The key to this promise is the 21-centimeter line of neutral hydrogen. When the electron in a hydrogen atom flips its spin relative to the proton, the atom emits or absorbs a photon with a wavelength of 21 centimeters, corresponding to a frequency of about 1420 megahertz. Because the universe was filled with neutral hydrogen throughout the dark ages, this transition stamps an imprint on the radiation that reaches us today, redshifted by the cosmic expansion to low radio frequencies of roughly a few megahertz to tens of megahertz. Crucially, unlike the cosmic microwave background, which offers only a two-dimensional snapshot of the universe at the moment it became transparent about 380,000 years after the Big Bang, the 21-cm signal encodes information across three dimensions. Each observed frequency corresponds to a different distance, and therefore a different cosmic epoch, allowing astronomers in principle to reconstruct a tomographic movie of cosmic evolution spanning hundreds of millions of years that no other observable can match.

The dark ages are especially attractive because they are astrophysically pristine. Before the first stars ignited, the hydrogen signal was shaped almost entirely by fundamental physics: the density of matter, the temperature of the gas, the expansion rate of the universe, and the properties of any particles or fields beyond the standard model. There are no galaxies, no quasars, and no complicated feedback processes to muddy the interpretation. Any deviation of the observed signal from the predictions of the standard Lambda-CDM model would therefore point directly at new physics, whether in the dark matter sector, in the inflationary epoch that seeded all structure, or in some unexpected background of low-frequency radiation.

In their new analysis, the Calicut team focused on three representative non-standard scenarios. The first is an excess radio background, a hypothetical sea of additional low-frequency photons beyond the cosmic microwave background, parameterized in the study by an amplitude of 0.001. Such a background would raise the radio temperature seen by hydrogen atoms and deepen the absorption features in the 21-cm signal, a possibility that has attracted attention since the EDGES experiment reported an unexpectedly strong absorption profile centered at 78 megahertz in 2018. The second scenario involves co-SIMP dark matter, a strongly interacting form of dark matter whose freeze-out mechanism, proposed by Smirnov and Beacom in 2020, differs from the standard weakly interacting picture. With an interaction parameter of 1.5, co-SIMP particles would scatter with baryons and alter the thermal history of the gas during the dark ages. The third scenario is a modification of inflation: the Starobinsky model featuring a localized suppression of the primordial power spectrum at a wavenumber of about 7.5 inverse megaparsecs, which would leave a characteristic dip in the distribution of matter fluctuations on small scales.

To assess whether any of these effects could actually be measured, the researchers simulated observations with two complementary approaches. The first is the global, or sky-averaged, 21-cm signal, which measures the mean brightness of the hydrogen emission or absorption as a function of frequency. This requires only a single radio antenna with an exceptionally well-calibrated spectrometer. The second is the power spectrum of 21-cm fluctuations, which measures how the signal varies across the sky on different angular and frequency scales. This demands an interferometric array of many antennas, but it delivers far richer information, because the pattern of fluctuations encodes the three-dimensional structure of the early cosmos and is far less vulnerable to certain systematic errors that plague global measurements.

The team’s forecasts assume a single-antenna spectrometer placed on the lunar far side, integrating for 1000 hours. Under these conditions, they find that the standard Lambda-CDM global signal could be detected relative to a null hypothesis at a significance of 4.12 sigma. More strikingly, the co-SIMP dark matter scenario produces a global signal detectable at 4.67 sigma, and, crucially, distinguishable from standard cosmology at 2.76 sigma. The excess radio background model, by contrast, would be detectable relative to null at 3.86 sigma but could only be separated from the standard prediction at a marginal 0.29 sigma, meaning a global-signal experiment alone would struggle to tell it apart from conventional cosmology. The lesson is subtle: detecting a signal is not the same as identifying which physics produced it, and different exotic models leave fingerprints of very different strength in the sky-averaged spectrum.

The power spectrum measurements tell a more encouraging story for model discrimination. For a compact interferometric array covering 25 square kilometers on the lunar far side, observing for 10,000 hours, the team forecasts detection significances relative to null of 1.97 sigma for Lambda-CDM, 2.26 sigma for the excess radio background, 4.50 sigma for co-SIMP dark matter, and 1.61 sigma for the Starobinsky inflationary model. When it comes to distinguishing these models from standard cosmology, the spatial power spectrum achieves 3.32 sigma for the co-SIMP scenario, a substantial improvement over the global signal’s 2.76 sigma, while the excess radio background remains difficult to separate at 0.29 sigma and the Starobinsky suppression is the hardest of all to detect at just 0.13 sigma. These numbers make clear that the interferometric route, though technically far more demanding, is the one that truly unlocks the dark ages as a precision probe of fundamental physics, particularly for dark matter models that modify the thermal and clustering history of the early universe.

Why the Moon? From Earth, the dark ages signal at frequencies below about 30 megahertz is hopelessly corrupted by the ionosphere, which reflects and absorbs low-frequency radio waves, and by human-generated radio interference from broadcast transmitters, radar, and satellites. The lunar far side is the only location in the inner solar system permanently shielded from Earth’s radio noise, and during the two-week lunar night it offers an environment of extraordinary radio quiet. This is why concepts such as the Dark Ages Radio Explorer, or DARE, proposed an orbiting spacecraft, while more recent designs envision arrays deployed directly on the surface, including in-situ manufactured antenna concepts like Farview and the LuSEE-Night lunar surface electromagnetics experiment. The new forecasts give these mission planners concrete targets: a single-antenna spectrometer with 1000 hours of integration can already deliver a detection of the global signal, while a 25-square-kilometer interferometer with 10,000 hours of observation can begin to discriminate between competing models of dark matter and inflation.

The implications extend well beyond any single model. The co-SIMP result demonstrates that the dark ages 21-cm signal can probe strongly interacting dark matter in a regime inaccessible to laboratory detectors and collider experiments, complementing direct-detection efforts with an entirely independent cosmological test. The excess radio background analysis shows that even if a global absorption feature is found, pinning down its origin may require the spatial information that only interferometry provides. And the Starobinsky forecast illustrates that features in the primordial power spectrum, which would reveal the physics of inflation itself, are at the edge of detectability, motivating even larger arrays and longer integration times. Earlier theoretical work by Loeb and Zaldarriaga in 2004 and by Lewis and Challinor in 2007 established the foundations of dark ages tomography, and recent studies by Mondal and Barkana have argued that the 21-cm signal from the dark ages could ultimately deliver precision cosmology of unprecedented reach. The new study turns those ambitions into quantitative benchmarks.

For now, the dark ages remain unobserved; no instrument has yet detected the signal from this primordial era, and the EDGES claim of a cosmic dawn absorption feature remains unconfirmed. But the trajectory is clear. As space agencies develop lunar surface infrastructure, low-frequency radio experiments on the far side are moving from concept studies toward flight programs. If the forecasts of the Calicut team hold, humanity’s first three-dimensional map of the cosmic dark ages would not merely fill a gap in the timeline of the universe. It would test the nature of dark matter, scrutinize the seeds laid down by inflation, and search for radiation backgrounds that should not exist in standard cosmology, all by listening to the faintest radio hum that hydrogen atoms ever produced. The darkest epoch of cosmic history, it turns out, may be the brightest idea yet for probing the foundations of physics.

Subject of Research: Observational forecasts for using the dark ages 21-cm signal from neutral hydrogen to test non-standard cosmological models including excess radio backgrounds, co-SIMP dark matter, and modified inflation.

Article Title: Probing fundamental cosmology with the dark ages 21-cm signal

Article References: Moorkanat, D., Krishna, A., Hiten, H., Sadhaiv, R. V., Dadaso, S. T., & Mondal, R. (2026). Probing fundamental cosmology with the dark ages 21-cm signal. Astrophysics and Space Science, 371(9), Article 100. https://doi.org/10.1007/s10509-026-04632-x

Image Credits: AI Generated

DOI: 10.1007/s10509-026-04632-x

Keywords: cosmic dark ages, 21-cm cosmology, neutral hydrogen, dark matter, co-SIMP, excess radio background, Starobinsky inflation, lunar far side, power spectrum, global signal, Lambda-CDM, early universe