Ultracold atoms recreate an effect too faint to observe directly

Physicists in China have used a cloud of atoms cooled to within a hair's breadth of absolute zero to reproduce, in miniature, a process that has never been observed in its original form: the spontaneous creation of particle-antiparticle pairs out of empty space. The effect, predicted by the physicist Julian Schwinger in 1951, requires an electric field so far beyond anything achievable in a laboratory that seeing it directly has been impossible for three-quarters of a century. The new work, published this week in Physical Review Letters, instead builds a scaled-down stand-in for the effect using laser-trapped rubidium atoms, and watches pairs of excitations appear in a way that mirrors the real process almost exactly.

The team, based primarily at the University of Science and Technology of China (USTC) in Hefei, included the theoretical physicist Frank Wilczek, who shared the 2004 Nobel Prize in Physics, along with collaborators in Italy, Germany and the United States. The experiment was first described in a 2024 preprint and has now cleared peer review; it was reported Friday by Phys.org, which spoke with the paper's senior authors about the result.

Building a vacuum out of atoms

The researchers started with a Bose-Einstein condensate of rubidium-87 atoms and loaded it into a one-dimensional optical lattice, a standing-wave pattern of laser light that traps atoms at regularly spaced sites, the way an egg carton holds eggs. By filling every other site with a pair of atoms and leaving the rest empty, they built a repeating pattern that stands in for what physicists call a "bare vacuum" in a simplified one-dimensional version of quantum electrodynamics. Atoms were then allowed to hop between neighboring sites, a process engineered to mimic the creation of a particle-antiparticle pair, while a quantum gas microscope tracked the position of individual atoms well enough to resolve single lattice sites.

In the experiment's simplest setting, with the simulated particle mass tuned to zero, the atoms' hopping rate (what the authors call the matter-gauge coupling) was about 75 hertz, and the resulting pair-production signal oscillated at roughly 80 hertz, with a weaker second peak near twice that frequency. The researchers tracked the system through about six full oscillation cycles, with the fraction of atoms converted into "pairs" reaching close to 70 percent at its peak. Pair production fell off sharply once the simulated mass grew large relative to that coupling, crossing a threshold the team calculated at roughly two-thirds of the coupling strength — the lattice analog of the enormous field needed to overcome a particle's rest-mass energy in the real Schwinger effect. Agreement between the measured atom dynamics and theoretical predictions held for roughly the first ten milliseconds of each run, the window in which the lattice simulator stayed well isolated from outside noise.

A process too faint to see directly

In ordinary space, the vacuum is not perfectly empty; quantum theory holds that it constantly produces pairs of "virtual" particles that pop in and out of existence too quickly to detect. A sufficiently strong electric field can, in principle, pull such a pair apart before it has time to re-annihilate, turning it into a real, detectable particle and antiparticle. The field required is enormous — on the order of 10^16 volts per centimeter, far beyond what any laser or accelerator on Earth can sustain — which is why the effect, despite being a standard prediction of quantum field theory, has never been seen in the form Schwinger originally described. Physicists have come to describe this kind of process more broadly as a form of vacuum decay: a seemingly stable, lowest-energy-looking state is in fact only locally stable, or "false," and can tunnel into a lower-energy configuration by producing particles, in close analogy to how a radioactive nucleus decays.

That broader framing is why the result has drawn interest beyond atomic physics. The same mathematics that describes pair production under a strong field also appears in cosmological models of how the universe's fundamental fields — including, in some theories, the Higgs field that gives elementary particles their mass — might have evolved, or could still evolve, out of a metastable state. A cold-atom platform that lets researchers dial the effective mass and field strength independently, and watch the resulting dynamics atom by atom, gives theorists a controllable analog system for questions that are otherwise confined to pen-and-paper estimates or supercomputer simulations that struggle with the underlying quantum many-body mathematics.

"This is an example of 'false vacuum decay,' and in its original form it has never been observed directly," said Zhen-Sheng Yuan, a physicist at USTC and one of the paper's senior authors, adding that the atom-based approach brings "otherwise hard-to-observe quantum-field processes into a controlled laboratory."

Who the result matters to

The immediate audience for the work is narrow: researchers who build and use cold-atom quantum simulators, and theorists working on lattice gauge theory, the mathematical framework used to describe how particles interact through forces like electromagnetism and the strong nuclear force. For that community, the appeal of the USTC platform is that it tracks individual atoms with single-site resolution, giving direct, atom-by-atom access to dynamics that are usually inferred indirectly or modeled numerically. Yuan's research group, housed at USTC's Hefei National Laboratory, has spent several years developing the cooling and imaging techniques that make this kind of experiment possible, including methods for entangling large numbers of trapped atom pairs.

There is no near-term application for the public, and the authors do not claim to have observed the Schwinger effect itself, only a faithful analog of its underlying physics in a different, more controllable medium. The result does not change any prediction of the Standard Model of particle physics, nor does it bear on whether the actual vacuum of our universe is stable. Its significance is as a proof of concept: a demonstration that a table-top apparatus of laser-cooled atoms can be tuned to explore a regime of quantum field theory that remains inaccessible to particle accelerators and largely intractable for classical computers.

What comes next

The team, which also includes researchers affiliated with Caltech, the Max Planck Institute of Quantum Optics, and Trieste's International Centre for Theoretical Physics, says the same lattice platform could be extended to two-dimensional gauge theories, to simulations involving different types of fundamental forces, and to related phenomena such as the breaking of a "string" of field lines between two charges or the scattering of composite particles analogous to mesons. The group's original preprint, posted to arXiv in November 2024, outlines several of those extensions as future work; the Physical Review Letters version adds the peer-reviewed data and analysis behind the published figures.

Jian-Wei Pan, the USTC physicist and Chinese Academy of Sciences academician who oversees the broader quantum research program the Yuan group sits within, has for two decades pushed Chinese labs toward large-scale quantum technology, from satellite-based quantum communication to programmable quantum processors. The new result is a smaller-scale undertaking by comparison, but it fits the same pattern: using precisely controlled quantum systems to answer questions that are otherwise out of reach, whether in cryptography, computing, or, in this case, a 75-year-old prediction about the stability of empty space.