How Superconductors Actually Work (and Why They Levitate)

Drop a small magnet over a chilled ceramic disc and something happens that looks like a magic trick: the magnet stops, hangs in mid-air, and refuses to fall. Nudge it sideways and it springs back. Spin it and it keeps spinning, locked at a fixed height as if held by an invisible hand. You are watching how superconductors work in their most theatrical form — but the floating magnet is only the surface of a far stranger story. Underneath it lies a quantum state of matter in which electrons stop behaving like individuals, pair up, and march in perfect lockstep, carrying electric current forever without losing a single joule to resistance. The levitation is just the visible shadow of that hidden order.

This article unpacks the physics from the ground up: why ordinary metals resist current, how pairs of electrons conspire to dodge that resistance, why a superconductor actively throws magnetic fields out of its body, and why — despite a century of progress — a superconductor that works at room temperature and normal pressure remains stubbornly out of reach.

What this covers: the discovery and the two defining properties; the BCS mechanism and Cooper pairs; the Meissner effect and flux pinning behind levitation; high-temperature cuprates and the open puzzle; the room-temperature quest; and where superconductors already matter.

Context and Background

Superconductivity was discovered by accident. In 1911, the Dutch physicist Heike Kamerlingh Onnes — who had recently become the first person to liquefy helium, reaching about 4 kelvin (−269°C) — was measuring how the electrical resistance of metals behaves as they get extremely cold. The expectation was that resistance would fall smoothly and perhaps level off. Instead, when he cooled solid mercury below 4.2 K, its resistance did not just drop. It vanished. Within a tiny temperature interval the measured resistance collapsed to something indistinguishable from zero. Onnes called the new state “superconductivity,” and the discovery earned him the 1913 Nobel Prize in Physics.

The first hallmark of a superconductor is therefore exactly zero DC resistance below a material-specific critical temperature, written Tc. This is not “very low” resistance, the way copper has low resistance. Experiments have run persistent currents in superconducting loops for years with no measurable decay; the theoretical lifetime of such a current is astronomically long. Below Tc, a direct current circulates as if friction has been switched off entirely.

The second hallmark took 22 more years to find. In 1933, Walther Meissner and Robert Ochsenfeld discovered that a superconductor does not merely fail to admit a magnetic field — it expels one that is already there. Cool a metal through Tc while it sits in a magnetic field, and the field is pushed out of the interior. This active expulsion, the Meissner effect, is a separate and deeper property than zero resistance, and it is what makes magnetic levitation possible. A material that has both zero resistance and full field expulsion is what physicists mean by a true superconductor.

For decades these two facts sat without explanation. Classical physics had no mechanism for them, and even early quantum theory struggled. The puzzle was only resolved in 1957, and the resolution turned out to be one of the most elegant arguments in twentieth-century physics. (For another tour of counterintuitive quantum behavior, see our explainer on quantum error correction and surface codes. For the historical record of Onnes’s work, the Nobel Prize in Physics 1913 archive is the primary source.)

Cooper Pairs and the BCS Mechanism

Superconductivity arises because, below Tc, electrons stop traveling as independent particles and instead bind into weakly coupled pairs — Cooper pairs — that condense into a single, coherent quantum state. That state flows through the metal without scattering, which is why resistance disappears. The binding is mediated not by direct attraction (electrons repel each other) but by the vibrating lattice of positive ions, and the theory that explains it is named BCS after Bardeen, Cooper, and Schrieffer, who published it in 1957 and shared the 1972 Nobel Prize for it.

Figure 1: A passing electron tugs the positive ions of the lattice slightly toward it, leaving a transient region of concentrated positive charge in its wake. A second electron is drawn to that region. The lattice distortion thus glues two electrons into a Cooper pair — an indirect attraction strong enough, at low temperature, to overcome their direct Coulomb repulsion.

To see why this is necessary, we first have to understand why normal metals resist current at all.

Why ordinary metals resist current

In a normal metal, electric current is a flow of electrons through a lattice of positive ions. Resistance is what happens when those electrons get scattered out of their orderly drift. Two things scatter them. The first is the thermal vibration of the lattice: the ions are never still, and their vibrations are quantized into packets of sound-like energy called phonons. An electron moving through the metal collides with phonons, transfers some momentum, and veers off course. The hotter the metal, the more phonons, the more scattering — which is why a metal’s resistance rises with temperature. The second cause is static imperfection: impurity atoms, missing ions, grain boundaries, and dislocations. Even at absolute zero, where the phonon contribution would vanish, these defects leave a floor of “residual resistance.”

Every scattering event randomizes an electron’s momentum and dumps a little of its directed energy into heat. Multiply that across the trillions of electrons carrying a current and you get Joule heating — the reason power lines lose energy and laptop chargers get warm. The remarkable thing about a superconductor is that, below Tc, this scattering stops mattering. The question is how electrons manage to ignore the very obstacles that resist them above Tc.

A useful clue hid in plain sight for years: the isotope effect. In the early 1950s, experimenters found that the critical temperature of a superconductor depends on the mass of its atomic nuclei. Swap a heavier isotope into the lattice and Tc drops, scaling roughly with the inverse square root of the ion mass. Heavier ions vibrate more slowly, so this was a direct fingerprint that the lattice vibrations — phonons — are not an incidental nuisance but the active ingredient. The very thing that causes resistance above Tc turns out to be the thing that, below Tc, glues electrons together. That paradox is the clue BCS theory had to explain.

The phonon-mediated attraction

The answer, worked out by Leon Cooper in 1956, is that electrons can attract one another — indirectly, through the same lattice that normally scatters them. Picture one electron speeding through the crystal. Its negative charge tugs the nearby positive ions inward, just slightly. The ions are heavy and sluggish, so they keep drifting together for a moment after the electron has moved on, briefly creating a small pocket of extra positive charge in the electron’s wake. A second electron, passing later, feels that pocket of positive charge and is drawn toward it. Net result: through the intermediary of the lattice distortion, two electrons that should repel end up weakly attracting each other. In the language of quantum field theory, the two electrons exchange a phonon.

This attraction is feeble — far weaker than the bare Coulomb repulsion between two electrons sitting side by side. But the two interactions operate on different scales. The Coulomb repulsion is strong and short-ranged; in a metal it is also heavily screened by the sea of other electrons. The phonon-mediated attraction is delicate but it acts between electrons of opposite momentum and opposite spin that are far apart in space and close together in energy, right at the surface of the “Fermi sea” of occupied electron states. Cooper showed mathematically that if there is any net attraction at all, no matter how small, the normal state of the electron sea becomes unstable: it is energetically favorable for electrons near the Fermi surface to bind into pairs. There is no minimum threshold; an arbitrarily weak attraction is enough.

A Cooper pair is therefore not two electrons stuck together like a molecule. The two partners are typically separated by hundreds of nanometers — a distance spanning thousands of other electrons. The pairing is a correlation in momentum space, a partnership in how the electrons move rather than where they sit. This characteristic separation, the coherence length, is enormous compared with the spacing between atoms, which is why the pairs heavily overlap: at any instant a given region of the metal contains a tangle of millions of mutually interpenetrating pairs, all sharing the same quantum bookkeeping. That overlap is precisely what makes the resulting state so robust and so coherent — it is not a gas of separate pairs but a single interwoven quantum fabric.

The energy gap and the coherent condensate

Bardeen, Cooper, and Schrieffer’s 1957 breakthrough was to handle not one pair but all of them at once, building a single quantum wavefunction in which every electron near the Fermi surface is paired. Two consequences follow, and together they explain everything.

First, pairing opens an energy gap. Breaking a Cooper pair costs a minimum amount of energy, 2Δ, because you must pull both partners out of the condensate. At low temperature there simply isn’t enough thermal energy lying around to pay that price. Scattering an electron requires changing its state by a tiny amount of energy — exactly the kind of small nudge a phonon or impurity delivers — but the gap forbids small nudges. There is no nearby state to scatter into. So the usual scattering processes are switched off, and current flows without resistance. As you warm the material toward Tc, thermal energy grows until pairs start breaking; at Tc the gap closes, the pairs dissolve, and ordinary resistance returns.

Second, and more subtly, Cooper pairs are bosons in their collective behavior. A single electron is a fermion, and the Pauli exclusion principle forbids two fermions from occupying the same quantum state. But a pair of electrons has integer total spin, and a system of pairs can pile into one shared quantum state — a macroscopic condensate described by a single wavefunction with a single, well-defined phase across the entire sample. Once trillions of pairs share one wavefunction, they move as one. To scatter the current you would have to scatter the whole condensate at once, which is statistically impossible. The supercurrent is rigid, coherent, and frictionless. The BCS theory predicts the size of the gap, the value of Tc, and the way both depend on the lattice — and for the classic low-temperature metals those predictions match experiment with striking precision.

The Meissner Effect, Flux Pinning, and Levitation

A superconductor expels magnetic fields from its interior — it is a perfect diamagnet, not merely a perfect conductor — and this active expulsion, the Meissner effect, is what lets a magnet hover. Type II superconductors take it further: they let the field thread through as quantized vortices that get pinned in place, which is why a magnet does not just float but locks rigidly in three dimensions above a cooled puck.

It is tempting to say “zero resistance explains levitation,” but that is wrong, and the distinction is the heart of the physics.

Figure 2: An idealized perfect conductor only opposes changes in magnetic field, so a field present before cooling stays frozen inside it. A real superconductor actively expels the field from its interior even if the field was already there when it was cooled — the Meissner effect, a genuine thermodynamic property, sustained by persistent surface currents that cancel the field within.

Perfect conductor versus superconductor

Imagine a hypothetical material with literally zero resistance but no other special properties — a “perfect conductor.” By Faraday’s law and Lenz’s law, a changing magnetic field induces currents that oppose the change, and with zero resistance those currents never decay, so they perfectly cancel any change in field. The catch: this only freezes the field at whatever value it had when resistance vanished. Cool such a material in a field, and the field stays trapped inside. Apply a field afterward, and it is blocked. The internal field depends on the history.

A real superconductor does something stronger and history-independent. Cool it through Tc while a field is already present, and the field is actively pushed out of the interior. The final state has zero field inside regardless of the path taken to get there. That cannot be explained by zero resistance alone; it is a distinct property — perfect diamagnetism — that the BCS condensate produces because the paired electrons set up persistent screening currents in a thin surface layer that exactly cancel the applied field within the bulk. The field does penetrate a short distance into that surface layer, falling off exponentially over a characteristic distance called the London penetration depth, typically tens to hundreds of nanometers. Beyond that skin, the interior is field-free. The Meissner effect, not zero resistance, is the true signature of superconductivity, and it is why a magnet can be supported: the superconductor behaves like a magnetic mirror, generating an image field that repels the magnet above it.

The reason field expulsion is thermodynamic rather than circumstantial comes back to the energy gap and the shared wavefunction. The condensate has a single, rigid quantum phase across the whole sample, and a magnetic field threading the interior would force that phase to wind in ways the coherent state will not tolerate at low field. It is energetically cheaper for the superconductor to summon screening currents and throw the field out than to let it in. The London brothers, Fritz and Heinz, captured this in 1935 with a pair of equations that predicted both the screening currents and the exponential penetration depth before BCS theory even existed — a reminder that the macroscopic behavior was being correctly described decades before the microscopic mechanism was understood.

Type I, Type II, and flux vortices

How a superconductor responds to a stronger field splits the family into two types.

Property	Type I	Type II
Field response	Full Meissner expulsion up to one critical field Hc, then abrupt loss of superconductivity	Meissner below Hc1; field enters as vortices between Hc1 and a much higher Hc2
Typical materials	Pure elemental metals (lead, tin, mercury, aluminium)	Alloys and compounds (niobium-titanium, niobium-tin, all cuprates)
Mixed state	None	Yes — a vortex lattice of normal cores in a superconducting sea
Useful for high-field magnets and levitation	No — quenches at low field	Yes — survives to very high fields; supports flux pinning

Type I superconductors expel the field completely until a single, relatively low critical field is reached, at which point superconductivity collapses all at once. They are useless for strong magnets and for stable levitation. Type II superconductors — which include every practical superconductor and every high-temperature one — behave differently above a lower critical field Hc1. Instead of quenching, they let the magnetic field punch through the body in thin, discrete tubes called flux vortices. Each vortex is a whisker of normal, non-superconducting material, and the magnetic flux carried by each one is quantized: it always equals one flux quantum, a fixed bundle of magnetism set by fundamental constants. Around each normal core, a swirl of supercurrent circulates. The bulk between the vortices stays superconducting, so the material keeps zero resistance to far higher fields, all the way up to a second, much larger critical field Hc2.

Why the magnet locks in place

Figure 3: Above Hc1, the magnetic field pierces a Type II superconductor as a lattice of quantized flux vortices, each with a non-superconducting core. Crystal defects and impurities trap — pin — those cores in fixed positions. Because moving the magnet would force the vortices to move and cost energy, the magnet is held rigidly in three dimensions: this is flux pinning, the mechanism behind quantum levitation.

There is a beautiful subtlety in that quantization. The amount of magnetic flux each vortex carries is not arbitrary — it is fixed at exactly one flux quantum, equal to Planck’s constant divided by twice the electron charge. The factor of two is itself evidence of pairing: the current circulating around a vortex is carried by Cooper pairs of charge 2e, not single electrons, and the measured flux quantum confirms it. The vortex lattice is a macroscopic object whose very geometry is dictated by a fundamental constant of nature, made visible at the millimeter scale.

This is where stable levitation, often called “quantum locking,” comes from. In a real Type II crystal, the vortices do not float freely. Wherever the material has a tiny defect — an impurity atom, a grain boundary, an oxygen vacancy — it is energetically cheaper for a vortex core to sit there, because the core is already non-superconducting and a defect is already a disruption. The defects therefore pin the vortices in place. Once pinned, the vortices form a frozen, three-dimensional map of exactly how the magnetic field threaded the sample at the moment it was cooled.

Now try to move the magnet. To shift it you would have to drag the vortices to new positions, ripping them off their pinning sites, which costs energy. So the system resists: push the magnet down, the vortices compress and push back; pull it up, they resist leaving; slide it sideways, they hold their lattice. The magnet is locked in all directions, even hanging below the superconductor if you cool it that way. Pure Meissner repulsion alone gives an unstable, slippery float (a consequence of Earnshaw’s theorem, which forbids stable static levitation by repulsion alone); it is flux pinning in a Type II material that turns the float into a rigid lock. The dramatic tabletop demos with a small magnet hovering and orbiting above a chilled yttrium-barium-copper-oxide (YBCO) puck are flux pinning made visible.

Trade-offs and Why Room-Temperature Is So Hard

If superconductivity is so powerful, why isn’t it everywhere? Because every known superconductor demands punishing conditions, and the trade-offs are brutal.

Figure 4: The known superconductor families. Conventional BCS metals work only below about 30 K at ambient pressure. The cuprates reach roughly 135 K at ambient pressure but are brittle ceramics with a mechanism still not fully understood. Hydrogen-rich compounds reach near room temperature — but only when squeezed to millions of atmospheres, far from anything practical.

The first cost is cooling. Conventional BCS superconductors like niobium-titanium need liquid helium, around 4 K, which is expensive, scarce, and finicky. The 1986 discovery of cuprate superconductors by Bednorz and Müller (Nobel Prize 1987) was revolutionary because some cuprates superconduct above 77 K, the boiling point of liquid nitrogen — a coolant that costs less than milk. But 77 K is still −196°C. Any application has to wrap the superconductor in a cryostat, which adds bulk, weight, and running cost.

The second cost is materials. The high-Tc cuprates that cleared the nitrogen barrier are brittle ceramics. You cannot easily draw a ceramic into the long, flexible, kilometer-scale wires that power cables and large magnets require; making usable cuprate tape is a feat of materials engineering. And there is a deeper problem: nearly 40 years after their discovery, the mechanism of high-temperature superconductivity in the cuprates is still not fully understood. The simple phonon-mediated BCS picture does not cleanly account for their high transition temperatures, and the pairing “glue” in these strongly correlated materials remains one of the major open questions in condensed-matter physics.

The third frontier is pressure. Theory has long suggested that hydrogen-rich compounds could superconduct near room temperature because light hydrogen atoms vibrate at very high frequencies, giving strong electron-phonon coupling. Experiments on hydrides such as hydrogen sulfide and lanthanum hydride have reported superconductivity at remarkably high temperatures — but only under megabar pressures, millions of atmospheres, generated inside a diamond anvil cell. A material that superconducts at 250 K but only when crushed harder than the center of the Earth is a triumph of physics and useless as a wire.

This is also the right place to be sober about hype. In 2023, a Korean group announced LK-99, a copper-doped lead-apatite, as a room-temperature, ambient-pressure superconductor. The claim spread explosively. Within weeks, laboratories worldwide tried to reproduce it, and the consensus that emerged was that LK-99 is not a superconductor: the apparent partial levitation and resistance drops were traced to ferromagnetic impurities and a sharp transition in copper sulfide present in the samples, not to superconductivity. LK-99 is now the textbook cautionary tale: extraordinary claims in this field demand independent reproduction, magnetization and Meissner-effect measurements, and not just a video of something twitching over a magnet.

Practical Takeaways

Superconductors are already woven into modern technology, despite the cooling burden, because for some jobs nothing else comes close. The biggest deployment is MRI scanners: nearly every hospital MRI uses a superconducting niobium-titanium magnet to produce the intense, ultra-stable magnetic field that images the body. Maglev trains use superconducting magnets for friction-free levitation and propulsion. The world’s largest particle accelerator, the LHC, steers its beams with superconducting magnets, and the new generation of fusion reactors — both tokamaks and compact designs — relies on high-temperature superconducting tapes to generate the colossal fields that confine the plasma. (See our companion piece on how tokamak fusion reactors confine plasma.) Superconducting circuits are also the leading hardware for many quantum computers, where Cooper-pair-based qubits exploit the macroscopic quantum coherence directly. A handful of cities run superconducting power cables that carry far more current than copper in the same footprint.

What actually makes a superconductor practical? A short checklist:

• A critical temperature high enough to reach with affordable cooling (ideally above 77 K, liquid nitrogen).
• Operation at ambient or near-ambient pressure — no diamond anvil required.
• A high critical field and critical current density, so it stays superconducting under real magnetic and electrical loads.
• Strong flux pinning, so vortices stay put and the material doesn’t dissipate energy under field.
• Mechanical workability into wires or tapes, plus chemical stability over years of thermal cycling.

The dream is a material that checks every box at room temperature and one atmosphere. That would transform power grids, transport, and computing. We are not there. But the path from Onnes’s vanishing mercury to liquid-nitrogen ceramics shows a field that keeps finding the impossible to be merely difficult.

Frequently Asked Questions

Why does a magnet float above a superconductor?

Two effects combine. The Meissner effect makes the superconductor expel the magnet’s field, creating a repulsive image field that pushes the magnet up. In a Type II superconductor, the field also enters as quantized flux vortices that get pinned by defects in the crystal — flux pinning. Pure repulsion alone would be unstable, but pinning locks the vortices, and therefore the magnet, rigidly in three dimensions. That is why the magnet doesn’t slide off and can even hang below the superconductor.

Are room-temperature superconductors real?

Not at normal pressure, no. Some hydrogen-rich compounds show superconductivity near room temperature, but only under millions of atmospheres of pressure inside a diamond anvil — completely impractical. At ambient pressure the record remains the cuprates, around 135 K (−138°C). High-profile ambient claims like LK-99 in 2023 failed independent reproduction and were explained by impurity effects, not superconductivity.

What is the difference between a superconductor and a perfect conductor?

A perfect conductor merely has zero resistance, which freezes any magnetic field at the value present when it was cooled. A superconductor additionally expels magnetic field from its interior — the Meissner effect — regardless of the field’s history. Perfect diamagnetism, not just zero resistance, is the defining signature of superconductivity.

What are Cooper pairs?

Cooper pairs are loosely bound pairs of electrons that form below the critical temperature. The binding comes indirectly from the lattice: one electron distorts the positive-ion lattice, and a second electron is attracted to the resulting pocket of positive charge. The pairs behave collectively as bosons and condense into a single coherent quantum state that carries current without scattering.

Why do superconductors need to be so cold?

Cooper pairs are held together very weakly, so any thermal energy can break them. Above the critical temperature, random thermal motion shatters the pairs faster than they can form, the energy gap closes, and ordinary resistance returns. Raising the critical temperature means finding a stronger pairing mechanism — which is exactly the hard, partly unsolved problem at the center of the field.

Does a superconductor really carry current forever?

For practical purposes, yes, for direct current. Persistent currents in superconducting loops have been observed without measurable decay over years, and the theoretical decay time is longer than the age of the universe. Alternating currents and time-varying fields do cause some losses, especially from vortex motion, which is one reason flux pinning and material quality matter so much in real devices.

Further Reading

• How tokamak fusion reactors work: plasma confinement physics — where high-temperature superconducting magnets do their heaviest lifting.
• How GPS actually works and why relativity matters — another case of subtle physics hiding inside everyday technology.
• Quantum error correction and surface codes explained — the quantum-coherence story that superconducting qubits depend on.
• External: the Nobel Prize in Physics 1972 archive for Bardeen, Cooper, and Schrieffer’s BCS theory.
• External: CERN’s overview of superconductivity and accelerator magnets for a real-world high-field application.

By Riju — about