WorksPhysics and Astronomy
High-temperature superconductivity in NdSrNiO membranes under pressure
Abstract
Lattice compression has emerged as a fundamental tuning parameter for nickelate superconductivity. Pressure acts as a trigger to induce superconductivity in bulk Ruddlesden-Popper nickelates. For infinite-layer nickelate thin films, compressive epitaxial strain and rare-earth ion chemical pressure have been used to substantially enhance the superconducting transition temperature (). Efforts to go further have been constrained by the limits of epitaxial stability or the challenges of measuring thin films in high-pressure environments. Here, we overcome this limitation by developing a technique to incorporate freestanding infinite-layer membranes into a diamond anvil cell. Using this platform, we observe a strong increase in up to our highest measurement pressure of 90 GPa, where a superconducting downturn can be observed near liquid nitrogen temperatures. Strikingly, we find a simple linear enhancement of at a rate of 0.65 K GPa, with no signs of saturation. This suggests that the pairing strength in infinite-layer nickelates can be raised to a surprisingly high scale, using an approach that can be broadly applied to many two-dimensional materials.
Introduction
Since the discovery of superconductivity in infinite-layer nickelates[3], considerable evidence has accumulated to indicate its unconventional nature, including the observation of a superconducting dome[4, 10], the presence of strong magnetic fluctuations[11, 12], and evidence for a nodal gap structure[13, 14, 15]. Already the initially reported superconducting transition temperature () was far above the scale expected from electron-phonon coupling[16]. Materials optimization in epitaxial thin films has progressively increased from the initial ~15 K (Ref. 3) to ~40 K (Ref. 6). Two developments have been central to this progress. First, moving to substrates better lattice-matched to the precursor perovskite phase improved crystallinity while imposing compressive strain on the infinite-layer phase[4, 5, 17]. Second, replacing the rare-earth ion with samarium better accommodated this smaller substrate lattice while introducing chemical pressure that shortened the -axis lattice parameter[6, 7]. Together, these advances establish a clear empirical trend – lattice contraction drives enhancement. This raises a fundamental question: what is the intrinsic upper bound of in infinite-layer nickelates, and does lattice-contraction-driven enhancement continue beyond the range currently accessible in thin films?
High pressure provides a natural route to address this question. The structural simplicity of the infinite-layer phase circumvents the complex structural responses associated with planar buckling in cuprates[18, 19] and octahedral tilts in bilayer nickelates[1]. In this sense, it offers a cleaner platform to investigate how the electronic structure and superconductivity evolve under compression. Progress in high-pressure studies has been constrained, however, by the absence of bulk superconducting infinite-layer nickelate samples[20]. While a seminal thin-film pressure study reported an increase in up to ~30 K, the stability of these substrate-bound films rapidly deteriorated well below ~12 GPa (ref. 8). This limitation motivates an alternative geometry: a freestanding membrane in which the superconducting infinite-layer nickelate film is fully released from its growth substrate[9, 21]. This sample geometry is similar to exfoliated flakes from a broad family of two-dimensional materials, for which high pressure studies are beginning to emerge[22]. Here we use such membranes in a diamond anvil cell (DAC) that enables us to access the extreme pressure regime.
Main
Optimally doped thin films (~6.7 nm) were synthesized by pulsed laser deposition followed by topotactic reduction. The films were encapsulated by protective capping layers and grown on a water-soluble sacrificial layer (synthesis and characterization described in Ref. 9). Dissolution of the sacrificial layer released a polymer-supported infinite-layer nickelate membrane, enabling direct integration of the superconducting membrane onto the diamond culet in a substrate-free geometry (Figure 1a (PDF p. 15),Figure 1b (PDF p. 15)). Electrical contacts were defined in a van der Pauw geometry using a patterned, suspended silicon nitride stencil mask. Sequential ion milling and electron-beam evaporation locally removed the insulating cap to form Ti/Au Ohmic contacts to the membrane (Figure 1c (PDF p. 15),d). The electrodes were extended using Pt foils, and the DAC was assembled with a metal gasket insulated by cubic boron nitride (cBN) powder (Fig. 1e) and silicone oil as the pressure-transmitting medium (PTM). The substrate-free membranes exhibit robust superconductivity, with rising continuously throughout stepwise pressurization. This pressure-driven enhancement culminates at ~91.5 GPa (our maximum achieved measurement pressure), where the temperature-dependent sheet resistance shows a superconducting transition with ~ 74.2 K (Fig. 1f; is taken as the onset temperature throughout this work including literature comparisons, see Extended Data Fig. 1).

Figure 1. High-pressure DAC platform for freestanding infinite-layer nickelate membranes. a, Schematic of the polymer-supported 10 u.c. / 20 u.c. / 10 u.c. / 10 u.c. heterostructure grown on the substrate, followed by water etching of the sacrificial layer. b, Transfer of the freestanding membrane onto a diamond culet. c, Argon ion milling and electron-beam evaporation of Ti (5 nm)/Au (125 nm) onto the transferred membrane using a silicon nitride stencil mask. d, Optical microscope image of the processed sample on the diamond culet. e, Schematic of the DAC assembly with a metal gasket coated with insulating cBN. Pt foil extends the evaporated metal electrodes. f, at GPa, showing K. Inset: an optical microscope image of the sample chamber within the DAC.
Figure 2a,Figure 2b displays the stepwise evolution of with pressure for two representative samples measured using 300 m and 200 m diamond culets, respectively; the 200 m culet was used to access higher pressures up to its 90 GPa mechanical limit. At ambient pressure, K falls in the established range reported for as-grown thin films and freestanding membranes transferred onto substrates[9]. With increasing pressure, rises monotonically, a trend observed across all six measured samples (Extended Data Fig. 2). The zero-resistance temperature (defined as the temperature at which the resistance reaches the measurement noise floor) lags the pressure evolution of , leading to a broadened superconducting transition. Such broadening is also observed in high-pressure transport measurements across cuprates[23] and Ruddlesden–Popper nickelates[1, 24]. A likely origin is the non-hydrostatic environment in the DAC,

Figure 2. Pressure evolution of superconductivity in infinite-layer nickelate membranes. a, b, as a function of pressure for Sample 1 on a 300 culet (a) and Sample 2 on a 200 culet (b). c, of Sample 3, measured at 71.8 GPa under out-of-plane magnetic fields from 0 to 13 T. d, Upper critical fields as a function of normalized temperature for a pristine thin film ( GPa, circles) and Sample 3 at GPa (squares), 65.5 GPa (upward triangles), and 71.8 GPa (downward triangles). values were determined using 50% of the extrapolated normal-state resistance criterion (). Solid lines represent linearized Ginzburg–Landau fits near . The inset shows the extracted in-plane coherence length plotted against zero-field (bottom axis) and the corresponding pressure (top axis).
arising from the solidification of the silicone oil PTM near its glass transition at a relatively low pressure (~3 GPa)[25]. The observed broadening cannot be explained by the small macroscopic pressure gradient alone (Extended Data Figure 3), but may reflect the nanometer-thick membrane geometry, which has limited ability to elastically average over localized load variations.

Extended Data Figure 3. Pressure determination and pressure gradients in high-pressure electrical transport measurements. Representative Raman spectra collected at ~56 GPa from different locations within the sample chamber are shown. A small pressure variation is observed across different locations. The reported pressure values represent the pressure in the immediate vicinity of the sample, estimated from Raman measurements taken at a nearby position (position B) to avoid direct laser exposure of the sample.
Despite this expected sensitivity to non-hydrostaticity, the superconducting transition remains robust up to the highest pressure reached. Notably, begins to increase more rapidly above ~60 GPa and narrows the gap to , leading to a stabilization of the transition width (Extended Data Fig. 4). Ultimately, the accessible pressure range is limited by diamond anvil fracture rather than an apparent saturation of the sample response. The concurrent accelerated rise in and reduction in normal-state resistance suggests that inhomogeneities becomes less limiting at elevated pressures.
Figure 2c shows a series of curves measured under increasing out-of-plane magnetic fields up to 13 T at GPa, where reaches ~64 K. Below , the magnetoresistive transition is progressively suppressed and broadened, characteristic of type-II superconductors. From these curves, we extract the upper critical field, , using the criterion of 50% of the extrapolated normal-state resistance (following the approach in Ref. 26), and plot the values as a function of normalized temperature in Figure 2d. From the -linear response in the critical regime near , we fit the data with the linearized Ginzburg-Landau form to obtain the in-plane coherence length, . Repeating these measurements at intermediate pressures, alongside a pristine thin film as a zero-pressure reference, reveals a clear change in the slope of the linear fits (Figure 2d). The extracted decreases from ~2.39 nm to ~1.54 nm at 71.8 GPa, exhibiting a monotonic reduction that inversely tracks the increase of under pressure, reflecting the strengthening of the superconducting state.
Discussion
Figure 3 compiles the pressure dependence for all six measured samples, which are remarkably consistent. The data are well described by a line with slope ~ 0.65 K GPa.

Figure 3. Distinct high-pressure trajectory relative to other superconducting families. Onset values of all six measured samples (closed symbols) are fitted together, yielding a combined linear increase of ~0.65 K GPa (dashed gray line). Open symbols provide comparison data for bilayer nickelate, various cuprates, and iron selenide: LaNiO (ref. 24), HgBaCaCuO (ref. 23), YBaCuO (ref. 27), YBaCuO (ref. 27), BiSrCaCuO (ref. 33), and -FeSe (ref. 28).
Surprisingly, exhibits no sign of saturation. This very simple linear increase motivates a comparison with the high-pressure response of other superconducting nickelates, cuprates, and iron selenides. Most high-pressure studies of high- superconductors report an eventual suppression of at elevated pressure. The bilayer nickelate LaNiO, for example, exhibits a sharp emergence of superconductivity near GPa followed by a gradual suppression beyond GPa (ref. 1,24). Cuprates such as (Ref. 23) and (Ref. 27) and - (Ref. 28) similarly exhibit an eventual saturation and downturn in (Figure 3).
In cuprates, the pressure evolution is widely understood through two competing effects: a non-doping structural contribution that enhances pairing, and a pressure-induced charge transfer that ultimately overdopes the system and suppresses superconductivity[29]. Depending on the balance of these effects, optimally doped cuprates may exhibit a transient initial rise[23, 30] or an almost immediate overdoping-driven suppression (e.g., (Ref. 27) in Figure 3). Theoretical models for infinite-layer nickelates propose a similar fate: a pressure-enhanced self-doping effect driven by -axis compression of rare-earth-derived three-dimensional electron pockets[31, 32] is predicted to cause in optimally doped to peak near 50 K at GPa before declining due to self-doping into the overdoped regime[31]. Against this context, not only does the over four-fold enhancement we observe for optimally doped far exceed the maximum increase seen in cuprates (), but the persistent, non-saturating trajectory has yet to exhibit any sign of an overdoping penalty that apparently limits cuprates. We note a rare exception to the usual dome-shaped is the resurgence in (ref. 33, Fig. 3), with similar trends found in monolayer and bilayer counterparts{citep}34`. Although the underlying mechanism remains debated, this suggests that the non-saturating response observed in is not an isolated anomaly, and it implies further opportunities to enhance .
Methods
Diamond Anvil Cell Preparation: High pressure was generated using a non-magnetic diamond anvil cell made of Be–Cu alloy, equipped with either or culet diamonds. Insulating gaskets were prepared from T301 stainless steel or Re ( thickness) and pre-indented to GPa. The full culet area was then laser-drilled, and a mixture of cBN powder and epoxy (1:10 by weight)[35] was packed into the indentation and compressed between the anvils to form a dense insulating layer. For culets, the insulating layer was subsequently compressed to a thickness of and a sample chamber of diameter was drilled using an in-house laser-drilling system. For culets, the insulating layer thickness was with the sample chamber diameter of . Following sample transfer and electrode fabrication (Figure 1b (PDF p. 15)-d), Pt foils and Cu wires were attached to extend the evaporated electrodes. The gasket was then realigned on the anvil, and the diamond anvil cell was assembled with an initial anvil separation of –, estimated from optical interference fringes. Silicone oil served as the pressure-transmitting medium to maintain quasi-hydrostatic conditions.
Pressure calibration: Pressure was calibrated at room temperature using ruby fluorescence (for 0.1–50 GPa) and the diamond Raman edge (for 50–100 GPa). The uncertainty is GPa for ruby fluorescence and, for the diamond Raman edge, increases with pressure but remains below 5% up to the highest pressure reached in this study. Pressures above 50 GPa were measured at positions adjacent to the sample rather than directly on the sample (see Extended Data Figure 3). Ruby fluorescence and diamond Raman spectra were collected using a Renishaw inVia Raman microscope with a spectral resolution of – and 514.5/633 nm excitation lasers at a laser power of mW to minimize possible damage to the sample.
Data availability
All source data supporting the findings of this study are provided.
Acknowledgements
Y.L., M.W., and X.W. contributed equally to this work. The authors acknowledge Feng Ke for his contribution to the initial stage of this work, and Bai Yang Wang, Kyuho Lee, Brian Moritz, Cheng Peng, Jiarui Li, and Florian Theuss for useful discussions. This work was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering under Contract No. DE-AC02-76SF00515. Development of device/electrical assembly was supported by the Kavli Foundation, Klaus Tschira Stiftung, and Kevin Wells.
Author contributions
Y.L. and X.W. synthesized freestanding infinite-layer nickelate membranes, transferred them onto diamond culets, and prepared electrical contacts. M.W. designed and assembled the diamond anvil cells for high-pressure resistance measurements. X.W. developed the dry-transfer method and stage for the precise integration of membranes onto diamond anvil cell. Y.L and X.W. carried out transport measurements with assistance from M.W. All authors contributed to the writing of the manuscript. Y.L., Y.Y., W.L.M., Y.Lin, and H.Y.H. supervised the project.
Conflict of Interest
The authors declare no conflict of interest.
Extended Data Figures
(a)

(b)

(c) Evaluation method for the onset critical temperature . a, The 99.5% criterion, defined as the temperature at which the resistance drops to 99.5% of the extrapolated normal-state linear fit. b, The intersection method, determined by the intersection of linear fits to the normal state and the superconducting transition region. c, The second-derivative method, defined as the zero-intercept of a linear fit to the curve near the onset region. Definition (a) is adopted throughout the manuscript.
Figure 1 (PDF p. 15).

Extended Data Figure 2. Full temperature range for all samples. a-f, sample 1–6, respectively.

Figure Extended Data Fig. 4. Enhancement in and corresponding saturation of the transition width. a, , defined as the temperature at which the resistance reaches the noise floor, as a function of pressure. Data are shown for Samples 2-4, where the zero-resistance state could be unambiguously resolved. b, Corresponding transition width, defined as .
References
- [1]Sun, H. et al. Signatures of superconductivity near 80 K in a nickelate under high pressure. Nature 621, 493–498 (2023).DOI
- [3]Li, D. et al. Superconductivity in an infinite-layer nickelate. Nature 572, 624–627 (2019).DOI
- [4]Lee, K. et al. Linear-in-temperature resistivity for optimally superconducting (Nd,Sr)NiO₂. Nature 619, 288–292 (2023).DOI
- [5]Ren, X. et al. Possible strain-induced enhancement of the superconducting onset transition temperature in infinite-layer nickelates. Commun. Phys. 6, 1–8 (2023).DOI
- [6]Chow, S. L. E., Luo, Z. & Ariando, A. Bulk superconductivity near 40 K in hole-doped SmNiO₂ at ambient pressure. Nature 642, 58–63 (2025).DOI
- [7]Yang, M. et al. Enhanced superconductivity and mixed-dimensional behaviour in infinite-layer samarium nickelate thin films. Nat. Commun. 17, 2761 (2026).DOI
- [9]Lee, Y. et al. Synthesis of superconducting freestanding infinite-layer nickelate heterostructures on the millimetre scale. Nat. Synth. 4, 573–581 (2025).DOI
- [10]Zeng, S. et al. Phase diagram and superconducting dome of infinite-layer Nd₁−xSrₓNiO₂ thin films. Phys. Rev. Lett. 125, 147003 (2020).
- [11]Lu, H. et al. Magnetic excitations in infinite-layer nickelates. Science 373, 213–216 (2021).DOI
- [12]Krieger, G. et al. Charge and spin order dichotomy in NdNiO₂ driven by the capping layer. Phys. Rev. Lett. 129, 027002 (2022).DOI
- [13]Gu, Q. et al. Single particle tunneling spectrum of superconducting Nd₁₋ₓSrₓNiO₂ thin films. Nat. Commun. 11, 6027 (2020).DOI
- [14]Cheng, B. et al. Evidence for d-wave superconductivity of infinite-layer nickelates from low-energy electrodynamics. Nat. Mater. 23, 775–781 (2024).DOI
- [15]Harvey, S. P. et al. Evidence for nodal superconductivity in infinite-layer nickelates. Proc. Natl Acad. Sci. USA 122, e2427243122 (2025).DOI
- [16]Nomura, Y. et al. Formation of a two-dimensional single-component correlated electron system and band engineering in the nickelate superconductor NdNiO₂. Phys. Rev. B 100, 205138 (2019).
- [17]Lee, K. et al. Effects of stoichiometry and epitaxial strain on the stabilization of infinite-layer nickelates. APL Mater. 13, 101105 (2025).
- [18]Jorgensen, J. D. et al. Structural features that optimize high temperature superconductivity. in Recent Developments in High Temperature Superconductivity (eds Klamut, J., Veal, B. W., Dabrowski, B. M. & Klamut, P. W.) 1–15 (Springer, Berlin, Heidelberg, 1996).
- [19]Yamada, N. & Ido, M. Pressure effects on superconductivity and structural phase transitions in La₂₋ₓMₓCuO₄ (M = Ba,Sr). Physica C 203, 240–246 (1992).DOI
- [20]Puphal, P. et al. Topotactic transformation of single crystals: from perovskite to infinite-layer nickelates. Sci. Adv. 7, eabl8091 (2021).DOI
- [21]Yan, S. et al. Superconductivity in freestanding infinite-layer nickelate membranes. Adv. Mater. 36, 2402916 (2024).DOI
- [22]Pimenta Martins, L. G. et al. High-pressure studies of atomically thin van der Waals materials. Appl. Phys. Rev. 10, 011313 (2023).
- [23]Gao, L. et al. Superconductivity up to 164 K in HgBa₂Caₘ₋₁CuₘO₂ₘ₊₂₊δ (m = 1, 2, and 3) under quasihydrostatic pressures. Phys. Rev. B 50, 4260–4263 (1994).pubmed.ncbi.nlm.nih.gov/9976724
- [24]Li, J. et al. Identification of superconductivity in bilayer nickelate La₃Ni₂O₇ under high pressure up to 100 GPa. Natl. Sci. Rev. 12, nwaf220 (2025).DOI
- [25]Klotz, S., Chervin, J.-C., Munsch, P. & Le Marchand, G. Hydrostatic limits of 11 pressure transmitting media. J. Phys. D: Appl. Phys. 42, 075413 (2009).DOI
- [29]Neumeier, J. J. & Zimmermann, H. A. Pressure dependence of the superconducting transition temperature of YBa₂Cu₃O₇ as a function of carrier concentration: A test for a simple charge-transfer model. Phys. Rev. B 47, 8385–8388 (1993).pubmed.ncbi.nlm.nih.gov/10004873
- [30]Chu, C. W. et al. Superconductivity above 150 K in HgBa₂Ca₂Cu₃O₈₊δ at high pressures. Nature 365, 323–325 (1993).DOI
- [31]Di Cataldo, S., Worm, P., Tomczak, J. M., Si, L. & Held, K. Unconventional superconductivity without doping in infinite-layer nickelates under pressure. Nat. Commun. 15, 3952 (2024).DOI
- [32]Sharma, S., Jung, M.-C., LaBollita, H. & Botana, A. S. Pressure effects on the electronic structure and magnetic properties of infinite-layer nickelates. Phys. Rev. B 110, 155156 (2024).DOI
- [35]Funamori, N. & Sato, T. A cubic boron nitride gasket for diamond-anvil experiments. Rev. Sci. Instrum. 79, 053903 (2008).DOI
