Superconducting Qubit
A circuit so cold and quiet that electricity learns to flow in two directions at once.
Holds a single bit of quantum information — but unlike an ordinary bit, it can be 0, 1, or a precisely-controlled blend of both. Hundreds or thousands of qubits networked together let physicists run algorithms that classical computers can't simulate efficiently.
A loop of superconductor (typically niobium or aluminum, cooled to roughly 10 mK) with a tiny break called a Josephson junction. Below its critical temperature, current flows without resistance and the loop's energy levels become discrete, like an atom's. Microwave pulses tuned to the gap between the two lowest levels rotate the qubit — that's a quantum gate.
In plain terms
Imagine a tiny swing that can sit still, swing forward, or somehow swing forward AND backward at the same time until you peek. That last trick is the only thing a regular swing can't do — and the only thing a qubit is for.
Also called: transmon, Josephson qubit.
Why it matters
- Manufacturable using thin-film techniques borrowed from regular chip fabs.
- Coherence times have grown from nanoseconds in 1999 to hundreds of microseconds today — a millionfold improvement.
- Powers IBM, Google, and Rigetti's quantum processors.
- Programmable via ordinary microwave electronics, not exotic optics.
Where it lives
- IBM Heron and Condor processors
- Google Sycamore (quantum-supremacy demo, 2019)
- Rigetti Ankaa-class chips
Key numbers
- Operating temp: ~10 mK (30,000× colder than the cosmic microwave background)
- Drive frequency: 4–8 GHz (Microwave, same band as Wi-Fi)
- Coherence (T₂): 100–500 µs (State-of-the-art, 2024)
- Gate time: 20–50 ns
Surprising facts
- A typical superconducting qubit lives at about 10 millikelvin — colder than deep space, which sits at roughly 2.7 K.
- When the qubit is in superposition it isn't 'somewhere between' 0 and 1 — it genuinely has both answers ready, and the one you measure depends on probability.
- The Josephson junction at its heart is only a few nanometers thick — a film of insulator the width of about ten atoms.
- If you tap the chip's package with a fingernail, the heat from your fingertip is enough to scramble every qubit on the wafer.
Open questions
- Two-level system jumps: Inside the amorphous oxide of every Josephson junction lurk atomic-scale defects called two-level systems (TLS). These defects randomly flip their quantum state and drag the qubit with them, causing relaxation events that appear at rates no single microscopic model fully predicts. The exact identity of the tunneling entity — whether a single atom, a dangling bond, or a hydrogen impurity — remains unresolved despite decades of spectroscopy.
- Unexplained 1/f flux noise: Superconducting qubits and SQUIDs both exhibit magnetic flux noise whose power falls as 1/f over thirteen orders of magnitude in frequency — a span from sub-hertz to gigahertz. The noise amplitude barely changes regardless of the metal used, the substrate, or the device area. A consensus has emerged attributing it to unpaired surface spins, but the microscopic identity of those spins and why they produce such a universal signature remains an open question.
- Non-equilibrium quasiparticle mystery: Even at 10 millikelvin — a temperature where all thermal quasiparticles should have frozen out of aluminum — a residual density of unpaired electrons persists inside superconducting circuits and sporadically tunnels across Josephson junctions, causing sudden qubit flips. The source of this excess quasiparticle population (cosmic rays, infrared photons, phonons from vibrations) is still actively debated, and no single mitigation has eliminated it.
A short history
The transmon — today's dominant superconducting qubit design — was proposed by Jens Koch and colleagues at Yale in 2007 to suppress charge noise that limited earlier 'Cooper-pair box' qubits. The Cooper-pair box itself was first demonstrated by Nakamura, Pashkin and Tsai at NEC in 1999, the first solid-state qubit ever to show coherent oscillations.
Sources
- Koch et al., Phys. Rev. A (2007) — transmon proposal (https://journals.aps.org/pra/abstract/10.1103/PhysRevA.76.042319)
- Nakamura et al., Nature (1999) — first Cooper-pair box qubit (https://www.nature.com/articles/19718)
- IBM Quantum hardware roadmap (https://www.ibm.com/quantum/roadmap)