T, demonstrating the downward shift in critical temperature . The value of was extracted as the intersection between linear fits to the normal state and superconducting transition as indicated for the 14 T curve. The XRD mapping of the sample is shown in the inset with the electrode configuration used for the measurements. Top inset plot shows the low temperature regime. Saturation of the resistance signal and variations in the noise level were caused by a sensitivity auto-ranging issue in the data acquisition program. At temperatures below 60 K in zero field, we observe a positive offset of approximately 125 µΩ in the resistance noise floor. We find that this offset is consistent with common-mode voltage arising from the finite resistance of the current leads, which is approximately 20 Ω per lead. At the applied excitation current of 100 µA, this offset corresponds to a common-mode voltage of 12.5 nV. The Stanford Research Systems Low-Noise Voltage Preamplifier SR560 has a common-mode rejection ratio (CMRR) of 100 dB up to 1 kHz. Using this specification, we estimate that a common-mode voltage of 12.5 nV would arise from a common-mode source corresponding to a resistance of approximately 12.5 Ω, which is of the same order as the measured resistance of our current leads. We therefore conclude that the finite resistance offset does not originate from the sample, but from imperfect rejection of the common-mode voltage associated with the current leads. (b) Isothermal field sweeps for the sample in DAC 1. The value of was extracted as indicated for the curve measured at K.
Estimates of from the WHH model are about a factor of 3 smaller than that expected from the weak-coupling Pauli limiting field ( T [30]) suggesting orbital pair-breaking effects are dominant in our LaH films, consistent with the findings for bulk LaH [?, 9] as well as other high- hydrides measured to date [4, 5, 6, 7, 31].
In line with previous measurements on bulk samples, we estimate the superconducting coherence length according to the Ginzburg-Landau equation for an orbital-limited type-II superconductor, where is the magnetic flux quantum. Substituting as estimated from GL and WHH models, we find nm for DAC 1, and nm for DAC 2, in line with values reported in previous studies [?, 9]. Differences observed in the coherence length and the slope near likely arise due to differences in sample purity and mean free path between the different studies. A rigorous classification of LaH as a type-II superconductor would require knowledge of both the superconducting coherence length, , and the magnetic penetration depth, , through the Ginzburg-Landau parameter . Since is not measured in the present work, the coherence length alone is insufficient to establish , which defines a type-II superconductor. Nevertheless, the short coherence lengths obtained here are commonly associated with strongly type-II superconductivity, and indeed they are comparable to those of type-II unconventional superconductors such as Bi-2223 ( nm) and smaller than those of type-II conventional superconductors including MgB ( nm) and NbSn ( nm) [32]. From the inferred values of , we estimate the weak-coupling BCS Fermi velocity using . Using the upper and lower values of , is found to be in the range m s and m s in DACs 1 and 2 respectively, within the range of universal Fermi velocities reported in Refs. [9, 33] and in good agreement with previous reports. A summary of the superconducting properties of LaH found in this work and in prior studies is given in Table 1.
| Study | BCS | |||||
| (GPa) | (K) | (T K) | (T) | (nm) | ( m s) | |
| DAC 1 (this work) | 168 5 | 246.3 | 120–167 | 1.40–1.66 | 2.50–2.96 | |
| DAC 2 (this work) | 176 5 | 246.6 | 116–159 | 1.44–1.68 | 2.57–3.00 | |
| Drozdov et al., [?] | 150 | 249 | 95–136 | 1.56–1.86 | 2.81–3.35 | |
| Sun et al., [9] | 136 | 246 | 143.5 | 1.51 | 2.77 |
Table 1. Superconducting parameters of -LaH from different high-pressure studies. Samples were synthesised from La films + AB in this work, while works of Drozdov et al., and Sun et al., used bulk La and pure H [?, 9]. The values of displayed in the table for this work are those extracted from warming curves at 0.7 K min. The lower and upper values given for in this work are those extracted from GL and WHH extrapolations respectively, while those from Drozdov et al., were obtained from GL fits to constructed using values defined by the 90% and 50% normal-state resistance criteria reported in Ref. [?]. The range of values for and BCS in each study are calculated from the corresponding range of values according to the equations given in the main text.
For type-II superconductors, a broadening of the resistive transition is expected in magnetic field due to dissipation associated with the motion of vortices in the mixed state [34, 35]. The apparent field dependence of the transition width can, however, depend sensitively on the criterion used to define the superconducting transition, particularly in the presence of additional resistive features. Using the conventional 90% and 10% resistance criteria [34], the transition widths of our LaH films initially decrease with magnetic field before gradually increasing above 14 T (see Fig. fig:S11a). Qualitatively similar behaviour displaying an initial narrowing of the resistive transition width followed by broadening toward higher fields was shown across various classes of both conventional and unconventional type-II superconductors in Ref. [34], including in HS, MgB thin films, pnictides and cuprates. In this work, the initial narrowing arises primarily from the field suppression of the broad resistive step near 240 K, which we attribute to inhomogeneity of the superconducting phase (see Fig. fig:S11b). When the transition width is instead defined using the 90% and 50% resistance criteria, similar to that employed in Ref. [4], thereby excluding this feature, it increases monotonically with magnetic field (see Fig. fig:S11a). Since the maximum applied field of 14 T corresponds to less than 10% of the extrapolated , more pronounced broadening would be expected at higher magnetic fields.
Temporal stability of the crystal structure and superconductivity of LaH10±δ films
One important question to address is the long-term stability of lanthanum hydrides over time, and whether hydrogen diffusion drives any temporal electronic or structural changes. Such phenomena have been observed previously in hydride compounds, for example in the high- hydride -LaH whereby it was proposed that hydrogen diffusion could be driving observed increases in the normal state resistance at temperatures above 250 K [5]. Further insight into the effects of hydrogen vacancies in LaH has recently been presented in Ref. [12]. There are also open questions about the long-term stability of -LaH. Recent NMR measurements have shown evidence of a continuous dehydrogenation of LaH () to a stoichiometry close to that of the precursor trihydride LaH over the course of about 70 days after laser heating and at constant pressure, with a consistent dehydrogenation timescale measured across four DACs [15]. The observed dehydrogenation was claimed to be further supported by electrical measurements that showed a tendency for to decrease over time in four DACs containing lanthanum hydride compounds.
In contrast to the NMR study, Minkov et al., demonstrated the long-term stability of both the crystal structure and superconductivity of -LaH using combined XRD and transport measurements [16]. The unit-cell volume was shown to remain stable for approximately 4.5 years, with a small decrease in volume of the order 0.3 Å per La atom attributed to a small pressure increase by about 5 GPa. Superconductivity with K was shown to be stable over timescales exceeding 5 years. In order to further address the questions surrounding the stability of -LaH, we performed a combined high-resolution synchrotron XRD and electrical transport study of our high- films over a period of about 300 days post laser-heating synthesis, beyond the timescale over which LaH was observed to decompose in NMR measurements [15]. A summary of our stability study is shown in Figure 5.

Figure 5. Stability of LaH films over time. (a) XRD data and structural refinements to the phase for DAC 1 at 168 GPa. Data are shown as open circles, with Pawley refinements shown as the red solid curves. All data were normalised to the intensity of the (111) reflection for comparison. Measurements were taken on the I15 beamline at Diamond light source over a period from 84 to 243 days after laser heating. The volumes obtained from refinements are displayed alongside their corresponding curves. Where present, peaks due to elemental Au from the electrodes and -BN from AB decomposition are indicated, and unassigned peaks are denoted with an asterisk (*). (b) Unit cell volume as a function of time after laser heating for both DACs, including data from beamlines I15 (see Figure S12 for DAC 2 refinement), ID11 (Figure S13) and ID27 (fig:1a). Shaded bands indicate the change in volume expected from the DFT equation of state in [17], given a 5 GPa pressure decrease in both DACs over the measurement period. (c) Resistance versus temperature curves for DAC 1 measured up to 301 days after laser-heating. (d) The stability of the superconducting transition temperature over time, extracted from cooling cycles. Data points correspond to extracted at 50% () of the resistance drop. Asymmetric error bars reflect the difference between and (upper error bar), and (lower error bar), as defined in (c).
Figure 5(a) shows XRD measurements performed on beamline I15 at Diamond Light Source for DAC 1 between 84 and 243 days post laser heating synthesis. Integrated patterns show clearly the presence of the phase up to 243 days after laser heating. The unit cell volumes extracted from structural refinements are displayed alongside each integrated pattern in (a). Unit cell volume is shown as a function of time in Figure 5(b) demonstrating stability of the crystal structure. Measurements presented in fig:5b were conducted on three different beamlines: I15 (circles) (fig:5a and Figure S12), as well as on ESRF beamlines ID11 (squares) (see Figure S13) and ID27 (triangles) (see fig:1a). Further, we observe no significant changes in the spatial extents of the hydride phases in either DAC, evidenced by comparing the high-resolution XRD spatial maps performed on beamlines ID11 and ID27, acquired with a 60 day interval between the measurements (see Figure S14). We emphasise that such a qualitative comparison does not rule out the possibility of the formation of an amorphous hydride phase over time, whether via reaction with surrounding free hydrogen or by decomposition of any of the identified hydride phases. Nevertheless, the retention of largely the same extent of -LaH and the absence of signatures of -LaH in any of our XRD measurements strongly imply that our high- films do not follow the decomposition pathway observed in the NMR study [15].

Figure S12. Measurements of DAC 2 on beamline I15 at Diamond Light Source 32 days after the second laser heating. The measured data are an average of five frames over the sample region to improve statistics due to poor resolution of the phase in the measurements and the reduced signal to noise from the I15 measurements compared to the ESRF measurements.

Figure S13. X-ray diffraction measurements performed on beamline ID11 at the ESRF. Pawley refinements of integrated patterns for (a) DAC 1 134 days after laser heating and (b) DAC 2 43 days after laser heating, showing reflections corresponding to -LaH. The positions on the sample at which the integrated frames were acquired are shown on the XRD mappings in fig:S14a and (c).

Figure S14. Comparison between high-resolution XRD spatial mappings from ESRF beamlines ID11 and ID27 for DACs 1 (a)-(b), and DAC 2 (c)-(d). Measurements were acquired about 60 days apart. Blue regions correspond to the (311) reflection of –LaH, localised black regions correspond to the (101) reflection of –LaH, and the red regions for DAC 1 correspond to the (101) reflection of –LaH. No traces of the phase were identified in DAC 2. The spatial map of the electrodes corresponds to the (111) reflection of Au. For each phase, the (hkl) were chosen to ensure no overlap between the identified phases. Frame numbers shown on the spatial maps indicate positions where integrated patterns were measured to refine the unit cell parameters of the phase. Visualisation of the spatial maps were performed using XDI [41].
We observe small changes in the unit cell volume of the phase over the course of the XRD measurements, with the largest variation of 1.1 Å in DAC 1 and 0.5 Å in DAC 2. This can very likely be attributed to small pressure decreases ( GPa) in both DACs over the measurement period (Figure S2 and Figure S3). Using the DFT equation of state from Ref. [17], we estimate that the observed increase in unit cell volume can likely be accounted for with a pressure decrease of 5 GPa in each DAC, illustrated by the shaded bands in fig:5b. We cannot rule out a further contribution to the volume increase due to hydrogen diffusion into the sample. We also note that integrated diffraction patterns acquired at the different beamtimes may not correspond exactly to the same position on the sample, and hence volumes may be slightly different due to pressure gradients in the DAC (see Figure S1 and Figure S15), or variations in hydrogen content across the sample.

Figure S15. Measurement of pressure at different positions over the sample in DAC 2 using the position of the diamond edge in Raman spectroscopy. Measurements shown are after the second laser heating of DAC 2. Raman laser spot ( nm) is visible at the sample centre. Dotted circle indicates the approximate culet position.
fig:5c shows four-point resistance measurements of for DAC 1 from immediately after laser heating up to 301 days after synthesis, demonstrating stability of the superconducting transition. Measurements were taken during cooling at a rate of K min in the contact configuration as shown on the inset XRD spatial map. An overall decrease in the normal state resistance over time was observed for both DACs (see Figure S16 for DAC 2 curves) while remained stable. Similar changes in the normal state were observed by Minkov et al., for bulk samples [16], and we attribute this behaviour to annealing effects at room temperature. We believe this further rules out the decomposition to LaH, since this would be expected to drive an increase in the normal state resistance over time due to its semi-metallic nature at similar pressures [?]. Figure 5 shows the measured values as a function of time for both DACs, extracted from cooling cycles. The value of in DAC 1 appears to show a slight increase in the first few days after laser-heating, but of both DACs is then stable over the course of about 300 days thereafter.

Figure S16. Resistance versus temperature curves for the sample in DAC 2, performed over a period up to 385 days after the second laser heating. remains very stable, while the magnitude of the normal state resistance decreases over time. This could be due to annealing at room temperature, or related to the 5 GPa pressure decrease in the DAC over time.
Our results reinforce the claims for the thermodynamic stability of . This is indeed consistent with the findings in Refs. [?, 36] whereby fcc formed spontaneously without laser heating when was pressurised to 140 GPa in a molecular environment. This demonstrates that is more stable than at high pressures in excess . One additional point to note is that in both this study and that of Minkov et al., the long-term retention of excess molecular hydrogen in the DACs after laser heating was reported (see Figure S2(c) and Figure S3(c) for this work). In the NMR study, no vibron was observed after laser heating [15], and it was noted by the authors that the observed dehydrogenation of the hydride sample could be related to the absence of chemical equilibrium between the hydride and the surrounding hydrogen reservoir. This remains an important open question that warrants further investigation. To further understand the decomposition observed in NMR measurements it would be desirable to perform combined XRD, Raman, and NMR measurements on samples synthesised via the same methods as in the NMR study, namely a laser-heated mixture of powder and ammonia borane. This would provide important insight into the synthesised crystal structures, reveal any sample heterogeneities and possible novel metastable phases, explore decomposition pathways and establish whether stabilisation requires a chemical equilibrium with excess molecular hydrogen in the sample chamber post laser heating.
Conclusion
We demonstrate the first successful formation of Fm={3}m- films in two diamond anvil cells (DACs) at pressures of 168 GPa and 176 GPa. The films were synthesised from an elemental lanthanum film deposited directly onto the anvils, and laser-heated at high pressures with purified ammonia borane as the hydrogen donor compound. High-resolution X-ray diffraction measurements confirmed the fcc lanthanum sublattice of , with unit cell parameters in excellent agreement with those reported for bulk samples in prior studies. Electrical transport measurements confirmed superconductivity with K, with the characteristic suppression observed in magnetic fields up to 14 T. Combined X-ray diffraction and electrical measurements reveal remarkable temporal stability of our films. We find no evidence of dehydrogenation during the 300-day period over which they were monitored following laser-heating, reinforcing the claims for the thermodynamic stability of - in an excess molecular environment. We further emphasise the importance of careful spatial diffraction mapping for reliable assignment of superconducting properties to the correct hydride phase, highly relevant for current ternary syntheses. Our work establishes thin-film methods as a promising platform for the future synthesis of ternary hydrides where accurate control of precursor stoichiometries will be critical. Thin-film techniques also provide a route toward the integration of micro-fabricated device geometries in diamond anvil cells for future detailed characterisation of the superconducting properties of hydrides.
Experimental details
The high-pressure synthesis of two films was carried out using custom DACs constructed from MP35N alloy, equipped with Type Ia Boehler and Type IIas Boehler (ultra-low fluorescence) anvils with culet diameters of , bevelled at to a diameter of . Gaskets were prepared by pre-indentation of rhenium to a thickness of followed by compression of an insulating boron nitride and epoxy mixture to a thickness of . A hole of diameter was laser-drilled in the insulation forming the sample chamber. For electrical measurements, five tungsten-gold bi-layer electrodes were deposited through a shadow mask directly onto one of the diamond anvils, with the smallest separation between the tips of the electrodes . Lanthanum films of thickness and were evaporated onto the electrodes in DACs 1 and 2 respectively at a rate of , and in a vacuum of following multiple pre-melting cycles to remove oxide impurities from the source material. The DACs were loaded with ammonia borane (AB, ) as the hydrogen donor and pressure transmitting medium (PTM). The AB was first purified by sublimation, and the DACs closed in an argon glovebox with residual and . DACs 1 and 2 were pressurised to and respectively, with the pressure determined from the Raman shift of the high-frequency diamond edge using the Akahama calibration [37] (see Figure S3 and Figure S17). Pressures determined from the position of the diamond edge typically yielded values 5–15 GPa higher than those inferred from the hydrogen vibron. Prior to laser heating, the metallic character of the elemental lanthanum films was verified in resistance versus temperature measurements (Figure S6), with residual resistance ratios (RRR) of 1.8 and 1.5 in DACs 1 and 2 respectively. The DACs were heated using a 1070 nm Yb-fiber laser in a one-sided geometry [38] with 0.2–0.3 s pulses. The four-point resistance of each sample was monitored during laser heating. Laser power was incremented after each pulse until an irreversible increase in the room temperature resistance was observed, after which laser heating was halted. This resistance increase indicated a chemical reaction between the lanthanum films and the hydrogen released from the ammonia borane, and occurred at a temperature below that necessary to observe any glowing. Hence, we could not perform a reliable temperature measurement using black-body radiation during laser heating. Resistance versus temperature measurements were performed using a low-frequency AC technique in a four-point configuration with excitation currents in the range –. Zero-field measurements were done using an SRS SIM921 AC resistance bridge (), and magnetic field measurements were performed in a Cryogenic Ltd 14 T magnet using a Keithley 6221 current source, a model SR560 preamplifier and SR830 lock-in amplifier (–) with an excitation current of . Resistances are presented as calculated from the in-phase component of the measured signal, with both the in-phase and quadrature components for each transport measurement provided in the data repository. X-ray diffraction measurements were performed on three beamlines: I15 beamline at Diamond Light Source, and ESRF beamlines ID11 and ID27. Details such as X-ray wavelength, beam diameter, detector model and sample-to-detector distance for measurements performed on each beamline are provided in the Supplementary Information Table S1. Integrated diffraction patterns and azimuthal projections of the detector images were produced using DIOPTAS [39] after masking of spurious reflections, and those from the diamond anvils. Background subtraction of the integrated patterns and subsequent structural refinements were performed using GSAS-II [40]. Spatial mapping of unique reflections corresponding to hydride phases within each DAC was performed using XDI [41].
| Beamline | Beam diameter at FWHM | Sample-to-detector distance | Detector | Calibrant | |
| I15 (DLS) | 0.30996 | 16 | 446 | Pilatus3 X CdTe 2M | |
| ID11 (ESRF) | 0.28436 | 1 | 148 | Eiger2 X CdTe 4M | |
| ID27 (ESRF) | 0.3738 | 0.7 | 200 | Eiger2 X CdTe 9M |
Table S1. Experimental details for XRD measurements performed on beamlines I15 at Diamond Light Source (DLS), and ID11 and ID27 at the European Synchrotron Radiation Facility (ESRF).

Figure S17. (a) Raman spectrum of DAC 1 before and immediately after laser heating. The edge of the first order diamond mode used to determine the pressure is labelled. After laser heating, a high-frequency vibron is observed, associated with free molecular released by thermal decomposition of the surrounding ammonia borane. (b) The derivative of the spectra in (a) in the wavenumber range of the diamond, where the pressure was determined from the wavenumber corresponding to the derivative minima according to the Akahama calibration [37]. (c) The position of the vibron observed in DAC 1 immediately after laser heating. The pressure inferred from the vibron according to Ref. [42] is indicated.
Acknowledgements
This work was supported by the EPSRC Open Fellowship of J.B. under Grant No. EP/Z533555/1. S.F. acknowledges support from the ERC Horizon 2020 programme under grant 715262-HPSuper and the EPSRC under grant EP/V048759/1. D.L. thanks the UKRI Future Leaders Fellowship (MR/V025724/1) for financial support. O.T.L. would like to acknowledge support from the Royal Society in the form of a University Research Fellowship (UF150057). We acknowledge Diamond Light Source for time on beamline I15 (proposals CY40128-2, CY40128-3, CY40128-4, CY40128-6, and CY40128-7). We acknowledge the European Synchrotron Radiation Facility (ESRF) for provision of synchrotron radiation facilities on beamlines ID11 and ID27 under proposals MA-6519 and HC-6190, respectively. We thank Jonathan Wright for assistance and support during the beamtime on ID11. We are grateful to Antony Carrington for useful discussions.
Data availability
Data will be made available at the University of Bristol data repository. X-ray diffraction data from ESRF beamlines ID11 (proposal: MA-6519) and ID27 (proposal: HC-6190) are available at DOI: 10.15151/ESRF-ES-2156316192 and DOI: 10.15151/ESRF-ES-2125107364, respectively.
References
Supplementary Information
Synthesis and stability of high– films at high pressures

Figure S11. Broadening of the superconducting transition under external magnetic fields. We calculate as defined in Ref. [34] using a 90% and 10% criteria: , and also using a 90% and 50% criteria for comparison: . Using the 90% and 10% criteria, we find that the transition widths for both DAC 1 and 2 slightly decrease in fields up to 10 T and begin to broaden at 14 T. This behaviour is similar to behaviour reported in Ref. [34]. However, using the 90% and 50% criteria, the transition widths increase monotonically in magnetic field. The maximum applied magnetic field here is 14T which is less than 10% of for DAC 1 and 2. (b) Resistance curve for DAC 1 in zero field, illustrating how the different values of were determined.
References
- [4]I. Osmond, O. Moulding, S. Cross, T. Muramatsu, A. Brooks, O. Lord, T. Fedotenko, J. Buhot, and S. Friedemann, Clean-limit superconductivity in Im3̄m H₃S synthesized from sulfur and hydrogen donor ammonia borane, Physical Review B 105, L220502 (2022), publisher: APS.DOI
- [5]S. Cross, J. Buhot, A. Brooks, W. Thomas, A. Kleppe, O. Lord, and S. Friedemann, High-temperature superconductivity in La₄H₂₃ below 100 GPa, Physical Review B 109, L020503 (2024), publisher: APS.
- [6]P. Kong, V. S. Minkov, M. A. Kuzovnikov, A. P. Drozdov, S. P. Besedin, S. Mozaffari, L. Balicas, F. F. Balakirev, V. B. Prakapenka, S. Chariton, and others, Superconductivity up to 243 K in the yttrium-hydrogen system under high pressure, Nature communications 12, 5075 (2021), publisher: Nature Publishing Group UK London.DOI
- [7]W. Chen, D. V. Semenok, X. Huang, H. Shu, X. Li, D. Duan, T. Cui, and A. R. Oganov, High-temperature superconducting phases in cerium superhydride with a Tc up to 115 K below a pressure of 1 Megabar, Physical Review Letters 127, 117001 (2021), publisher: APS.DOI
- [8]J. Guo, D. Semenok, G. Shutov, D. Zhou, S. Chen, Y. Wang, K. Zhang, X. Wu, S. Luther, T. Helm, and others, Unusual metallic state in superconducting A15-type La₄H₂₃, National Science Review, nwae149 (2024), publisher: Oxford University Press.DOI
- [9]D. Sun, V. S. Minkov, S. Mozaffari, Y. Sun, Y. Ma, S. Chariton, V. B. Prakapenka, M. I. Eremets, L. Balicas, and F. F. Balakirev, High-temperature superconductivity on the verge of a structural instability in lanthanum superhydride, Nature Communications 12, 6863 (2021), publisher: Nature Publishing Group UK London.DOI
- [10]J. A. Flores-Livas, L. Boeri, A. Sanna, G. Profeta, R. Arita, and M. Eremets, A perspective on conventional high-temperature superconductors at high pressure: Methods and materials, Physics Reports 856, 1 (2020), publisher: Elsevier.
- [11]F. Peng, Y. Sun, C. J. Pickard, R. J. Needs, Q. Wu, and Y. Ma, Hydrogen clathrate structures in rare earth hydrides at high pressures: possible route to room-temperature superconductivity, Physical review letters 119, 107001 (2017), publisher: APS.DOI
- [12]I. Osmond, L. J. Conway, M. A. Kuzovnikov, C. Stevens, T. Marqueño, H. A. Shuttleworth, A. Huxley, C. J. Pickard, G. J. Ackland, R. T. Howie, and M. Peña-Alvarez, Hydrogen Vacancy Induced Superconductivity Collapse in A15 Lanthanum Hydride, Physical Review Letters 136, 086102 (2026).
- [13]H. Liu, I. I. Naumov, R. Hoffmann, N. W. Ashcroft, and R. J. Hemley, Potential high-Tc superconducting lanthanum and yttrium hydrides at high pressure, Proceedings of the National Academy of Sciences 114, 6990 (2017).DOI
- [14]I. Errea, F. Belli, L. Monacelli, A. Sanna, T. Koretsune, T. Tadano, R. Bianco, M. Calandra, R. Arita, F. Mauri, and others, Quantum crystal structure in the 250-kelvin superconducting lanthanum hydride, Nature 578, 66 (2020), publisher: Nature Publishing Group UK London.DOI
- [15]Y. Zhou, Y. Fu, M. Yang, I. Osmond, R. Jana, T. Nakagawa, O. Moulding, J. Buhot, S. Friedemann, D. Laniel, and T. Meier, Diffusion-driven transient hydrogenation in metal superhydrides at extreme conditions, Nature Communications 16, 1135 (2025).
- [16]V. S. Minkov, M. A. Kuzovnikov, P. Kong, A. P. Drozdov, F. Du, J. Yan, J. Kim, S. Chariton, V. B. Prakapenka, M. Mezouar, B. Wehinger, G. A. Smith, F. F. Balakirev, and E. F. Talantsev, Long-term stability of superconducting metal superhydrides, Nature Communications 17, 8171 (2026).
- [17]D. Laniel, F. Trybel, B. Winkler, F. Knoop, T. Fedotenko, S. Khandarkhaeva, A. Aslandukova, T. Meier, S. Chariton, K. Glazyrin, and others, High-pressure synthesis of seven lanthanum hydrides with a significant variability of hydrogen content, Nature Communications 13, 6987 (2022), publisher: Nature Publishing Group UK London.
- [18]J. Buhot, O. Moulding, T. Muramatsu, I. Osmond, and S. Friedemann, Experimental evidence for orthorhombic Fddd crystal structure in elemental yttrium above 100 GPa, Physical Review B 102, 104508 (2020), publisher: American Physical Society.DOI
- [19]F. F. Balakirev, V. S. Minkov, A. P. Drozdov, and M. I. Eremets, Evidence of superconductivity in electrical resistance measurements of hydrides under high pressure, Journal of Superconductivity and Novel Magnetism 37, 1779 (2024).DOI
- [20]G. S. Pawley, Unit-cell refinement from powder diffraction scans, Journal of Applied Crystallography 14, 357 (1981).DOI
- [21]Z. M. Geballe, H. Liu, A. K. Mishra, M. Ahart, M. Somayazulu, Y. Meng, M. Baldini, and R. J. Hemley, Synthesis and stability of lanthanum superhydrides, Angewandte Chemie 130, 696 (2018), publisher: Wiley Online Library.
- [22]A. M. Shipley, M. J. Hutcheon, M. S. Johnson, R. J. Needs, and C. J. Pickard, Stability and superconductivity of lanthanum and yttrium decahydrides, Physical Review B 101, 224511 (2020).DOI
- [23]I. A. Kruglov, D. V. Semenok, H. Song, R. Szcześniak, I. A. Wrona, R. Akashi, M. M. D. Esfahani, D. Duan, T. Cui, A. G. Kvashnin, and others, Superconductivity of LaH₁₀ and LaH₁₆ polyhydrides, Physical Review B 101, 024508 (2020).
- [24]F. Datchi, A. Dewaele, Y. Le Godec, and P. Loubeyre, Equation of state of cubic boron nitride at high pressures and temperatures, Physical Review B 75, 214104 (2007).DOI
- [25]U. B. Demirci, Ammonia borane, a material with exceptional properties for chemical hydrogen storage, International Journal of Hydrogen Energy 42, 9978 (2017).DOI
- [26]C. V. Storm, C. R. Roy, K. A. Munro, and M. I. McMahon, Crystal structures of lanthanum to 230 GPa, Physical Review B 110, 024107 (2024).DOI
- [27]F. A. Gorelli, P. Dalladay-Simpson, G. Marchese, Z.-y. Cao, P. Barone, L. Benfatto, G. Garbarino, and F. Mauri, Raman fingerprint of high-temperature superconductivity in compressed hydrides, arXiv:2511.10784 10.21203/rs.3.rs-8105210/v1 (2025).arxiv.org/abs/2511.10784
- [28]R. Li, T. Hashimoto, K. Ohta, and H. Okamoto, Thermal conductivity and electrical resistivity of very high-purity copper at low temperatures, in Proc. 12th Int. Cryogenic Eng. Conf. (Elsevier, 1988) pp. 414–417.DOI
- [29]T. Baumgartner, M. Eisterer, H. W. Weber, R. Flükiger, C. Scheuerlein, and L. Bottura, Effects of neutron irradiation on pinning force scaling in state-of-the-art Nb₃Sn wires, Superconductor Science and Technology 27, 015005 (2013), publisher: IOP Publishing.DOI
- [30]A. M. Clogston, Upper Limit for the Critical Field in Hard Superconductors, Physical Review Letters 9, 266 (1962).DOI
- [31]S. Mozaffari, D. Sun, V. S. Minkov, A. P. Drozdov, D. Knyazev, J. B. Betts, M. Einaga, K. Shimizu, M. I. Eremets, L. Balicas, and F. F. Balakirev, Superconducting phase diagram of H₃S under high magnetic fields, Nature Communications 10, 2522 (2019).
- [32]C. Yao and Y. Ma, Superconducting materials: Challenges and opportunities for large-scale applications, iScience 24, 102541 (2021).DOI
- [33]E. F. Talantsev, Universal Fermi velocity in highly compressed hydride superconductors, Matter and Radiation at Extremes 7, 058403 (2022).
- [34]E. Talantsev, V. Minkov, F. Balakirev, and M. Eremets, Comment on “Nonstandard superconductivity or no superconductivity in hydrides under high pressure”, Physical Review B 110, 186501 (2024).DOI
- [35]J. E. Hirsch and F. Marsiglio, Nonstandard superconductivity or no superconductivity in hydrides under high pressure, Physical Review B 103, 134505 (2021).DOI
- [36]V. S. Minkov, M. A. Kuzovnikov, P. Kong, A. P. Drozdov, F. Du, J. Yan, J. Kim, S. Chariton, V. B. Prakapenka, M. Me zouar, B. Wehinger, G. A. Smith, F. F. Balakirev, and E. F. Talantsev, Long-Term Stability of Superconducting Metal Superhydrides (2025), arXiv:2507.08009 [cond-mat].arxiv.org/abs/2507.08009
- [37]Y. Akahama and H. Kawamura, Pressure calibration of diamond anvil Raman gauge to 310 GPa, Journal of Applied Physics 100 (2006), publisher: AIP Publishing.DOI
- [38]O. T. Lord, E. T. Wann, S. A. Hunt, A. M. Walker, J. Santangeli, M. J. Walter, D. P. Dobson, I. G. Wood, L. Vočadlo, G. Morard, and M. Mezouar, The NiSi melting curve to 70 GPa, Physics of the Earth and Planetary Interiors 233, 13 (2014), publisher: Elsevier.
- [39]C. Prescher and V. B. Prakapenka, DIOPTAS: a program for reduction of two-dimensional X-ray diffraction data and data exploration, High Pressure Research 35, 223 (2015), publisher: Taylor & Francis.DOI
- [40]B. H. Toby and R. B. Von Dreele, GSAS-II: the genesis of a modern open-source all purpose crystallography software package, Journal of Applied Crystallography 46, 544 (2013), publisher: International Union of Crystallography.DOI
- [41]R. Hrubiak, J. S. Smith, and G. Shen, Multimode scanning X-ray diffraction microscopy for diamond anvil cell experiments, Review of Scientific Instruments 90, 025109 (2019).DOI
- [42]M. I. Eremets, V. S. Minkov, P. P. Kong, A. P. Drozdov, S. Chariton, and V. B. Prakapenka, Universal diamond edge Raman scale to 0.5 terapascal and implications for the metallization of hydrogen, Nature Communications 14, 907 (2023).DOI