Rydberg Atom Array
Hundreds of single atoms held in checkerboards of laser light, then nudged into states the size of a cell.
Holds anywhere from a few dozen to over a thousand individual neutral atoms (rubidium, cesium, strontium, ytterbium) in a programmable two- or three-dimensional pattern of optical tweezers. Carefully timed laser pulses promote selected atoms into Rydberg states with principal quantum number n ≈ 50–100, where they interact strongly enough to entangle in under a microsecond. The same hardware can run quantum simulations of magnetism, perform digital quantum gates, and host the world's most accurate optical-lattice clocks.
A magneto-optical trap cools millions of atoms to microkelvin temperatures; a single tightly-focused laser tweezer (≈1 µm waist) then catches one atom at a time. Acousto-optic deflectors and spatial light modulators steer arrays of hundreds of tweezers into arbitrary patterns and rearrange them on the fly. To make atoms interact, two laser pulses (typically 420 nm + 1013 nm for rubidium) drive a two-photon transition into a Rydberg state. Two Rydberg atoms within ~10 µm experience a van-der-Waals shift large enough that promoting one suppresses the other (the 'Rydberg blockade'), which acts as a native two-qubit entangling gate. Read-out is destructive: a fluorescence image taken after a state-selective push reveals which atoms ended up in which qubit state.
In plain terms
Imagine catching individual atoms with invisible chopsticks made of laser light, lining them up in a perfect grid, then making each one balloon up to the size of a virus for just a flicker of time. While they are bloated, neighbours feel each other and shake hands; when they shrink back down, they remember the conversation.
Also called: neutral-atom array, Rydberg tweezer array.
Why it matters
- Same atom species used everywhere on Earth — no per-chip variability the way superconducting circuits have.
- Largest programmable quantum systems demonstrated to date: arrays of 1,180 (Atom Computing) and 256+ (QuEra) atoms.
- Dual-use: identical hardware runs world-record optical lattice clocks (10⁻¹⁹ fractional uncertainty).
- Rearrangeable connectivity — any atom can be moved next to any other, side-stepping the fixed-graph limitation of solid-state qubits.
Where it lives
- QuEra Aquila (256-qubit cloud machine)
- Atom Computing Phoenix (1,180-atom array, 2023)
- Pasqal Orion-class neutral-atom processors
- JILA & Boulder strontium optical lattice clocks
Key numbers
- Atom count (record): 1,180 (Atom Computing, 2023)
- Tweezer waist: ~1 µm (Single-atom optical traps)
- Rydberg n: 50–100 (Principal quantum number used)
- Blockade radius: ~10 µm (Range of strong dipole-dipole interaction)
- Two-qubit gate time: <1 µs
Surprising facts
- A single atom in a Rydberg state with n=100 has an electron orbiting roughly 1 µm from the nucleus — comparable to the size of a small bacterium, and 10,000 times wider than a normal atom.
- Atom Computing operated 1,180 single neutral atoms in a programmable array in 2023 — the largest single-step register of qubits ever assembled.
- Two Rydberg atoms 10 µm apart can shift each other's energy by gigahertz, enabling entangling gates in well under a microsecond.
- Strontium optical-lattice clocks built on the same hardware reach fractional uncertainties below 10⁻¹⁹ — they would gain or lose less than a second over the 13.8-billion-year age of the universe.
Open questions
- Anomalous Rydberg-state lifetime shortening: Rydberg atoms in dense arrays decay measurably faster than predicted by single-atom radiative-lifetime calculations. The discrepancy depends on array density and trap geometry in ways that no purely thermal black-body or single-atom model captures cleanly. Cooperative emission, near-resonant micro-electric-fields from neighbour-atom dipoles, and stray ions are all suspects, but no consensus mechanism has emerged.
- Dipole-blockade leakage at the few-percent level: Rydberg blockade is supposed to give a perfect 'no two excitations at once' rule, yet experiments routinely see a few percent of forbidden double excitations even in well-isolated pairs. The leakage scales unexpectedly with laser intensity and is suspected to involve doubly-excited 'pair states' that intersect the desired Rydberg energy at certain interatomic distances — a Förster-resonance pathway whose contribution is only partially modelled.
A short history
The Rydberg blockade as a route to a quantum logic gate was proposed by Lukin, Fleischhauer, Côté, Duan, Jaksch and Zoller in 2001 (PRL 87, 037901). The first demonstration of entanglement between two neutral atoms via blockade came in 2010 (Wisconsin and Institut d'Optique). The leap from a handful of atoms to thousands followed Bakr, Browaeys and Lukin's optical-tweezer rearrangement techniques (2016–2018). Atom Computing's 1,180-atom array (2023) and QuEra's Aquila cloud service (2022) established neutral atoms as a leading qubit platform alongside superconducting circuits.
Sources
- Lukin et al., PRL (2001) — Rydberg blockade gate proposal (https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.87.037901)
- Atom Computing — 1,180-atom array announcement (2023) (https://atom-computing.com/atom-computing-sets-new-record-with-1180-qubit-quantum-computer/)
- Bluvstein et al., Nature (2024) — logical processor with neutral atoms (https://www.nature.com/articles/s41586-023-06927-3)
- QuEra Aquila documentation (https://www.quera.com/aquila)