ScienceNot yet confirmed elsewhere1 publisher2 min readPublished
Metal-poor galaxies yield a primordial helium fraction of 24.58% by mass
Astronomers on the LBT YP Project measured primordial helium at 24.58% of early matter by mass, against 24.67% from Planck's cosmic microwave background data. The agreement counts as a check on the Big Bang model only once helium made later by stars is separated from the original helium.
The Scientist · Science desk
Planck's cosmic microwave background estimate of primordial helium is 24.67%, 0.09 percentage points above the LBT YP value. About 90% of cosmic helium is primordial; the other 10% was formed through stellar fusion. The sample: interstellar gas regions within 54 galaxies.
| Figure | Value | Note | Claim |
|---|---|---|---|
| Planck cosmic microwave background helium estimate | 24.67% | 6 | |
| Planck above LBT YP | 0.09 points | 12 | |
| Cosmic helium that is primordial, about | 90% | other 10% formed through stellar fusion | 7 |
| Galaxies in sample | 54 galaxies | 4 |
What happened
- Helium is the most common byproduct of stellar fusion, so its relative abundance has risen over billions of years and skews measurements of the original amount.
- The sample was interstellar gas regions within 54 galaxies.
- Set against more local helium measures, the result implies about 90% of cosmic helium is primordial and about 10% came from stellar fusion.
- The work appeared in The Astrophysical Journal in 2026 as a series of papers, with Skillman, Aver and Yeh as first authors on Projects I, IV and V.
Why it matters
- constraint phys.org says the result strengthens one of the Big Bang model's three pillars, and that claim is only as good as the metallicity screen that removes stellar helium.
- capability The same primordial-versus-stellar split gives clues to the rate of star formation in galaxies and could help show how different elements are formed, according to the account.
- precedent Paper V is titled around the cosmological implications of the new value, so the screening assumption also sits under the series' wider conclusions.
Planck's figure is the higher of the two, by 0.09 percentage points, or about 0.36% of its value [12][13]. phys.org calls the match good [9]. The thing this summary doesn't tell you is how tight either number is, or how many gas regions sit behind the 54 galaxies [4]. Without error bars, a 0.09-point gap cannot be called large or small.
The direction of the gap is informative, within limits. Stars only add helium, and that accumulation over billions of years is what skews the raw data [1]. Contamination left in the sample would therefore push a measured value up, and this one sits below Planck's [12]. Other errors can go either way, so the gap is not evidence of unremoved stellar helium, but it does not show the screen is complete either.
The screen itself is indirect. The team chose galaxies with little heavy-element content and mostly early stars [2], on the idea that low metals mean no contamination from stellar helium [3]. Metal content is what gets observed. Stellar helium is what the team wants gone. How any stellar helium left in even the lowest-metal galaxies is handled is a question for the methodology paper, which covers sample selection, observations and methods [11].
The 90% primordial share is a second-stage number. It comes from setting the new primordial value against more local measures of helium [7], so it carries the uncertainty of both ends.
What to watch
- The uncertainties in the Project IV paper: whether the 0.09-point gap to Planck falls inside the error bars.
- How the Project I methodology treats stellar helium left in even the lowest-metal galaxies.
- Whether an independent sample of metal-poor galaxies reproduces 24.58% by mass.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence62
- Adoption
- Insufficient
- Hype gap+10
- Incentives
- Insufficient
- Confidence55
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
Not all observed helium formed in the Big Bang: helium is the most common byproduct of stellar fusion, so its relative abundance has been increasing over billions of years, and this skews the data.
- [2]
The LBT YP Project measured hydrogen and helium levels within metal-poor galaxies, which contain mostly early stars and not much in the way of heavy elements.
- [3]
The idea is that if the galaxies are low in metals, they also are not contaminated by stellar helium.
- [4]
The measurement is based on a sample of interstellar gas regions within 54 galaxies.
- [5]
The team found primordial helium makes up 24.58% of the initial matter of the cosmos by mass.
- [6]
The estimate from Planck observations of the cosmic microwave background gave a primordial helium result of 24.67%.
- [7]
Comparing the result to more local measures of helium abundance, the team found about 90% of cosmic helium is primordial and the other 10% was formed through stellar fusion.
- [8]
The primordial-versus-stellar split gives clues to the rate of star formation within galaxies and could help better understand how different elements are formed.
- [9]
The 24.58% result agrees well with the Planck estimate of 24.67%.
- [10]
The result further strengthens one of the three pillars of the big bang model.
- [11]
The results were published in The Astrophysical Journal (2026) as a series: Skillman et al., Project I (project description, sample selection, observations, methodology); Aver et al., Project IV (a new value of the primordial helium abundance); Yeh et al., Project V (cosmological implications of a new determination of primordial 4He).
- [12]
Planck's 24.67% exceeds the LBT YP value of 24.58% by 0.09 percentage points.
- [13]
The 0.09-point gap is about 0.36% of Planck's 24.67% value.
Sources
1 independent publisher whose own reporting we read for this story.
- phys.orgPrimordial helium further confirms our understanding of the Big Bang
1 article · October 12, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.