Source-verifiedLevel 3

Rabi Oscillation

Rabi oscillation is the sinusoidal cycling of a qubit's population between |0⟩ and |1⟩ under a resonant drive: P₁(t) = sin²(Ωt/2). Its first peak defines the π-pulse — the calibrated NOT (X) gate.

What it means

Drive a qubit with a microwave pulse exactly on resonance with its 0→1 transition and the population does not simply drift to |1⟩ — it oscillates.The probability of measuring |1⟩ follows P₁(t) = sin²(Ωt/2), where the Rabi frequency Ω is proportional to the drive amplitude: push harder and the oscillation speeds up.This makes Rabi oscillation the workhorse of gate calibration.Prepare the qubit in |0⟩, sweep the pulse amplitude or duration, plot the measured P(1), and a sinusoid appears; the first maximum is the π-pulse — the amplitude-duration pair that flips |0⟩ fully to |1⟩, i.e.a perfectly calibrated NOT (X) gate.That pair is stored as calibration data, alongside the transition frequency found by spectroscopy, and the pulse phase then selects the rotation axis (0° gives X, 90° gives Y).Because fabrication makes every qubit slightly different and its properties drift with time and temperature, the π-pulse must be periodically re-found by repeating the Rabi experiment — the X gate is not a fixed piece of hardware but a continuously maintained calibration.

Everyday analogy

Pushing a swing at exactly the right rhythm: push in time with the swing's own frequency and each push adds up — the swing goes higher and higher, then comes back down, in a smooth cycle. Drive a qubit on resonance and the probability of finding it in |1⟩ oscillates sinusoidally the same way as you vary the push strength or duration. Stop exactly at the first peak and you have flipped 0 to 1 completely — that first peak is the π-pulse, a perfect NOT (X) gate.
Rabi is a person's name: the American physicist Isidor Rabi won the Nobel Prize for his magnetic-resonance work, in which two-level systems oscillate between states under a resonant drive. And 'pulse' comes from Latin pulsus, 'a beating, the heartbeat' — a fittingly biological word for the short rhythmic taps of microwave energy that make a qubit's state swing.

Common misconceptions

  • A Rabi π-pulse is NOT a fixed hardware property — the X gate is just a well-calibrated pulse. Its correct amplitude and duration drift with time and temperature, and must be re-found by repeating the Rabi experiment.
  • One qubit's π-pulse does NOT work on its neighbor — fabrication makes every qubit a snowflake, with slightly different frequency, anharmonicity and coupling, so Q1's NOT pulse is not Q2's NOT pulse.

Key takeaways

  • On resonance, P₁(t) = sin²(Ωt/2): the excited-state probability oscillates sinusoidally, and the Rabi frequency Ω is proportional to the drive amplitude — push harder, oscillate faster.
  • The first maximum of the Rabi oscillation is the π-pulse: the amplitude-duration pair that flips |0⟩ fully to |1⟩ — a calibrated NOT (X) gate, stored as calibration data.
  • Rabi sits inside a sequential calibration stack: spectroscopy finds the qubit frequency, Rabi finds the π-pulse, Ramsey refines phase/frequency and measures T2 — and the stack is re-run because everything drifts.
  • The pulse phase chooses the rotation axis on the Bloch sphere: 0° gives an X rotation, 90° gives Y.

Check your understanding

In a Rabi calibration experiment you sweep the drive amplitude or duration and plot P(1). What does the first peak of the resulting sinusoid give you?

  1. A.The qubit's resonance frequency
  2. B.The T1 relaxation time
  3. C.The π-pulse parameters — a calibrated NOT (X) gate
  4. D.The readout fidelity
Show the answer

Answer: C. The π-pulse parameters — a calibrated NOT (X) gate

Why: The first maximum of P(1) marks the amplitude-duration pair that fully flips |0⟩ to |1⟩. That pair defines the π-pulse, i.e. the calibrated X gate. The frequency comes from spectroscopy, and T1 from a separate decay experiment.

For a resonant drive, P₁(t) = sin²(Ωt/2). If you double the drive amplitude, what happens?

  1. A.Nothing — Ω is a fixed property of the qubit
  2. B.The oscillation gets slower
  3. C.Ω doubles, so the population oscillates twice as fast and the π-pulse gets shorter
  4. D.The qubit immediately leaves the |0⟩,|1⟩ subspace
Show the answer

Answer: C. Ω doubles, so the population oscillates twice as fast and the π-pulse gets shorter

Why: The Rabi frequency is proportional to the drive amplitude, so doubling the amplitude doubles Ω — the same rotation completes in half the time. (In practice, very strong short pulses broaden the spectrum and raise the leakage risk.)

Builds on

QubitBloch Sphere

Primary source: I. I. Rabi, Space Quantization in a Gyrating Magnetic Field, Phys. Rev. 51, 652 (1937), doi:10.1103/PhysRev.51.652

Rabi oscillation physics per Rabi 1937; transmon calibration protocol per Krantz 2019 SIV.

Learn it hands-on

This concept is part of a 46-level curriculum with an interactive simulator and Lumen, a tutor whose answers are verified before you see them. Levels 1–5 are free.

Start freeSee the curriculum