A new paper in Physical Review Letters suggests that dark photons — one of the leading candidates for dark matter — never heated the early universe the way physicists assumed, reopening a vast region of parameter space for experimental searches.
Dark photons are hypothetical particles that would interact with ordinary matter only through a faint mixing with regular photons. Physicists long assumed that in the hot plasma of charged particles filling the early cosmos, dark photons would convert into ordinary light, heating the plasma further and leaving detectable traces. Cosmological measurements were then used to rule out a broad range of dark photon masses and couplings.
But new computer simulations by Perimeter Institute researchers Junwu Huang and Mohamad Shalaby, with Anson Hook of the University of Maryland, show that the conversion process shuts itself off before significant heating can occur.
"The treatment for the last 15 years is a linear treatment. If you use that approximation, you can compute the amount of energy transfer, and it's very large," said Huang. "And I realized it's not possible."
Shalaby, a plasma-physics specialist, ran simulations showing that as soon as dark photon energy begins flowing into the plasma, the system becomes violently nonlinear. "As you are converting energy into the standard model plasma, the plasma actually goes crazy," Huang explained. "These nonlinearities basically shut off the energy conversion after a tiny amount of energy is converted."
According to the new analysis, the conventional constraint is invalid across roughly ten orders of magnitude in mass — from about 10⁻¹⁵ to 10⁻⁶ electron volts, corresponding to frequencies from roughly kilohertz to gigahertz.
"These exclusions were saying the strength of dark matter had to be 10⁸ weaker than it actually can be," said Hook. "This paper opens up a lot of new possibilities to look for dark matter."
The result may reach beyond dark photons: the same linear approximations underpin other astrophysical probes, such as searches using neutron star or white dwarf magnetospheres. "This is a test case in cosmology. A lot of astrophysical systems have also been used to look for similar effects, and we need to rethink all of them," Huang said.
"By calculating the early universe plasma correctly, experiments will probe new parameter spaces and potentially actually see something," Shalaby said.
