Reference › Applications of superconductors

Applications of superconductors

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

Applications of superconductors are the technologies that use materials carrying electric current with zero resistance below a critical temperature, T_c. The largest group is electromagnets: the magnets of magnetic resonance imaging (MRI) scanners, particle accelerators and fusion experiments are wound from superconducting wire. MRI is the largest commercial use: in 2016 more than 3,000 superconducting scanners were being installed each year.^[1][2] Other uses are power cables, generators, magnetic sensors, photon detectors and the circuits of quantum computers.

Every superconductor in service is cooled, from about 0.02 kelvin (K) for quantum circuits to 77 K (−196 °C) for power cables.^[4][3] Removing one watt of heat takes at least 70 W of refrigerator power at 4.2 K and 2.9 W at 77 K. A room-temperature superconductor would need no refrigeration. Used throughout a power grid, it would save at most a few percent of the electricity generated.

Such a material would need a T_c well above 300 K, 375 to 400 K by one estimate.^[5] It would also have to carry a high current density in a magnetic field, be made into kilometers of wire or tape, withstand large forces and be affordable. A material that superconducts above 200 K only at pressures above 100 gigapascals (GPa; about a million atmospheres), as the known High-pressure hydrides do,^[6] cannot be made into a wire or a device.

Magnets

A superconducting coil holds a steady magnetic field without dissipating power in its winding, and the niobium-based superconductors used in magnets carry about a hundred times the current density of copper conductors.^[3] The niobium-titanium (NbTi) alloy in magnet wire can carry 4,000 to 5,000 amperes (A) per square millimeter without resistance at 4.2 K in a field of 5 tesla (T).^[7][3] The wire as a whole carries less, because much of it is copper and it is run below that limit.^[2]

Medical imaging and spectroscopy

By 2016 close to 50,000 MRI scanners were installed worldwide, more than 35,000 of them with superconducting magnets. These magnets are wound from NbTi and run at the temperature of liquid helium, 4.2 K (−269 °C). The winding is closed on itself through superconducting joints, so the current persists without being driven, and the field is guaranteed to decay by less than 0.1 part per million per hour. Almost all whole-body superconducting scanners operate at 1.5 T or 3 T, and a 1.5 T magnet contains 30 to 40 km of wire.^[2]

Nuclear magnetic resonance (NMR) spectrometers, used to study molecular structure, are rated by the frequency at which hydrogen nuclei resonate in their field. The first 1.2 GHz instrument, with a 28.2 T magnet, was installed at the University of Florence in 2020. Fields that high are beyond conventional low-temperature superconductors, and the magnet depends on a high-temperature superconductor.^[8]

Particle accelerators

The Large Hadron Collider (LHC) at CERN, the European particle physics laboratory, bends its proton beams around a 27 km ring with 1,232 dipole magnets, each 15 m long and weighing 35 metric tons.^[9] Their NbTi coils need a current of 11,850 A to reach the full field of 8.3 T, and they operate in superfluid helium at 1.9 K (−271.3 °C).^[10][11] The collider has been shut down since June 2026 for an upgrade expected to last up to four years.^[12]

Accelerators give particles their energy in radio-frequency cavities, hollow metal resonators that hold an oscillating electric field. Superconducting walls lose very little of the radio-frequency power as heat. The European XFEL, an X-ray laser, accelerates electrons to 17.5 gigaelectronvolts with 768 niobium cavities along 1.7 km, cooled to 2 K (−271 °C).^[13]

Fusion

ITER, under construction in southern France, is a tokamak, a fusion device that confines a plasma (a hot ionized gas) with magnetic fields. It has 18 toroidal-field coils of niobium-tin (Nb₃Sn), which produce the main confining field; each weighs 330 metric tons, and the maximum field is 11.8 T. Its central solenoid, which drives the current in the plasma, is also Nb₃Sn and is designed for 13 T, and the poloidal-field coils that shape the plasma use NbTi. The magnets are cooled with helium at about 4 K and contain more than 100,000 km of Nb₃Sn strand.^[14]

A toroidal-field model coil for the SPARC fusion project, built by the Massachusetts Institute of Technology and Commonwealth Fusion Systems, was wound from 270 km of rare-earth barium copper oxide (REBCO) tape, a high-temperature superconductor. It reached 20.1 T at the conductor while carrying 40,500 A at 20 K (−253 °C). The project's paper describes REBCO as the only conductor that met the requirements for field and for heating by fusion neutrons, because a low-temperature superconductor at 4 K can warm very little before it stops superconducting.^[15]

Maglev

The SCMaglev (superconducting magnetic levitation) train of Central Japan Railway Company carries NbTi magnets cooled with liquid helium to 4 K (−269 °C). The magnetic force between them and coils attached to the guideway lifts the train about 10 cm.^[16] A test train reached 603 km/h on 21 April 2015,^[17] and the planned service speed is 500 km/h.^[16]

Electric power

Cables and fault current limiters

AmpaCity, a cable in Essen, Germany, was connected to the city grid in April 2014 for a two-year field test. It was about 1 km long, ran at 10,000 V, was designed for 40 MW and replaced a conventional 110,000 V line between two transformer stations. Its ceramic superconductor was cooled with liquid nitrogen to 73 K (−200 °C).^[18][19] The Karlsruhe Institute of Technology, a project partner, stated that a superconducting cable carries five times as much electricity as a copper cable of the same size.^[19] Cables of this kind work at 65 to 77 K. The first commercial one, a 23,000 V link over 1 km, was completed in South Korea in 2019.^[3]

AmpaCity included a superconducting fault current limiter, a device that protects a grid from short circuits. When a short circuit drives the current above the critical current of the superconducting element, the element becomes resistive, which limits the current, and it recovers once it has cooled.^[20][3] The European Commission's research service reported that the Essen limiter was too expensive to commercialize.^[20]

Generators and motors

Superconducting windings carry more current than copper windings of the same size, which can make a machine of a given power lighter.^[21] The EcoSwing project fitted a 3.6 MW wind turbine in Thyborøn, Denmark, with a rotor of 40 REBCO coils made from more than 20 km of tape. Cryocoolers, closed-cycle refrigerators, were mounted on the rotor and held the coils below their design temperature of 30 K, and the generator fed the grid for more than 650 hours.^[22] The National Aeronautics and Space Administration (NASA) is developing a 1.4 MW aircraft motor with superconducting rotor coils kept at 50 K (−223 °C) and a target of 16 kW per kilogram at 99% efficiency; in 2025 its components were being tested separately.^[21]

Sensors and quantum circuits

A superconducting quantum interference device (SQUID) is a superconducting loop interrupted by Josephson junctions, thin barriers between two superconductors, and is among the most sensitive detectors of magnetic field.^[23][24] SQUIDs made of niobium films and cooled in liquid helium at 4.2 K reach a noise level of 1 to 5 femtotesla per root hertz (1 fT is 10⁻¹⁵ T). That is enough for magnetoencephalography, the recording of the magnetic fields of nerve currents in the brain, which are about 100 fT, on the order of a billionth of Earth's field.^[24]

Superconducting quantum bits (qubits) are circuits built around the same junction. A 2019 processor fabricated in aluminum on silicon ran a computation on 53 qubits,^[25] and one reported in 2024 ran an error-correcting code on 101 qubits.^[26] Qubits resonate near 5 GHz and are operated at about 20 millikelvin (mK), or 0.02 K, so that thermal energy stays far below the energy separating their two states.^[4]

A superconducting nanowire single-photon detector is a thin superconducting wire laid in a winding path. An absorbed photon heats a spot of the wire, which stops superconducting there and produces a voltage pulse. Detectors made mainly of tungsten silicide at the United States National Institute of Standards and Technology (NIST) run at 1 to 2 K and register up to 98% of incoming photons, against about 50% for typical commercial semiconductor detectors.^[27]

Comparison of applications

The table collects the examples above, with the sources given there.

Cooling

Liquid helium boils at 4.2 K and liquid nitrogen at 77 K.^[10][3] A cryocooler can take the place of either liquid. Moving one watt of heat from a temperature T to surroundings at 300 K takes at least (300 − T)/T watts of work, the Carnot limit, which grows as T falls: 70 W at 4.2 K and 2.9 W at 77 K. The step from 4.2 K to 77 K therefore removes 67 of the 70 W, and operation at 300 K would save the remaining 2.9 W. Real refrigerators need more. Small cryocoolers had reached up to 24% of the Carnot efficiency at 80 K by 2002, about 11 W of input per watt of cooling,^[28] and NIST gives about 0.8% for cryocoolers at 4 K, or 1 kW of input for 0.1 W of cooling.^[29] The high-temperature superconductors discovered from 1986 cut refrigeration cost about tenfold compared with helium-cooled ones.^[1] The cryogenic system of the LHC holds 120 metric tons of helium and draws about 40 MW of electricity.^[10]

MRI, with its helium-cooled NbTi magnets, accounted for about 20% of helium use in 2016. Conductors with a higher T_c would simplify the cryogenics, but a review published late that year found that none of them met the minimum specification for commercial MRI magnets and that their cost was a barrier.^[2] A higher T_c would not change the cooling of quantum circuits, which is set by the energy separating the qubit states.^[4]

A 2002 review noted that power applications of superconductors need a kilowatt or more of refrigeration at 70 K and called the cost of cryocoolers a serious problem for many applications.^[28] The EcoSwing rotor took 14 days to cool down and needed a rotating coupling to connect its coolers to their stationary compressors.^[22]

What room-temperature operation would change

A conductor that worked at 300 K would need no refrigeration and no cryostat, the insulated vessel that holds the cold parts. Motors and generators would carry no cold rotating parts, and MRI magnets would need no helium. Magnetic sensors would not need the cooling system that limits their portability, although SQUIDs made of high-temperature superconductors are so far less sensitive than low-temperature ones.^[24]

In the United States, transmission and distribution losses averaged about 5% of the electricity carried from 2018 through 2022.^[30] The world figure for 2024, in International Energy Agency data published by the World Bank, was about 6.5% of output, a total that includes theft.^[31] Only part of the technical loss is resistive: grid operators separate variable losses, the heating of cables and transformer windings by the current they carry, from fixed losses, which occur in transformers and other equipment whenever it is energized.^[32] A superconductor removes most of the first kind, and only in the conductors it replaces; under alternating current it still dissipates some power as the changing magnetic field penetrates it.^[33] Replacing every conductor in a grid would therefore save at most a few percent of the electricity generated.

Requirements other than critical temperature

Current in a magnetic field

The superconductors used for wire are of type II: a magnetic field enters them as threads of flux called vortices. Current flows without loss only while the vortices stay pinned to defects, which holds up to the irreversibility field. In the cuprate YBa₂Cu₃O₇ at 77 K that field is about half the upper critical field, at which superconductivity ends, and in the bismuth cuprate Bi-2223 it is far lower.^[7] Bi-2223 wire therefore has to be cooled to about 30 K, and YBa₂Cu₃O₇ wire to about 50 K, to be used in a strong field.^[1]

Margin below the critical temperature

The critical current density falls to zero at T_c, so conductors run well below it.^[1] NbTi has a T_c of about 9.5 K and runs at 1.9 K in the LHC.^[3][10] YBa₂Cu₃O₇, the original REBCO compound, has a T_c of 93 K,^[34] and REBCO ran at 20 K in the SPARC coil.^[15] Both high-field magnets operate at about a fifth of T_c. In weaker fields conductors run nearer to it: NbTi in MRI magnets at 4.2 K, a little under half of its T_c, and cables, in fields below 1 T, at 65 to 77 K.^[2][3] One review puts the T_c needed for use at room temperature at 375 to 400 K, and higher for large currents.^[5]

Thermal fluctuations

Heat shakes vortices loose from their pinning sites, which makes flux pinning harder as the temperature rises. A 2006 United States Department of Energy workshop estimated, by scaling from known materials, that a room-temperature superconductor would have a coherence length below 1 nm and fluctuation effects rising roughly as the fourth power of T_c, and that a strongly layered one would probably be ineffective for power applications.^[1] A 2017 study proposed a lower limit on the rate at which a pinned current decays (flux creep), valid at low temperature and low field. The limit rises with temperature and with the strength of thermal fluctuations in the material, and the authors concluded that any superconductor with a high T_c will show fast decay.^[35] Both estimates extrapolate from materials with far lower T_c.

Wire

Applications need conductor by the kilometer: 30 to 40 km in an MRI magnet and 270 km in the SPARC coil.^[2][15] In cuprates the current that crosses a boundary between crystal grains falls exponentially with their misalignment, on a scale of 3 to 5 degrees. REBCO is therefore grown as an aligned film on a metal tape.^[7] The film is 1 to 2 micrometers thick, about 1% of the thickness of the finished conductor, which is 0.1 to 0.2 mm.^[36]

Strength and cost

The conductor in the SPARC coil bore a magnetic load of more than 800 kilonewtons per meter.^[15] In 2016 NbTi cost about 1 dollar for enough wire to carry 1,000 A over one meter at 4.2 K and 4 T.^[2] In 2021, for use at 4.2 K and 10 T, Nb₃Sn cost about 5 dollars on the same measure and REBCO 100 to 200 dollars, against 25 dollars regarded as ideal for large-scale use.^[3]

Superconductors that need high pressure

The highest reproduced T_c, about 250 K in lanthanum hydride (LaH₁₀), occurs at about 170 GPa (about 1.7 million atmospheres).^[6] In the laboratory such static pressure is reached between the tips of two diamonds in a diamond anvil cell, and the sample is small: the sulfur hydride (H₃S) sample used for magnetic measurements at 155 GPa was a disk about 85 micrometers across and 2.8 micrometers thick.^[37] No wire or coil can be made in such a cell, so a material that superconducts only at megabar pressure has no application. A usable material has to exist at ambient pressure and remain stable at room temperature.

See also

• Room-temperature superconductor
• Superconductivity
• Superconducting materials
• High-pressure hydrides
• Glossary

References

1. U.S. Department of Energy, Office of Basic Energy Sciences. Basic research needs for superconductivity: report of the Basic Energy Sciences Workshop on Superconductivity, May 8-11, 2006. U.S. DOE Office of Science technical report, OSTI 899129 (2006).
2. M. Parizh, Y. Lvovsky and M. Sumption. Conductors for commercial MRI magnets beyond NbTi: requirements and challenges. Supercond. Sci. Technol. 30, 014007 (2017).
3. C. Yao and Y. Ma. Superconducting materials: Challenges and opportunities for large-scale applications. iScience 24, 102541 (2021).
4. P. Krantz et al. A quantum engineer's guide to superconducting qubits. Appl. Phys. Rev. 6, 021318 (2019).
5. W. E. Pickett. Colloquium: Room temperature superconductivity: the roles of theory and materials design. Rev. Mod. Phys. 95, 021001 (2023).
6. A. P. Drozdov et al. Superconductivity at 250 K in lanthanum hydride under high pressures. Nature 569, 528 (2019).
7. A. Gurevich. To use or not to use cool superconductors? Nat. Mater. 10, 255 (2011).
8. P. Wikus, W. Frantz, R. Kümmerle and P. Vonlanthen. Commercial gigahertz-class NMR magnets. Supercond. Sci. Technol. 35, 033001 (2022).
9. CERN. Pulling together: Superconducting electromagnets. CERN (web page) (2026).
10. CERN. Cryogenics: Low temperatures, high performance. CERN (web page) (2026).
11. C. Krishna. Superconductivity for sustainability: a new superconducting link for the High-Luminosity LHC. CERN News, 3 March 2023 (2023).
12. CERN. The road to HiLumi. CERN (web page on Long Shutdown 3) (2026).
13. European XFEL. Accelerator. European XFEL (web page) (2026).
14. ITER Organization. Magnets. ITER Organization (web page) (2026).
15. Z. S. Hartwig et al. The SPARC toroidal field model coil program. IEEE Trans. Appl. Supercond. 34, 0600316 (2024).
16. N. Ueno. Superconducting Maglev – Development and Progress Toward Revenue Service. Plenary presentation abstract, Applied Superconductivity Conference 2014, IEEE Council on Superconductivity (2014).
17. Guinness World Records. Fastest maglev train. Guinness World Records (record entry for 21 April 2015) (2015).
18. Karlsruhe Institute of Technology. World's Longest Superconductor Cable. KIT Press Release 012/2012, 19 January 2012 (2012).
19. Karlsruhe Institute of Technology. Operation of Longest Superconducting Cable Worldwide Started. KIT Press Release 050/2014, 30 April 2014 (2014).
20. CORDIS, European Commission. Superconducting fault current limiter protects national electricity grids. CORDIS EU research results, Publications Office of the European Union, updated 23 April 2021 (2021).
21. NASA. High-Efficiency Megawatt Motor (HEMM). NASA Electrified Aircraft Propulsion (web page), updated 13 April 2025 (2025).
22. A. Bergen et al. Design and in-field testing of the world's first ReBCO rotor for a 3.6 MW wind generator. Supercond. Sci. Technol. 32, 125006 (2019).
23. National Institute of Standards and Technology. Magnetic Attraction: Physicists Pay Homage to the SQUID at 50. NIST News, 11 March 2014 (2014).
24. C. Bonavolontà, A. Vettoliere, P. Sorrentino and C. Granata. Superconducting Quantum Magnetometers for Brain Investigations. Sensors 25, 4625 (2025).
25. F. Arute et al. Quantum supremacy using a programmable superconducting processor. Nature 574, 505 (2019).
26. Google Quantum AI and Collaborators (R. Acharya et al.). Quantum error correction below the surface code threshold. Nature 638, 920 (2025).
27. National Institute of Standards and Technology. Superconductive Nanowire Single-Photon Detectors. NIST (web page), created 18 August 2025, updated 25 September 2026 (2025).
28. R. Radebaugh. Refrigeration Methods for Superconductors. in Handbook of Superconducting Materials, Vol. 1 (Taylor & Francis, London); abstract in the NIST publication record (2002).
29. National Institute of Standards and Technology. Low Temperature Regenerator and Pulse Tube Losses. NIST project page, created 23 April 2009, updated 26 March 2025 (2009).
30. U.S. Energy Information Administration. How much electricity is lost in electricity transmission and distribution in the United States? EIA Frequently Asked Questions, updated 7 November 2023 (2023).
31. World Bank. Electric power transmission and distribution losses (% of output). World Bank Open Data, indicator EG.ELC.LOSS.ZS, from International Energy Agency statistics (2026).
32. National Grid Electricity Distribution. What causes Losses? National Grid (web page) (2026).
33. F. Grilli. Compact HTS cables for low-loss AC power transmission: possibilities and limits. Abstract of a talk at the 16th European Conference on Applied Superconductivity (EUCAS 2023, Bologna, 3 to 7 September 2023), Institute for Technical Physics, Karlsruhe Institute of Technology (web page) (2023).
34. M. K. Wu et al. Superconductivity at 93 K in a new mixed-phase Y-Ba-Cu-O compound system at ambient pressure. Phys. Rev. Lett. 58, 908 (1987).
35. S. Eley et al. Universal lower limit on vortex creep in superconductors. Nat. Mater. 16, 409 (2017).
36. National Academies of Sciences, Engineering, and Medicine. The current status and future direction of high-magnetic-field science and technology in the United States. The National Academies Press, Washington, DC; chapter 5, Accelerator magnets, and chapter 6, Superconducting materials and wires/tapes for high-field magnets (2024).
37. V. S. Minkov et al. Magnetic field screening in hydrogen-rich high-temperature superconductors. Nat. Commun. 13, 3194 (2022).