Imagine venturing into the universe's most mysterious chapter—a time when the cosmos transformed from a simple, uniform haze into the vibrant, complex web of stars and galaxies we know today. This isn't just any story; it's the thrilling quest for the earliest whispers from hydrogen atoms, a signal that could redefine our understanding of cosmic origins. But here's where it gets controversial: what if this discovery challenges everything we think we know about the universe's birth?
As a cosmologist at the University of Manchester's Jodrell Bank Centre for Astrophysics, Phil Bull captures the essence of this pursuit. 'We're standing on the brink of one of astronomy's last great unexplored territories,' he says. 'This era is crucial, yet our grasp on it remains frustratingly incomplete.'
Bull speaks of that pivotal stretch in the universe's timeline—from roughly 380,000 years post-Big Bang to about a billion years later—when the cosmos evolved from homogeneity to intricate diversity. To uncover the secrets of this phase, scientists worldwide, spanning from the Australian outback to the icy Arctic, are competing to capture a faint but pivotal radio signal from the universe's first hydrogen atoms. Such a find might validate our current theories of cosmic evolution or, intriguingly, overturn them.
Hydrogen reigns as the universe's most plentiful element. When these neutral hydrogen atoms shift their energy states, they release or capture photons, resulting in a radio signal at a precise 21-centimeter wavelength. This '21cm line,' as it's known, can be triggered by radiation or other interactions. By hunting for these ancient photons from primordial hydrogen, researchers aim to reconstruct the universe's formative events.
Yet, despite an ever-growing fleet of teams joining the fray annually, no conclusive detection has emerged. So, who will claim victory in this race, and how exactly are they conducting their searches?
Let's rewind to approximately 380,000 years after the Big Bang, when the universe had cooled below 3000 Kelvin and expanded sufficiently. At this point, neutral atoms, including hydrogen, could coalesce for the first time. Without free-roaming electrons jamming the scene, regular matter decoupled from light, letting it stream freely through space. This relic radiation, saturating our skies, is the cosmic microwave background (CMB).
But what transpired next? For hundreds of millions of years, our knowledge hits a wall. The oldest galaxy we've spotted, MoM-z14—dating back about 280 million years post-Big Bang—was unveiled in April 2025 via the James Webb Space Telescope. Thus, a gap of nearly 280 million years persists in our early universe observations. 'This represents one of the final uncharted voids,' notes Anastasia Fialkov, an astrophysicist at the University of Cambridge's Institute of Astronomy.
This 'void' acts as a crucial link between the primordial simplicity and the modern universe's elaborate architecture. During this span, the cosmos transitioned from a dense fog of neutral hydrogen to a realm teeming with stars, black holes, and all manner of structures. It encompasses the cosmic dark ages, the cosmic dawn, and the reionization era—arguably the most dynamic chapter in cosmic history.
In the cosmic dark ages, following the CMB's pervasive glow, the only 'ordinary' matter consisted of neutral hydrogen (accounting for 75% of mass) and neutral helium (25%), with no stars to illuminate the darkness. Gravity likely amplified tiny density variations, causing this ancient gas to cluster and ignite the first stars and galaxies—a phase dubbed the cosmic dawn. Subsequently, during reionization, ultraviolet and X-ray radiation from these nascent celestial bodies ionized the hydrogen, converting the neutral gas into a plasma of charged particles.
The 21cm signal under scrutiny was generated as hydrogen spectral transitions were spurred by gas collisions in the dark ages, later by the initial starlight during dawn. To gauge its strength, scientists compare it against the CMB, which provides a constant backdrop of 21cm photons.
When hydrogen gas was cooler than the CMB, fewer collisions meant atoms absorbed more 21cm photons than they emitted, creating an absorption dip against the CMB's glow. Conversely, hotter gas led to more emissions, producing a bright emission peak. These shifts hinge on gas density, temperature, and the radiation from early sources, essentially encoding the universe's transformation in the signal.
One detection strategy involves measuring the average, or 'global,' signal across the entire sky, tracking its evolution from absorption to emission relative to the CMB. Typically, 21cm radio waves oscillate at around 1420 MHz, but this ancient signal, stretched by cosmic expansion, now spans 1 to 200 MHz frequencies—with lower frequencies tied to earlier times—and stretches to meter-long wavelengths.
Crucially, the signal's profile over time could corroborate the lambda-cold dark matter (ΛCDM) model, our prevailing cosmic theory, or challenge it fundamentally. Astronomers have poured careers into this, but obstacles abound.
The signal is extraordinarily weak, with brightness temperature deviations from the CMB's 2.7 Kelvin baseline barely reaching 0.1 Kelvin.
[Figure 1: A simulation of the sky-averaged (global) signal over time (horizontal axis) and space (vertical axis). B: A model of the global 21cm line highlighting key cosmic events. Experiments like the Radio Experiment for the Analysis of Cosmic Hydrogen (REACH) target the 50–170 MHz band (shown in blue).]
Each global 21cm search zeroes in on specific frequency bands. For instance, REACH probes 50–170 MHz.
The signal lacks a single origin; like the CMB, it's ubiquitous. 'If it stood alone in the sky, we'd have pinpointed it ages ago,' remarks Eloy de Lera Acedo, director of the University of Cambridge's Cavendish Radio Astronomy and Cosmology group. But our universe brims with interference, chiefly from our Milky Way. They're pinpointing a mere 0.1 Kelvin change amid a million-times-brighter backdrop.
Earth's atmosphere adds distortion before the signal even arrives. 'This demands impeccable calibration and modeling,' explains Rigel Cappallo, a researcher at MIT's Haystack Observatory.
Detections teased but unconfirmed
In 2018, the Experiment to Detect the Global EoR Signature (EDGES), a partnership between Arizona State University and MIT Haystack, made headlines with a claimed global 21cm detection (Nature 555 67).
EDGES uses a dipole antenna, akin to a ping-pong table with a central slit, perched on a 30x30 meter metal sheet. Observations occurred in remote western Australia, shielded from radio noise.
Yet, seven years on, replication remains elusive.
EDGES' detected dip deviated sharply from predictions. 'Various models forecast specific signal ranges,' says Ravi Subrahmanyan, a CSIRO researcher in Australia. 'Measurements help validate or discard these.'
Standard models foresee a 0.1-0.2 Kelvin absorption trough due to cooler hydrogen versus warmer CMB, followed by rapid heating from stars, spiking emissions, especially from X-ray binaries. The signal fades as reionization ionizes hydrogen, halting the transition. Models estimate star counts and cosmic unfolding.
[Figure 2: Predicted 21cm signals from cosmological models (colored lines) versus EDGES' detection (dashed black line).]
'The signal's curve encapsulates myriad physical processes,' Fialkov notes. Its timing, slope, and depth mirror cosmic milestones.
But EDGES reported a dip twice the expected depth at 78 MHz, startling experts.
'This could necessitate groundbreaking physics,' de Lera Acedo says. 'First, confirm it's genuine.'
A game-changer in motion
EDGES' announcement stirred the field. 'It sparked a frenzy,' Bull recalls. 'It's ignited serious, thrilling research.' Some aim to replicate EDGES; others pursue model-aligned signals.
REACH, a Cambridge-Stellenbosch collaboration, targets 50–170 MHz on South Africa's Northern Cape plains. It employs a dipole and spiral cone antenna atop a star-shaped metal mesh to curb ground reflections.
'Precision engineering meets cosmology here,' de Lera Acedo, REACH's lead, states. Ground echoes, calibration flaws, soil signals—'these are the nemeses,' he warns. 'We refine noise reduction, analysis, and source filtering.'
From deserts to lakes to snow
SARAS, pioneered by India's Raman Research Institute in the late 2000s, has evolved to combat noise. Morphing from ground-based dipole to water-floating cone, it scans 40-200 MHz (Exp. Astron. 51 193).
Post-EDGES, SARAS shifted to verify the claim. 'We floated our radiometer to minimize ground issues,' says RRI's Saurabh Singh. In 2022, they cast doubt on EDGES (Nature Astronomy 6 607), yet many persist.
Singh emphasizes that non-detections are valuable: 'They eliminate models, constraining star and galaxy properties.'
Raul Monsalve Jara of UC Berkeley, an EDGES veteran, advocates diverse approaches. 'Multiple experiments with varied methods build trust.'
MIST, his co-led project with Chilean-Canadian-Australian-American teams, probes 25-105 MHz (MNRAS 530 4125). 'We simplified: no ground plate, portable for remote sites,' he explains. Sites must be pristine; irregular terrain complicates data. MIST's dipole has visited California and Nevada deserts, the Arctic, with Chile next. 'Mobility lets us test environments.'
Aaron Parsons of UC Berkeley sought utter isolation: suspending a rotating antenna over Utah's canyon, 100 meters aloft.
His EIGSEP (Electromagnetically Isolated Global Signal Estimation Platform) observes 50-250 MHz. 'Rotation aids calibration,' he notes. Ground antennas cross-check. Operations started recently.
More initiatives loom. Manchester's RHINO uses a mesh horn antenna, akin to skyscraper building material, for accuracy. It draws inspiration from the Holmdel Horn that serendipitously found the CMB in 1965. Starting at Jodrell Bank, it may roam.
Subrahmanyan, now at CSIRO after SARAS, designs GINAN—a self-calibrating antenna for 40-160 MHz. 'It delivers truer sky readings.'
EDGES persists in its third version: desk-sized, box-like top, bigger ground plate. They've observed from Canadian archipelago and Alaskan islands.
'Verification demands independent confirmation,' Cappallo stresses. EDGES replicated internally, but externally, no luck.
Scientists applaud diversity. 'Consensus requires multiple efforts,' Parsons asserts.
Heading lunarward
Some bypass Earthly hurdles, eyeing the Moon. Earth's ionosphere scrambles the signal; lunar far side blocks Earth's radio din.
'That's why lunar experiments exist,' Parsons says, involved in NASA's LuSEE-Night, launching next year.
In July's National Astronomical Meeting, de Lera Acedo proposed 'Cosmocube,' a lunar-orbiting nanosatellite, though a decade away. 'The future frontier.'
Meanwhile, terrestrial teams advance sensitivity, modeling, and insight. 'Proper execution will yield results,' Monsalve believes. The question lingers: which group nails it?
And this is the part most people miss: while EDGES' potential discovery promises revolution, its uniqueness raises skepticism—could it be an anomaly, or does it hint at unknown forces? Do you agree the signal demands exotic explanations, or might calibration quirks explain it? Is lunar detection the key to breakthroughs, or should we trust Earth-based persistence? Share your views below—let's debate the cosmos!
