A University of Birmingham team uses ~24,000 ultracold rubidium atoms to construct an 'entropic time' from entropy exchange between a bright and dark sector, parameterizing a Schrödinger equation that matches experimental measurements — reviving relational time ideas rooted in…
Ninety years after physicist Nevill Mott proposed that time might arise from quantum correlations between parts of a system rather than exist as a fundamental backdrop, a University of Birmingham team has given that idea its most concrete experimental test yet. The group, led by Giovanni Barontini, built what the team calls a "toy universe": a Bose–Einstein condensate of roughly 24,000 rubidium-87 atoms laser-cooled to a fraction above absolute zero, partitioned by a thin optical barrier into a "bright" sector and a "dark" sector — the latter acting as an unobserved environment analogous to dark matter. The setup, described in a paper posted to arXiv in September 2025 and revised in March 2026 (Barontini, arXiv:2509.07745), is described as a controllable stand-in for the kind of quantum correlations Mott had in mind, and a way to put a nearly century-old thought experiment on a lab bench.
The mechanism is what separates this from a generic quantum-weirdness demo. The team constructs an entropic time from a coarse-grained entropy defined via the exchange of atoms between the bright and dark sectors. This internal time variable is then plugged into a Schrödinger equation as the time parameter; the predicted quantum states of the bright sector match what the experiment measures across repeated cycles of expansion and recollapse. Time, in this framework, is something the system generates internally through entropy flow — not a coordinate the experimenter supplies from outside.
That is a meaningful step beyond earlier tests. The lineage goes back to Mott, who in 1929 analyzed how cloud-chamber tracks arise from entanglement between a decaying particle and the surrounding atoms, with the global quantum state remaining pure even as a classical record emerges. The paper cites this as "an early prototype of this mechanism." Barontini's contribution, confirmed by the paper, is to push the idea into a tunable many-body platform — ultracold atoms rather than photons — where the sector geometry, coupling strength, and internal clock observable can all be dialed. The paper's subjects are General Relativity and Quantum Cosmology (gr-qc), Quantum Gases (cond-mat.quant-gas), and Quantum Physics (quant-ph), and the work is explicitly framed within the Wheeler–DeWitt problem of finding a time variable in a time-independent Hamiltonian system.
The scope is narrower than the headlines suggest. This is a tabletop analogue, not a claim that time is an illusion in our universe. The bright/dark sector toy model is a constructed system with a designed coupling; the paper itself states the goal is to establish "a controlled experimental setting in which relational-time constructions can be quantitatively tested." It does not, by itself, overturn thermodynamics, relativity, or the everyday experience of time's arrow. New Scientist's framing of "time could be a quantum illusion" is closer to a research-direction provocation than to a settled finding, and the paper's more careful language — "emergent internal time from entropy exchange" — better reflects what the experiment actually demonstrates.
One biographical detail worth noting: Barontini has discussed the project's origin in interviews, including describing watching his young son build a "small universe" with toys as a seed for thinking about artificial model universes in the lab. That anecdote is part of Barontini's personal scientific origin story and is not a finding of the paper.
What to watch next: the published version of the paper, expected to clarify the specific observable used for the internal clock read-out across different barrier heights, the detailed entropy operationalization, and whether the platform can be tuned into regimes where emergent time behaves differently. Independent quantum-foundations and AMO physicists will be the right audience for a sanity check on whether this counts as a clean test of Mott's original idea or a cousin of it. Either way, the cold-atom toy universe gives the Mott question — whether time is relational rather than fundamental — a place to be asked, and partially answered, in a lab.