Every atom in your body and every grain of dust in every galaxy owes its existence to a tiny surplus left by the early universe. CERN describes the excess as approximately one extra matter particle for every billion antiparticles. Almost all the paired matter and antimatter annihilated, while the surplus became the ordinary matter from which stars, planets and living things eventually formed.
The imbalance has a name: baryon asymmetry. Measurements of the cosmic microwave background indicate that surviving baryons are vastly outnumbered by photons, preserving the scale of that primordial excess. Yet physicists still do not know which mechanism tipped the balance toward matter.
The visible universe is, in that narrow sense, what remained after an almost complete cancellation.
What the equations demanded
For every known matter particle, physics provides a corresponding antiparticle with the same mass but opposite charge or other reversed quantum numbers. An electron has the positron. A proton has the antiproton. When particles and antiparticles meet, their mass is converted into other forms of energy, commonly photons and other particles.
The hot early universe should have produced particles and antiparticles in pairs. If those populations had remained perfectly equal, their later annihilation would have left radiation but essentially no ordinary matter. Something during the universe’s first moments must therefore have created a small but durable bias.
That unknown process, called baryogenesis, had to operate before the surviving quarks combined into protons and neutrons. Nucleosynthesis, hydrogen, stars, chemistry and biology all followed from the excess it left behind over the next 13.8 billion years.
Sakharov’s three conditions
In 1967, Soviet physicist Andrei Sakharov set out the ingredients needed to generate a matter-antimatter imbalance. His proposal required baryon-number violation, violations of C and CP symmetry, and a period when the universe was out of thermal equilibrium. Sakharov’s original paper connected baryon asymmetry with CP violation and a nonstationary expanding universe.
CP violation is real. A 1964 neutral-kaon experiment demonstrated that matter and antimatter do not always behave as perfect mirror images. In 2025, CERN’s LHCb collaboration also observed CP violation in the decay of a beauty-lambda baryon.
The problem is scale. The CP violation contained in the Standard Model is many orders of magnitude too small to explain the matter excess observed across the universe. Physicists therefore continue searching for additional particles or interactions that could have supplied the missing asymmetry.
Enter the neutrino
Neutrinos are electrically neutral particles that pass through ordinary matter with very little interaction. They come in three flavours, electron, muon and tau, and can change from one flavour to another as they travel. This process, called oscillation, revealed that neutrinos have mass, something the original formulation of the Standard Model did not accommodate.
If neutrinos and antineutrinos oscillate differently, that would constitute CP violation in the lepton sector. It would provide another asymmetry for theories of the early universe to work with, although oscillation measurements alone could not explain baryogenesis without additional physics.
In October 2025, the NOvA and T2K collaborations published their first joint analysis. NOvA sends a beam 810 kilometres from Fermilab in Illinois to its detector in Ash River, Minnesota. T2K sends a beam 295 kilometres from the J-PARC accelerator to the Super-Kamiokande detector in Japan.
The combined data showed no strong preference for either neutrino mass ordering. If the inverted ordering is assumed, the results provide evidence for CP violation in the lepton sector. Under the normal ordering, however, CP-conserving values remain within the reported three-sigma interval. The result narrows the possibilities without settling the question.
The heavy cousin nobody has seen
Behind the search for neutrino CP violation is a broader idea called leptogenesis. Many versions propose that the three known neutrinos have extremely heavy, right-handed partners that existed only during the universe’s earliest, hottest moments.
These hypothetical heavy neutrinos would have decayed rapidly. If their decays produced slightly different numbers of leptons and antileptons, Standard Model processes could have converted part of that imbalance into the baryon asymmetry observed today.
No right-handed heavy neutrino has been detected. One important related search is neutrinoless double-beta decay, in which two neutrons would transform into two protons and two electrons without emitting the two antineutrinos expected in ordinary double-beta decay.
Experiments using xenon and germanium are looking for a narrow peak at the decay’s fixed energy. With no neutrinos carrying energy away, the two electrons would retain the full available decay energy. LEGEND-200 is taking data at Gran Sasso while testing whether neutrinos are their own antiparticles.
A confirmed signal would establish lepton-number violation and, under the standard interpretation, show that neutrinos have a Majorana component. That would be a major clue for leptogenesis, but it would not by itself identify the process that generated the universe’s matter excess.
Cosmic knots and 19th-century ghosts
Not every proposed explanation begins with neutrino oscillations. In October 2025, Gizmodo reported on a modern revival of Lord Kelvin’s 1867 idea that knots might have a fundamental role in physics. Kelvin’s aether theory was discarded, but the mathematics of knots survived.
In a Physical Review Letters paper published on August 29, 2025, Minoru Eto, Yu Hamada and Muneto Nitta showed that stable knot-like solitons can arise in an extension of the Standard Model containing a QCD axion and right-handed neutrinos.
Their model proposes that cosmic strings created during early phase transitions could become tangled into metastable knots. Those knots could temporarily dominate the universe’s energy density before collapsing through quantum tunnelling. Their collapse would produce heavy right-handed neutrinos whose later decays could generate a matter-antimatter asymmetry.
This remains a theoretical scenario, not evidence that a knot-dominated era occurred. Its advantage is that it offers an observational target: the altered early history could leave a gravitational-wave background detectable by future observatories such as LISA.
Reading the helium
Another clue may be preserved in the primordial abundances of hydrogen and helium. These elements formed during the universe’s first minutes, and their relative quantities depend partly on the behaviour and abundance of neutrinos and antineutrinos at the time.
A theoretical analysis based on Subaru Telescope observations of ten metal-poor galaxies found that the measured helium abundance was consistent with an early excess of neutrinos over antineutrinos. The authors argued that this neutrino asymmetry could have been transferred to ordinary matter through known and proposed particle processes.
The researchers stressed that the agreement does not prove this was the mechanism responsible. It demonstrates a viable route that fits the Subaru data, leaving further observations and independent tests to determine whether the clue survives. The original account explicitly described the result as a hint rather than a definitive solution.
What DUNE will do
The Deep Underground Neutrino Experiment is now moving from excavation into detector construction. Fermilab’s June 2026 project schedule calls for the first DUNE detector to begin operating in 2029.
DUNE will eventually send an intense neutrino beam about 1,300 kilometres from Fermilab to the Sanford Underground Research Facility in South Dakota. Its far detector will use tens of thousands of tons of liquid argon to record neutrino interactions, while a smaller near-detector complex at Fermilab will measure the beam before oscillation.
By comparing how neutrinos and antineutrinos change over that distance, DUNE will sharply constrain the CP-violating phase in neutrino oscillations. A large effect would strengthen some models of leptogenesis. A result near zero would rule out that particular low-energy source of CP violation, but neither outcome would independently prove or disprove every version of leptogenesis.
The size of the leftover
The visible universe is what remained after nearly all primordial matter and antimatter annihilated. With an exactly balanced starting population and no later symmetry-breaking process, ordinary matter would not have survived in anything resembling its present abundance.
That makes the matter excess small in ratio but enormous in consequence. Galaxies, stars, planets and living cells all descend from a difference that amounted to approximately one surviving matter particle per billion annihilating pairs.
Every day, detectors beneath mountains and across long underground baselines collect evidence from particles that barely interact with matter. Some measure neutrino oscillations. Others search for forbidden nuclear decays or gravitational-wave traces of speculative early-universe structures. Each approach tests a different route through the same unresolved problem.
The radiation of the hot early universe remains visible as the cosmic microwave background. ESA describes it as a 2.7-kelvin glow containing roughly 400 photons in every cubic centimetre of space. Against that vast sea of radiation, the surviving matter is rare, but it was enough to build everything we can see.
