Fast Radio Bursts Reveal 76% of Universe's Missing Ordinary Matter Lurks in Intergalactic Gas

Cosmologists have long faced a puzzle: the universe should contain a certain amount of ordinary matter—the stuff of atoms, stars, and galaxies—based on measurements from the early cosmos and the Big Bang itself. Yet when astronomers added up everything they could see directly, the total fell short. Most of that missing material was never thought to have vanished. Instead, it was believed to be real, widespread, and simply too faint to detect with conventional methods.

A new analysis of 69 precisely located fast radio bursts, published in June 2025 in Nature Astronomy, has now placed most of that elusive ordinary matter in the thinnest reaches of space. The study, led by Liam Connor and colleagues, estimates that 76 percent of ordinary matter—with an uncertainty range of plus 10 to minus 11 percentage points—resides in the diffuse gas between galaxies. Another 15 percent occupies the extended halos surrounding galaxies, while just 9 percent exists as stars and cold gas within galaxies themselves. This is not yet settled consensus, but it represents the most discriminating partition of the late-universe baryon budget that fast radio burst measurements have yet provided.

Fast radio bursts are brief flashes of radio energy arriving from distant galaxies. Their physical origins remain uncertain, but what matters for this census is how they travel. As a burst passes through ionized gas—regions where electrons float freely—lower radio frequencies arrive slightly later than higher ones. This delay, called the dispersion measure, is proportional to the total number of free electrons the signal encountered along its entire path. Gas too faint to photograph directly still leaves a measurable timing signature. Once astronomers identify a burst's host galaxy and measure how far away it is, they can convert that electron column into a statement about how much plasma the signal passed through. The challenge lies in separating contributions from the Milky Way, the host galaxy, and intervening galactic halos before attributing the remainder to intergalactic space.

Localization proved crucial. Thirty-nine of the 69 bursts came from the Deep Synoptic Array-110 at California's Owens Valley Radio Observatory, an instrument designed specifically to pinpoint fast radio bursts accurately enough for follow-up observations. The remaining 30 came from telescopes worldwide, primarily the Australian Square Kilometre Array Pathfinder. Optical observations from facilities including Keck and Palomar observatories helped establish the distances to these host galaxies. The 69 localized bursts ranged from roughly 11.74 million to 9.1 billion light-years away, with the most distant being FRB 20230521B. This long baseline provided both nearby and distant probes of the accumulated electrons between galaxies rather than relying on a narrow slice of cosmic history.

The intergalactic gas the study describes is highly ionized and extraordinarily diffuse—with a density below roughly 0.001 particles per cubic centimeter and a neutral fraction below one part in 10,000. Those properties explain why it has been so difficult to detect: the gas can dominate the ordinary-matter inventory while producing almost no light of its own. The 76 percent figure is a statistical inference, not a direct count of individual particles. It depends on foreground estimates, simulations of where electrons reside, and assumptions about how gas distributes itself in the cosmic web. The quoted uncertainties reflect the statistical model, but they do not guarantee that every modeling assumption is complete.

This result carries implications for how galaxies form and evolve. If roughly three-quarters of ordinary matter lies outside galactic halos, galaxies have not retained all the gas that gravity pulled toward them. Supernovae, stellar winds, and accreting supermassive black holes can heat gas and drive it outward into the broader cosmic web. The study argues that relatively efficient feedback is needed to reproduce the inferred split, giving galaxy-formation models a new observational constraint. Matching the number and appearance of galaxies is not sufficient if a simulation stores too much ordinary matter inside their halos. The baryon census tests the otherwise faint aftermath of billions of years of outflows.

The method's power lies in its simplicity: electrons that are nearly invisible in emission still delay radio waves. Its limitations follow the same reasoning. Each burst mixes contributions from multiple sources, and errors in any component can shift the inferred partition. Selection effects also matter—only localized bursts with usable host information enter this kind of census, and that subset may not perfectly represent the full fast radio burst population. Sixty-nine events are enough to sharpen the fractions, but not enough to erase every ambiguity. Future arrays should supply thousands of localized sightlines, allowing the budget to be tested across different redshifts, galaxy types, and foreground environments. For now, the clearest conclusion is deliberately bounded: most ordinary matter long missing from visible inventories is consistent with very diffuse, ionized gas between galaxies.