Researchers tested axion models by searching for extra cooling in white dwarfs from archival Hubble images of 47 Tucanae and found no signal. By adding axion-emission physics to stellar-evolution simulations and comparing predicted cooling sequences with observations, they constrained the electron-to-axion conversion rate to be no more than roughly one in a trillion. The result narrows the viable models for axion dark matter, especially those with direct electron coupling, while leaving other axion scenarios still possible.
What Ancient White Dwarfs Reveal About Axions As Dark-Matter Candidates
Could stellar evolution reveal more about dark matter?. | Credit: Nazarii Neshcherenskyi/Getty Images
New analysis of Hubble archival images uses cooling white dwarfs to probe invisible particles called axions.
Axions are a decades-old theoretical particle initially proposed to solve a puzzle in the strong nuclear force and later revived as a possible component of dark matter. Although collider experiments have not found them, certain axion models predict subtle astrophysical effects that can be tested using precise observations of stars.
The recent study (published as a preprint on arXiv in November 2025) looks for one such effect: extra cooling of white dwarfs caused by axions produced by energetic electrons. White dwarfs are the dense, slowly cooling remnants of stars: they can contain roughly the Sun’s mass in a volume smaller than Earth and are supported against collapse by electron degeneracy pressure. Deep inside these objects, electrons can attain relativistic speeds, creating favorable conditions for some electron–axion production mechanisms.
How Axions Would Reveal Themselves
In models where electrons can convert into axions, each conversion carries energy away in the form of escaping axions. Because white dwarfs have no internal energy source (like fusion), any additional loss channel accelerates their cooling and dimming. By comparing observed white-dwarf temperatures and luminosities with predictions from stellar-evolution models that include axion emission, researchers can test whether axion cooling is present.
Data and Method
The team used archival imaging from the Hubble Space Telescope of the globular cluster 47 Tucanae. Globular clusters are ideal laboratories for this test because their stars formed at nearly the same time, producing a coeval population of white dwarfs whose cooling sequence can be statistically compared with models. The researchers implemented axion-emission physics into an advanced stellar-evolution code to predict how white-dwarf temperatures and brightness evolve with age both with and without axion losses.
Result: No detectable axion-induced extra cooling was found in 47 Tucanae’s white-dwarf population.
From this null result the team derived a new astrophysical constraint on the efficiency of electron-to-axion conversion: an electron in a white dwarf cannot produce an axion more often than roughly one time in a trillion attempts. This bound does not exclude axions in all theoretical variants, but it strongly disfavors models with a direct, efficient coupling between electrons and axions.
These findings tighten the allowed parameter space for axions and complement laboratory searches. In short: axions remain viable in some models of dark matter, but direct coupling to electrons appears unlikely based on this Hubble-based test. Future searches will need still more inventive observational and experimental approaches.
Key details: arXiv preprint (November 2025); target: white dwarfs in 47 Tucanae; method: stellar-evolution simulations including axion losses; outcome: no evidence of axion cooling; constraint: electron-to-axion conversion <~1 in 10^12.
