Free the news
ThinkPol
Science

Dark energy may come from the force that holds atoms together, UBC-led study proposes

By ThinkPol · · 4 min read

A small telescope silhouetted against a star-filled night sky
Dreamstime

Dark energy, the unexplained something that is making the universe expand faster and faster, might not need any new physics at all, according to physicists at the University of British Columbia and the University of Sheffield. They propose that it comes from the same theory that describes the force holding atomic nuclei together.

Their paper, published Thursday in the Astrophysical Journal, tests the idea against the latest cosmological data and finds it fits at least as well as the standard model of cosmology, and in some comparisons slightly better.

The problem

Since 1998, astronomers have known that the expansion of the universe is speeding up. Whatever drives it, called dark energy, makes up about 70 per cent of the universe's total energy, but no one knows what it is.

The simplest explanation is a "cosmological constant," an energy of empty space that never changes. That is the standard model, known as ΛCDM, and it fits most data very well. But it has two awkward features. Particle physics predicts a vacuum energy enormously larger than what is observed. And it is a coincidence that dark energy only came to dominate the universe recently, in cosmic terms, just as we arrived to measure it.

The question became more pressing last year. Results from the Dark Energy Spectroscopic Instrument (DESI), which maps millions of galaxies, hinted that dark energy may be weakening over time rather than staying constant. If that holds up, a fixed cosmological constant cannot be the whole story.

The proposal

The idea comes from UBC theorist Ariel Zhitnitsky, who has developed it over more than a decade. It draws on quantum chromodynamics, or QCD, the established theory of the strong nuclear force that binds quarks inside protons and neutrons.

In QCD, empty space is not truly empty: it has a complex structure that includes so-called topological features. Zhitnitsky argues that in an expanding universe, this structure leaves behind a small amount of energy that depends on the expansion rate itself. A rough calculation using the known strength of the strong force gives a dark-energy density of the right order of magnitude, without introducing any new particle or field.

"Our theoretical framework suggests no new particles may be required at all," Zhitnitsky said in a UBC release.

The model also offers an answer to the "why now?" question. In the early universe, the authors argue, the effect was suppressed, because the universe was changing too quickly. It switches on only as expansion slows enough, which happens late in cosmic history.

Testing it against the data

The new paper, led by Sheffield PhD researcher Dong Ha Lee, is the first full test of the idea against observations. The team compared it with the standard model and with a popular general description of evolving dark energy, using cosmic microwave background measurements from the Planck satellite and ground-based telescopes in Chile and at the South Pole, DESI's latest galaxy maps, and two catalogues of exploding stars used to measure distance.

The model reproduces the weakening dark energy that DESI hints at. With only early-universe data, or only late-universe data, it is statistically indistinguishable from the standard model. With everything combined, it is mildly favoured. In the most demanding comparison, the Bayesian odds come out at roughly seven to one in its favour, a level the authors describe as moderate evidence but which many cosmologists would call weak. It also performed more consistently than the general evolving-dark-energy description, whose support depends on which supernova catalogue is used.

Reasons for caution

UBC's release says the dark-energy scale emerges "without the need for any additional fitting parameters." That is true of the overall size of the effect. But to fit the data, the paper adds two parameters that describe when and how quickly the effect switches on, the same number of extra parameters as the generic model it is compared with. The authors call the switch a "phenomenological" description and say a more fundamental version is future work.

The headline calculation is also an order-of-magnitude estimate, and parts of the theory rest on calculations that the authors acknowledge are technically very difficult to verify, including in simulations of QCD.

Most important, the evidence that dark energy is evolving at all is not settled. The DESI hints depend partly on supernova data and analysis choices, and more careful Bayesian treatments have generally found weaker support than the original headlines.

What would settle it

The test will come from new surveys of the expansion history: further DESI data releases, Europe's Euclid space telescope, NASA's Nancy Grace Roman Space Telescope and the Vera C. Rubin Observatory. If the hints of evolving dark energy fade, the case for this model weakens with them.

The paper also points to a possible laboratory test. The same mechanism predicts a tiny extra contribution to the Casimir effect, the attraction between closely spaced metal plates caused by the quantum vacuum. That "topological Casimir effect" has not been measured, but the authors say detecting it would directly support their account of dark energy.

The Canadian researchers are funded in part by the Natural Sciences and Engineering Research Council of Canada.

Source: Lee, D. H., van de Bruck, C., Di Valentino, E., Van Waerbeke, L. and Zhitnitsky, A., "Evolving Dark Energy Is Vacuum Energy After All," The Astrophysical Journal, Nov. 20, 2026 (doi:10.3847/1538-4357/aeb9cc); Van Waerbeke and Zhitnitsky, "DESI results and dark energy from QCD topological sector," Physical Review D; University of British Columbia news release, Nov. 20, 2026.

Front PageScience#Dark energy#Cosmology#UBC#physics#DESI