roomtsc

Reference › Verifying a superconductor

Verifying a superconductor

41 sources, listed at the end · last revised 7 October 2026

For the stories of individual claims, see Claims of room-temperature superconductivity.

Verifying a superconductor means showing by measurement that a material has entered the superconducting state and that no ordinary effect accounts for the data. Two signatures are expected together: an electrical resistance that falls to zero, and the diamagnetism of the Meissner effect, in which the material expels a magnetic field from its interior as it becomes superconducting.[1] Supporting measurements cover the response of the transition to a magnetic field and to a change of isotope, the specific heat, the energy gap, and the identity of the compound responsible.

A fall in resistance can come from a structural change in an impurity or from a superconducting filament that links the contacts, and a magnetic signal can depend on how the signal of the apparatus (the background) is subtracted. Claims are tested by whether a second laboratory can make the material and measure the same transition. The measurements are hardest in the High-pressure hydrides, hydrogen-rich compounds held at pressures above one million atmospheres, where the sample is a few tens of micrometers across and its magnetic signal is far smaller than that of the apparatus.

Zero resistance

Resistance is measured with four contacts on the sample, a four-probe measurement. Two carry a known current and two read the voltage between them, which keeps the resistance of the wires and contacts out of the result.

A measurement gives only an upper limit on a resistance, set by the smallest voltage detectable at the current used. Persistent currents give tighter limits: the decay time of a current circulating in a superconducting solenoid was found to exceed 100,000 years (see Superconductivity).[2][3] In the 2015 paper that reported superconductivity at 203 kelvin (K) and 155 gigapascals (GPa; about 1.5 million atmospheres) in compressed hydrogen sulfide (H2S), whose superconducting phase was later identified as H3S, a sample about 25 micrometers across and 1 micrometer thick had a resistivity of at most about 10−11 ohm meters (Ω m), about a hundredth of that of pure copper at the same temperature.[4] In LK-99, a copper-substituted lead apatite claimed in 2023 to superconduct at room temperature and ambient pressure, the resistivity fell tenfold at 378 K (104.8 °C), from about 0.02 to about 0.002 Ω cm (2 × 10−5 Ω m), and did not reach zero.[5] An onset temperature marks where the resistance begins to fall and does not show that it reaches zero.

The voltage contacts report on the best-conducting path between them, so a superconducting filament or a minor phase (a second compound or crystal structure in the sample) that links them gives a reading of zero even if the rest of the sample is not superconducting.[6] In cerium hydride (CeH9) the resistance fell to near zero while magnetic imaging found superconducting regions of 10 micrometers or less in only part of the area that a laser had heated to form the hydride.[1] In crystals of the nickel oxide La3Ni2O7 under pressure, one study found the superconductivity near 80 K to be filamentary,[7] and superconductivity throughout the crystal, with zero resistance and a Meissner effect, was later reported in crystals in which samarium replaces part of the lanthanum.[8]

A four-probe measurement can also return a negative value, which a resistance cannot have. The 2025 preprint that reports resistive onsets of up to 298 K (25 °C) in the hydride LaSc2H24 shows negative resistance between 230 and 277 K in a run at 195 GPa, which its authors call a circuit abnormality and attribute to partial decomposition of the sample. They report checking that all four electrodes remained conductive, to exclude a short circuit.[9]

A superconductor carries current without resistance only up to a critical current, above which the superconducting state is lost.[10] In the yttrium hydride YH6, current-voltage measurements suggested that it may exceed 1.75 amperes at 4 K.[11]

Magnetic evidence

Zero resistance alone would not remove a magnetic field: a steady field inside a perfect conductor would stay there on cooling, because an unchanging field induces no currents. A superconductor cooled through its transition temperature (Tc) in a field pushes the field out.[12][13] Measurements are made after cooling in two ways, and only the second separates the two cases.

Alternating-current susceptibility measurements detect shielding with small coils that apply an oscillating field to the sample. A group at Jilin University that shared no authors with the original 2015 paper used the method on compressed hydrogen sulfide and reported a Tc of 183 K at 149 GPa in a 2016 preprint.[15]

In a trapped-flux measurement a field is applied and removed, and a superconductor that pins flux keeps a magnetic moment in zero field. In H3S at 155 GPa, cooled in zero field, no flux was trapped until the applied field passed about 45 mT, as expected if the sample shields weaker fields completely, and the trapped moment vanished at a transition near 195 K.[16] A ferromagnet such as iron also keeps a moment after a field is removed, and Jorge Hirsch and Frank Marsiglio have argued that the moment measured in these experiments is an experimental artifact or comes from ferromagnetism of the sample or its surroundings.[17][18]

Supporting measurements

A magnetic field weakens superconductivity, so the resistive transition moves to lower temperature as a field is applied. In H3S it shifted downward in fields up to 7 tesla (T).[4] Lutetium hydride with a little nitrogen added (nitrogen-doped) was claimed to superconduct at 294 K (21 °C) and 1 GPa in a paper since retracted.[19][20] A second group's samples showed no transition, and their resistance curves did not shift at 1.6 GPa in fields up to 9 T.[20]

The specific heat, the heat needed to warm a unit mass of a material by one degree, jumps at Tc.[21] The size of the jump shows how much of the sample superconducts: in niobium boride samples whose zero resistance came from filaments of a minor phase, it was about 12% of the theoretical value.[6] A 2020 letter in Physics Today described calorimetry, the measurement of heat, under high pressure as a technical hurdle still to be overcome,[21] and the summaries of evidence for the hydrides published in 2022 and 2024 contain no specific-heat measurement.[22][23]

Superconductivity arises when electrons bind into pairs (Cooper pairs). If lattice vibrations (phonons) do the binding, a heavier isotope usually lowers Tc, the isotope effect. In Bardeen-Cooper-Schrieffer (BCS) theory Tc is proportional to M−α, where M is the atomic mass and α is 0.5. Replacing hydrogen with deuterium, which has twice the mass, gave α of about 0.3 in H3S and 0.4 in YH6.[4][11] A superconductor paired by other means need not show the effect. In palladium hydride, where phonons do the pairing, it is inverse: replacing hydrogen with a heavier isotope raises Tc.[24]

Pairing opens an energy gap, a range of energies in which the material has no electronic states, and tunneling spectroscopy, in which electrons cross a thin barrier into the material, measures it directly. In 2025 tunneling spectra gave a gap (2Δ) of about 60 millielectronvolts (meV) in H3S and about 44 meV in the deuteride D3S, with no states inside it.[25]

Identifying the superconducting phase

In LK-99 the resistance drop came from a copper sulfide impurity, a minor component of the samples, and single crystals of the pure lead apatite were transparent insulators.[5][26]

For H3S, X-ray diffraction, which reveals crystal structure, combined with resistance measurements showed that compressed hydrogen sulfide had decomposed into H3S and elemental sulfur.[27] A later magnetic measurement showed a second transition near 15 K, which its authors assign to the sulfur.[16] X-rays locate the metal atoms of a hydride and barely register hydrogen, so the hydrogen content of LaSc2H24 was inferred from the volume of the crystal's repeating unit.[9]

Reproduction and shared data

For H3S, experiments in Mainz, Los Alamos and Bristol gave transition temperatures that agree within experimental error, as do the upper critical fields measured in two of them.[23] A failed attempt to reproduce a result refutes it only if the second sample is the same material. The refutation of the lutetium hydride claim used samples whose lattice spacing was within 0.1% of the original and found no superconductivity down to 2 K between 0.4 and 40.1 GPa.[20] For LaSc2H24, claimed at 260 GPa, a second group made seven attempts at 250 to 280 GPa, did not obtain the compound and saw no superconductivity between 245 and 300 K, which leaves the claim neither confirmed nor refuted.[9][28]

The 2020 report of superconductivity at 288 K (15 °C) in carbonaceous sulfur hydride was retracted on 26 September 2022 because the background had been subtracted from its magnetic susceptibility data by a "non-standard, user-defined procedure" that the paper did not describe.[29][30] A university investigation later concluded that the leader of the group had fabricated data (see Claims of room-temperature superconductivity).[31]

A 2022 study of the magnetization of H3S and the lanthanum hydride LaH10, their magnetic moment in an applied field, received an author correction in 2023 stating that a linear background had been subtracted from the curves shown, a step its authors said did not move the field at which flux begins to enter the sample.[32] A second correction and an addendum, published on 6 October 2026, describe smoothing, averaging and filtering steps that the paper had left out and make the raw data public.[33][34] The addendum re-evaluates that field at zero temperature as 108 ± 7 mT for H3S and 35 ± 3 mT for LaH10, where the 2022 paper gave 96 ± 2 and 41 ± 2 mT, and its authors state that their main result is unchanged.[14][34]

A sample held above 100 GPa cannot leave its cell, so sharing a sample means shipping the loaded cell. One LaH10 cell lost pressure in transit from a synchrotron X-ray source, falling from 130 to 120 GPa, and its transition moved from 231 K to about 200 K.[16]

What imitates superconductivity

Apparent signature Cause Documented case
Sharp drop in resistance Structural transition of an impurity Cu2S in LK-99: near 385 K (112 °C) in a second group's samples, with no zero resistance.[35]
Resistance anomaly above 200 K Electronic change without superconductivity Nitrogen-doped lutetium hydride, on warming only, unmoved by 3 T.[36]
Diamagnetic step Background subtracted by a procedure the paper did not describe Carbonaceous sulfur hydride.[30]
Partial levitation over a magnet Ferromagnetism LK-99 fragments.[5]

The copper sulfide transition occurred at different temperatures on warming and on cooling, in both resistivity and magnetic susceptibility, which a measurement in one direction would miss.[35]

Megabar pressure

The highest reproduced transition temperature, about 250 K (−23 °C) in LaH10, was measured at about 170 GPa, above one megabar (100 GPa).[37] Such pressure is reached between the flat tips of two diamonds in a diamond anvil cell. In the 2022 magnetization study the tips were about 75 and 90 micrometers across, and the H3S sample was a disk 85 micrometers wide and 2.8 micrometers thick.[14] The sample has about one part in 108 of the mass of the cell.[23]

A magnetometer based on a superconducting quantum interference device (SQUID) measures cell and sample together, and the response of the cell is much larger. The 2015 paper on H3S included such a measurement, and its magnetic transition is the 203 K quoted for the compound.[4] In the 2022 study the cell was first measured with the unreacted starting materials already under pressure, before laser heating turned them into the hydride, and that signal was subtracted.[14] Two newer methods depend less on such a subtraction.

The LaSc2H24 preprint, with samples 10 to 20 micrometers across, contains no measurement of magnetization or susceptibility.[9] One of its authors told Physics World that the work damaged at least 70 pairs of diamonds.[40]

Published standards

A comment published in Nature Reviews Physics in December 2024, written by fifteen researchers who had not worked directly on hydride superconductivity, bases the case for H3S on the agreement between laboratories described above and on magnetization that is diamagnetic before any background is subtracted. It asks that data be made public and that doubts be tested by new experiments.[23] Hirsch replied in 2025 that the magnetization curves published in 2015 and 2022 do not behave as those of a superconductor should and that the raw data behind the 2015 curves are unavailable.[41]

See also

References

  1. P. Bhattacharyya et al. Imaging the Meissner effect in hydride superconductors using quantum sensors. Nature 627, 73 (2024).
  2. J. File and R. G. Mills. Observation of persistent current in a superconducting solenoid. Phys. Rev. Lett. 10, 93 (1963).
  3. D. M. Brink and R. A. Broglia. Nuclear Superfluidity: Pairing in Finite Systems, chapter 1, Introduction. Cambridge University Press, Cambridge, pp. 1-32 (2005).
  4. A. P. Drozdov et al. Conventional superconductivity at 203 kelvin at high pressures in the sulfur hydride system. Nature 525, 73 (2015).
  5. D. Garisto. LK-99 isn't a superconductor — how science sleuths solved the mystery. Nature 620, 705 (2023).
  6. F. Abud et al. Absence of superconductivity in NbB. Phys. Rev. Materials 1, 044803 (2017).
  7. Y. Zhou et al. Investigations of key issues on the reproducibility of high-Tc superconductivity emerging from compressed La₃Ni₂O₇. Matter Radiat. Extremes 10, 027801 (2025).
  8. F. Li et al. Bulk superconductivity up to 96 K in pressurized nickelate single crystals. Nature 649, 871 (2026).
  9. Y. Song et al. Room-temperature superconductivity at 298 K in ternary La-Sc-H system at high-pressure conditions. arXiv:2510.01273 (2025).
  10. C. Yao and Y. Ma. Superconducting materials: Challenges and opportunities for large-scale applications. iScience 24, 102541 (2021).
  11. I. A. Troyan et al. Anomalous high-temperature superconductivity in YH₆. Adv. Mater. 33, 2006832 (2021).
  12. W. Meissner and R. Ochsenfeld. Ein neuer Effekt bei Eintritt der Supraleitfähigkeit. Naturwissenschaften 21, 787 (1933).
  13. C. R. Nave. The Meissner Effect. HyperPhysics, Department of Physics and Astronomy, Georgia State University (2017).
  14. V. S. Minkov et al. Magnetic field screening in hydrogen-rich high-temperature superconductors. Nat. Commun. 13, 3194 (2022).
  15. X. Huang et al. High-temperature superconductivity in sulfur hydride evidenced by alternating-current magnetic susceptibility. Natl. Sci. Rev. 6, 713 (2019).
  16. V. S. Minkov et al. Magnetic flux trapping in hydrogen-rich high-temperature superconductors. Nat. Phys. 19, 1293 (2023).
  17. J. E. Hirsch and F. Marsiglio. Evidence against superconductivity in flux trapping experiments on hydrides under high pressure. J. Supercond. Nov. Magn. 35, 3141 (2022).
  18. J. E. Hirsch and F. Marsiglio. On the interpretation of flux trapping experiments in hydrides. arXiv:2309.14952 (2023).
  19. N. Dasenbrock-Gammon et al. Retraction Note: Evidence of near-ambient superconductivity in a N-doped lutetium hydride. Nature 624, 460 (2023).
  20. X. Ming et al. Absence of near-ambient superconductivity in LuH2±xNy. Nature 620, 72 (2023).
  21. J. C. Ho. Notes on superconducting hydrides. Phys. Today 73 (1), 12 (2020).
  22. M. I. Eremets et al. High-temperature superconductivity in hydrides: experimental evidence and details. J. Supercond. Nov. Magn. 35, 965 (2022).
  23. G. S. Boebinger et al. Hydride superconductivity is here to stay. Nat. Rev. Phys. 7, 2 (2025).
  24. I. Errea, M. Calandra and F. Mauri. First-principles theory of anharmonicity and the inverse isotope effect in superconducting palladium-hydride compounds. Phys. Rev. Lett. 111, 177002 (2013).
  25. F. Du et al. Superconducting gap of H₃S measured by tunnelling spectroscopy. Nature 641, 619 (2025).
  26. P. Puphal et al. Single crystal synthesis, structure, and magnetism of Pb₁₀₋ₓCuₓ(PO₄)₆O. APL Mater. 11, 101128 (2023).
  27. M. Einaga et al. Crystal structure of the superconducting phase of sulfur hydride. Nat. Phys. 12, 835 (2016).
  28. D. V. Semenok et al. Stability analysis of superconductivity in P6/mmm-LaSc₂H₂₄ and its experimental reproducibility from La-Sc alloys. arXiv:2605.29985 (2026).
  29. E. Snider et al. Room-temperature superconductivity in a carbonaceous sulfur hydride. Nature 586, 373 (2020).
  30. E. Snider et al. Retraction Note: Room-temperature superconductivity in a carbonaceous sulfur hydride. Nature 610, 804 (2022).
  31. D. Garisto. Exclusive: official investigation reveals how superconductivity physicist faked blockbuster results. Nature 628, 481 (2024).
  32. V. S. Minkov et al. Author Correction: Magnetic field screening in hydrogen-rich high-temperature superconductors. Nat. Commun. 14, 5322 (2023).
  33. V. S. Minkov et al. Author Correction: Magnetic field screening in hydrogen-rich high-temperature superconductors. Nat. Commun. 17, 10451 (2026).
  34. V. S. Minkov et al. Addendum: Magnetic field screening in hydrogen-rich high-temperature superconductors. Nat. Commun. 17, 10452 (2026).
  35. S. Zhu et al. First-order transition in LK-99 containing Cu₂S. Matter 6, 4401 (2023).
  36. X. Xing et al. Observation of non-superconducting phase changes in nitrogen doped lutetium hydrides. Nat. Commun. 14, 5991 (2023).
  37. A. P. Drozdov et al. Superconductivity at 250 K in lanthanum hydride under high pressures. Nature 569, 528 (2019).
  38. Y. Chen et al. Imaging the Meissner effect in lanthanum hydride using diamond quantum sensors. Proc. Natl. Acad. Sci. U.S.A. 123, e2536915123 (2026).
  39. Q. Hao et al. Diamond quantum sensing at record high pressure up to 240 GPa. arXiv:2510.26605 (2025).
  40. I. Dumé. Ternary hydride shows signs of room-temperature superconductivity at high pressures. Physics World, 7 November 2025 (2025).
  41. J. E. Hirsch. Hydride superconductivity: here to stay, or to lead astray and some day go away? J. Supercond. Nov. Magn. 38, 130 (2025).