Big Bang Nucleosynthesis
Linas JuozėnasShare
Knowledge Ark · Universe · The Grand Beginning
Before the stars, the first nuclei.
For a brief interval, the expanding universe was a nuclear furnace. Protons and neutrons assembled into light nuclei, leaving a mixture dominated by hydrogen and helium. That ancient mixture still helps us test what happened during the first minutes of cosmic history.
A chemical beginning we can still investigate.
Long before a star could shine, the universe had already made much of its helium. Understanding that early production means following a competition between nuclear reactions, energetic radiation, and the cooling caused by expansion.
Big Bang nucleosynthesis, usually shortened to BBN, describes this first episode of light-nucleus assembly. Ordinary hydrogen nuclei were mostly protons already present before fusion began. The newly built nuclei remained immersed in a hot plasma of particles and radiation.[1]
The window for building nuclei
Protons and neutrons were available before efficient nuclear assembly. What changed was the balance between reactions that could bind them together, radiation that could break fragile nuclei apart, and expansion that steadily reduced temperature and density.[1]
| Age after the hot Big Bang | Main change |
|---|---|
| Around the first second | Weak reactions gradually lose the ability to keep the neutron-to-proton ratio in equilibrium.[1] |
| The first few minutes | Deuterium survives long enough for reactions to build substantial helium-4.[1] |
| By roughly twenty minutes | The main episode is largely over as cooling and dilution make further nuclear assembly inefficient.[1] |
These are approximate stages, not exact boundaries shared by every reaction.[1]
Why helium ended up near one quarter
At high temperatures, weak interactions repeatedly converted neutrons into protons and protons into neutrons. Because a neutron is slightly heavier, cooling favored a smaller neutron fraction. Around the first second, interconversion became too slow to maintain equilibrium—a gradual process called freeze-out.[2]
The approximate ratio fell from about one neutron for six protons near freeze-out to one for seven when efficient nuclear assembly began. Nearly all the surviving neutrons then entered helium-4 nuclei, each containing two protons and two neutrons.[2]
Free-neutron beta decay contributed to that decline. It produces a proton, an electron, and an electron antineutrino. A free neutron’s mean lifetime is about fifteen minutes, corresponding to a half-life of about ten minutes.[3]
How does a 1:7 ratio give about 25% helium?
Let r be the neutron-to-proton ratio. If nearly every neutron enters helium-4, the helium share of all nucleons is approximately Yp ≈ 2r / (1 + r). With r = 1/7, this gives 1/4. Treating proton and neutron masses as approximately equal makes this nearly the mass fraction too.[2]
Why being hot was not enough
Deuterium, or hydrogen-2, is a nucleus containing one proton and one neutron. It provides the essential bridge to larger light nuclei. But in the early plasma, energetic photons could break newly formed deuterium apart almost as soon as it appeared.[4]
Deuterium’s binding energy is about 2.2 MeV. Nevertheless, it did not survive efficiently until the thermal energy scale kBT had fallen to roughly 0.08–0.1 MeV, around a billion kelvins. The reason is the enormous number of photons per baryon: even a small high-energy fraction could keep destroying it.[4]
From deuterium to helium—and a little lithium
Once deuterium could survive, several connected reactions rapidly built helium. The symbols below name nuclei: D is deuterium, T is tritium or hydrogen-3, and γ is a photon.[5]
The superscript counts nucleons: ³He has three, while ⁴He has four.[5]
Why much of the lithium began as beryllium
At the baryon density favored by observations, much of the primordial mass-seven material was made as beryllium-7. It later captured electrons and converted into lithium-7. That delayed conversion is part of what astronomers mean by the primordial lithium yield.[6]
Two distinct production routes
T + ⁴He → ⁷Li + γ
Helium-3 plus helium-4 makes beryllium-7; tritium plus helium-4 makes lithium-7 directly.[5]
The later electron-capture reaction is ⁷Be + e⁻ → ⁷Li + νe, with an electron neutrino emitted. The ionization state of the early plasma delays this process, so a neutral beryllium atom’s laboratory half-life is not the appropriate cosmic clock.[6]
Why the reactions did not fill the periodic table
There are no stable nuclei with mass numbers five or eight, interrupting easy steps beyond helium. Extended calculations produce extraordinarily small heavy-element traces, but almost no carbon, nitrogen, or oxygen compared with later stellar production.[7]
Mass fractions and nucleus counts are different
The broad outcome was approximately 75% hydrogen-1 and 25% helium-4 by ordinary-matter mass, with small amounts of other light nuclei. These percentages describe baryonic material; they do not include dark matter or dark energy.[8]
How we measure an ancient nuclear episode
Deuterium in distant gas
Gas clouds in front of bright quasars leave absorption lines in their spectra. The isotope shift separates deuterium’s signature from ordinary hydrogen. Carefully selected systems with little chemical enrichment give D/H around 2.5 × 10⁻⁵ by number—about one deuterium nucleus per forty thousand hydrogen nuclei.[10]
Helium in small galaxies
Emission lines from ionized gas reveal helium in galaxies with low abundances of heavier elements. Allowing for stellar enrichment, analyses infer a primordial helium mass fraction near 0.245, close to the simple quarter-by-mass picture. Temperature, density, and line modeling affect the precision.[11]
Deuterium is especially useful as a baryometer: in standard BBN, a higher baryon density generally allows more deuterium to be burned into heavier light nuclei, leaving less behind. Its residual abundance therefore constrains the cosmic supply of ordinary matter.[12]
The microwave background provides a separate measurement. Its acoustic pattern constrains the baryon density hundreds of thousands of years after nucleosynthesis, within a cosmological model. Comparing this with BBN tests whether one thermal history can explain both epochs.[13]
Broad agreement between deuterium, helium, and the CMB is a major success. At higher precision, the result depends on adopted nuclear reaction rates and observational uncertainties; some rate treatments yield a deuterium–CMB discrepancy. “Perfect agreement” would hide the work still being done.[12]
Why lithium remains a problem
Standard calculations predict more lithium-7 than is inferred from the surfaces of many old, metal-poor stars—roughly a factor of three more. This cosmological lithium problem is a real qualification to the otherwise successful light-element picture.[16]
A stellar surface is not an untouched sample of primordial gas. Mixing, diffusion, and nuclear destruction can change the lithium visible there. Researchers also test nuclear reaction inputs and possible early-universe physics, but no explanation has achieved an unambiguous, broadly accepted resolution.[16]
A proposed solution must preserve the successful parts of BBN. Changing an early reaction or adding a decaying particle can alter deuterium and helium as well as lithium, so reducing one mismatch is not enough.[16]
What the first minutes reveal about physics
BBN depends on more than nuclear fusion. Weak interactions set the neutron fraction; gravity determines the expansion rate; electromagnetic effects influence the plasma and nuclear collisions. That combination lets light-element abundances test the consistency of several parts of physics at once.[17]
The relativistic energy density is often expressed using Neff, the effective number of neutrino species. It is a way of describing radiation density, not a direct count of neutrino flavors. Standard neutrino decoupling itself gives a value slightly above three because the process is gradual.[18]
Additional radiation would speed expansion. Weak reactions would then lose equilibrium earlier, typically preserving more neutrons and increasing helium production. Measurements can constrain such changes, along with scenarios involving unstable particles or altered interactions, under the assumptions of each model.[17]
From nuclei to stars
For hundreds of thousands of years after BBN, the universe remained an ionized plasma. Neutral atoms became abundant near 380,000 years after the hot Big Bang, greatly reducing photon scattering. The CMB photons already existed in an older thermal radiation bath; they were not all created when atoms formed.[9]
The first stars followed much later, on the scale of hundreds of millions of years. Nuclear fusion inside stars began a new chapter of element production, building on the hydrogen and helium mixture inherited from the early universe.[19]
A few minutes left a record that spans the cosmos.
Every measurement of primordial deuterium or helium connects a distant cloud of gas with the same early nuclear history. BBN is powerful because its predictions can be tested—and because its remaining disagreements show us where the next questions lie.
Sources and further reading
Original studies, scientific reviews, and collaboration reports. Scientific context checked in September 2026. Nuclear diagrams show selected processes. The hydrogen–helium comparison uses a simplified 12-to-1 nucleus count, neglecting trace isotopes and small differences in nucleon mass.
- Gary Steigman (2003), Big Bang Nucleosynthesis: Probing the First 20 MinutesFollows the cooling plasma from neutron–proton interconversion to the end of efficient nuclear assembly.
- Fields, Molaro and Sarkar, Particle Data Group (2026), Big Bang NucleosynthesisConnects weak interactions, neutron survival, and the approximate primordial helium fraction.
- Particle Data Group (2025) — Neutron particle listingLists the neutron mass, mean lifetime, and beta-decay channel.
- Pitrou, Coc, Uzan and Vangioni (2018), Precision Big Bang NucleosynthesisCalculates gradual freeze-out and explains why deuterium survives only after substantial cooling.
- Alain Coc (2012), Primordial NucleosynthesisIdentifies the principal nuclear reactions that build deuterium, helium, and mass-seven nuclei.
- Khatri and Sunyaev (2011), Time of Primordial Beryllium-7 Conversion into Lithium-7Explains why primordial beryllium-7 converts to lithium only after capturing electrons later.
- Coc and colleagues (2012), Big Bang Nucleosynthesis up to Carbon, Nitrogen and OxygenTests an extended reaction network and finds extremely small primordial heavy-element yields.
- Cyburt, Fields, Olive and Yeh (2016), Big Bang Nucleosynthesis: 2015Explains the approximate helium mass fraction and the much smaller yields of other light nuclei.
- Olive and Peacock, Particle Data Group (2025), Big-Bang CosmologyDistinguishes nuclear assembly, later neutral-atom formation, and CMB last scattering.
- Cooke, Pettini and Steidel (2018), One percent determination of the primordial deuterium abundanceMeasures deuterium in nearly pristine gas silhouetted against distant quasars.
- Aver and colleagues (2022), Chemical abundances in the Leoncino Dwarf galaxyUses helium and hydrogen emission lines in metal-poor galaxies to estimate primordial helium.
- Pitrou, Coc, Uzan and Vangioni (2021), A new tension from primordial deuterium?Shows why nuclear reaction rates matter when comparing deuterium with precision cosmology.
- Planck Collaboration (2020), Planck 2018 results. VI. Cosmological parametersInfers the cosmic baryon density from the microwave background.
- Pisanti, Mangano, Miele and Mazzella (2021), Primordial deuterium after LUNATraces how underground measurements of deuterium burning sharpen cosmological calculations.
- Mossa and the LUNA collaboration (2020), Measuring deuterium burning at BBN energiesDescribes the underground apparatus and control of uncertainties in a key BBN reaction.
- Fields and Olive (2022), Lithium depletion and the primordial lithium problemExamines why old stellar atmospheres contain less lithium than standard nucleosynthesis predicts.
- Yeh, Shelton, Olive and Fields (2022), BBN and CMB tests of new physicsCompares the first minutes with the later microwave background and tests changes between the two epochs.
- Froustey, Pitrou and Volpe (2020), Neutrino decoupling and primordial nucleosynthesisExplains how neutrino energy and incomplete decoupling influence early expansion and nuclear yields.
- NASA Science, Cosmic HistoryPlaces the first stars and galaxies hundreds of millions of years after the first nuclei.
The Grand Beginning
- The Singularity and Moment of Creation
- Quantum Fluctuations and Inflation
- Big Bang Nucleosynthesis · You are here
- Matter vs. Antimatter
- Cooling and the Formation of Fundamental Particles
- The Cosmic Microwave Background (CMB)
- Dark Matter
- Dark Energy
- Recombination and the First Atoms
- The Dark Ages and First Structures
- Reionization: Ending the Dark Ages