The cosmic microwave background is the afterglow of the Big Bang, which we see at millimeter wavelengths. But it also has a long-wavelength tail extending into the radio range. At wavelengths longer than 21 centimeters, this light begins to be absorbed by neutral hydrogen scattered across distant space. Hydrogen acts like a frosted glass: some photons get stuck in it, making the sky map at these frequencies slightly smoother than expected.
The effect depends on two things: how hot the gas is (its spin temperature) and how many neutral atoms are there, as opposed to ionized ones. The colder the hydrogen, the stronger the absorption. The absorption peak occurs at frequencies around 100 megahertz: this corresponds to the cosmic "dark ages," before the first stars ignited. The fraction of absorbed light can reach a few percent.
Directly detecting such a faint radio signal is a daunting challenge. But the authors propose a workaround: compare radio maps with the already precise millimeter-wave maps of the cosmic microwave background. If the radio map shows a slight smoothing that the millimeter map does not, then hydrogen is indeed absorbing. This cross-analysis could be easier than catching hydrogen's own glow.
This approach offers a new way to measure the ratio of neutral hydrogen fraction to its spin temperature at different epochs. This is important for understanding how gas heated up in the early universe and when the first stars ignited. In essence, we get a thermometer and an atom counter for times when the universe was dark and cold.
as written by the authors arXiv:2605.22934
At wavelengths longer than 21 cm, photons from the long-wavelength tail of the cosmic microwave background (CMB) have a non-zero probability of being absorbed by distant neutral hydrogen. This provides an additional suppression of the observed CMB clustering in addition to the usual Thomson scattering. The optical depth as a function of frequency is sensitive to the 21 cm spin temperature $$T_s$$ of the gas as a function of cosmic time, with the excess optical depth peaking at a level of a few percent around 100 MHz. The details depend on the specifics of the heating of cosmic gas and the evolution of the neutral fraction $$x_{HI}$$. It is likely difficult to detect the CMB at these long radio wavelengths, but the cause is aided by the ability to cross-correlate with the already well-characterized fluctuations at cm/mm frequencies. We find that detecting CMB fluctuations at radio wavelengths corresponding to the 21 cm ``dark ages'' in cross-correlation with mm-wave maps may be easier than detecting the intrinsic 21 cm fluctuations. Measurement of the amplitude of CMB fluctuations as a function of radio wavelength provides a path for a new type of direct measurement of the combination $$x_{HI}/T_s$$ as a function of redshift.
The 21 cm wavelength is hydrogen's natural "radio beacon," arising from the flip of the electron's spin. Astronomers use it to map galaxies.
In Carl Sagan's novel "Contact," a signal at the hydrogen frequency becomes the key to deciphering a message from an extraterrestrial civilization.