Pasqal's 256-qubit processor finds a quantum-magnet slowdown theory did not predict
A Pasqal-led preprint reports an unexpected relaxation slowdown in a 2D quantum magnet on 256 neutral atoms, where three classical tensor-network methods lose control.
Pasqal says its 256-qubit neutral-atom processor has revealed behavior in a two-dimensional quantum magnet that current theory did not anticipate. In a preprint released with the Flatiron Institute's Center for Computational Quantum Physics, the University of Augsburg, Forschungszentrum Jülich, the University of Regensburg and the Institut d'Optique–CNRS, the team reports that under certain conditions the magnet takes far longer to settle after a sudden disturbance than expected, in a regime where leading classical simulation methods stop agreeing with one another.
What was announced
Pasqal announced the result on October 27 and posted the paper, Exploring the Relaxation Landscape of a 2D Quantum Magnet on a 256-Qubit Processor, on arXiv. The company frames it as a shift from using quantum processors to reproduce known physics toward using them to uncover behavior researchers did not foresee. It builds on earlier work in which the same machine reproduced the behavior of a real magnetic crystal.
The technical details
The experiment arranged 256 rubidium-87 atoms in a 16×16 square array and used the ground and Rydberg states of each atom as a qubit. The system simulated the two-dimensional transverse-field Ising model, a reference model of magnetism. Starting from a fully polarized state, the researchers abruptly changed the Hamiltonian, a "quench", and watched how correlations relaxed for different ratios of transverse field to interaction strength.
They found three regimes. Near the dynamical phase transition, around a ratio of 2.5, the system thermalized quickly. At a strong field of 6.0 it settled on a long-lived prethermal plateau described by an effective XY-type model. At an intermediate value of 3.5, relaxation slowed markedly, with correlations staying far from their thermal values well beyond the usual timescale. The authors write that they do not yet have a clear explanation for this slowdown.
To check the data, the team compared it with three tensor-network methods: two-dimensional tensor networks, matrix product states and tree tensor networks. All three matched the processor at early times, then drifted apart from the measurements and from each other at later times. The processor's results stayed consistent as the system size grew. Noise modeling, measurement-error correction and exact simulations on 5×5 lattices supported the conclusion that the slowdown reflects many-body physics rather than a hardware artifact.
Why it matters
How interacting quantum systems relax far from equilibrium in two dimensions is a hard open problem, and it is the kind of question where analog quantum simulators are expected to be useful. Pasqal and its partners say explicitly that the work does not demonstrate a computational quantum advantage. Instead, they present it as evidence that such processors can serve as complementary scientific tools where theory and classical numerics struggle.
"We are not simply reproducing what we already know, but uncovering behavior we did not expect," said Loïc Henriet, Pasqal's CTO.
What to watch
The paper is a preprint, so peer review and attempts by other groups, both experimental and classical, to explain the slowdown are the next steps. The authors present Rydberg atom arrays as a key platform for the open questions they identify, and say ongoing improvements in coherence times offer a path toward systematic finite-size and finite-time scaling analyses.
Sources
Facts on this page come from the sources below, retrieved by DesignQuantum.
- Pasqal's Neutral-Atom Processor Uncovers Quantum Magnet Behavior That Theory Did Not Predict (Pasqal press release) pasqal.com
- Exploring the Relaxation Landscape of a 2D Quantum Magnet on a 256-Qubit Processor (arXiv preprint) arxiv.org
- Pasqal 256-Qubit Neutral-Atom Processor Observes Unpredicted Slowdown in 2D Quantum Magnet Dynamics (Quantum Computing Report) quantumcomputingreport.com