Dark matter is an invisible window bending the light of galaxies. It was discovered by Vera Rubin, who noticed that stars were moving too fast. Perhaps this window is made of axions — particles billions of times lighter than an electron, invented by Roberto D. Peccei and Helen Quinn to fix the Standard Model. Axions are elusive: they can travel through a light-year of lead without a single collision. If dark matter is glass, axions are the scratches on it. Scientists searched for these scratches in the cosmic microwave background — the light of the Big Bang. Massive objects curve it, creating gravitational lenses. Three telescopes measured the distortions to figure out the fraction of axions. The result: for one axion mass, their share is no more than 1.5% of all dark matter. That’s like looking for a handful of sand in an Olympic swimming pool. For another type, a faint hint flickered — like a glint on glass — but it’s probably just a fluke. The window of the universe keeps its secrets.
as written by the authors arXiv:2606.06410
Ultralight axions (ULAs) are well-motivated dark matter particle candidates that arise in many extensions of the Standard Model of particle physics. ULAs with mass $$m_\mathrm{a} \lesssim 10^{-27}$$ eV have been strongly constrained by cosmic microwave background (CMB) observations in temperature and polarization. We fit recent measurements of gravitational lensing of the CMB from \textit{Planck}, the Atacama Cosmology Telescope (ACT) and the South Pole Telescope (SPT-3G) using a state-of-the-art simulation-calibrated nonlinear clustering model for ULAs. We derive the strongest constraints on ULAs in the mass range $$10^{-26}\;\mathrm{eV}\leq m_\mathrm{a}\leq 10^{-24.5}\;\mathrm{eV}$$. ULAs of this mass have been shown to alleviate tensions between inferences of the matter clustering if they compose a few percent of the total dark matter content of the Universe. We conclude that ULAs with a mass of $$10^{-26}$$ eV make up less than 1.5\% of the dark matter and $$10^{-25}$$ eV axions make less than 9\% (both at 95\% confidence level). We identify a slight preference for non-zero axion density at $$10^{-24.5}$$ eV at $$2.1\sigma$$. We find that the preference for ULAs is largely driven by a few data points and that further investigation of non-linear ULA physics is needed to confirm or rule out this signal.
The name 'axion' came from a joke: physicists borrowed the name of a laundry detergent because the new particle 'cleaned up' a troubling mathematical paradox in the theory.