Imagine two careful surveyors measuring the same stretch of road. One uses an old, trusted map. The other walks it with a tape measure. Both are meticulous, and yet their answers don't match.
That, in essence, is one of the biggest puzzles in modern cosmology. Two of our best ways of measuring how fast the universe is expanding give stubbornly different results. Now a team of physicists thinks a surprising ingredient, extremely faint magnetic fields left over from the Big Bang, might help bring the two numbers closer together.
Two rulers, two answers
The universe's expansion rate is called the Hubble constant, after Edwin Hubble, who showed in the 1920s that distant galaxies are moving away from us. It is quoted in kilometres per second per megaparsec. A megaparsec is about 3.26 million light years, so the number tells you how much faster a galaxy recedes for each extra stretch of that size.
The first method is indirect. Telescopes such as the Planck space telescope have studied the cosmic microwave background, the faint afterglow of the Big Bang. Tune the standard model of cosmology to match the tiny patterns in that ancient light, and it predicts a Hubble constant of about 67.
The second method is more direct. Astronomers watch a certain kind of exploding star, a type Ia supernova, in distant galaxies. These blasts are known as "standard candles" because they shine with the same intrinsic brightness, so how dim one appears reveals how far away it is. Calibrated with the Hubble and James Webb space telescopes, this approach gives a higher value, around 73.
A gap between 67 and 73 might sound small, but the researchers describe it as statistically highly significant. If both methods are right, the standard model must be missing something.
A magnetic idea
Enter Karsten Jedamzik of the French National Centre for Scientific Research, Levon Pogosian of Simon Fraser University in Canada, and Tom Abel of SLAC National Accelerator Laboratory and Stanford University. Their idea was first proposed to solve a different mystery: where the magnetic fields threading galaxies and galaxy clusters came from. Perhaps, they suggest, tiny "primordial" magnetic fields were born in the first instants after the Big Bang.
Here is why that could matter. About 380,000 years after the Big Bang, electrons and protons joined to form the first hydrogen atoms, an event called recombination. The universe turned from an opaque fog into a transparent space, and the light released then is the microwave background we see today.
The researchers argue that magnetic fields would have pushed and pulled on charged particles, making matter slightly clumpy. Think of a crowded dance floor: where people are packed together, they pair up faster. In the same way, hydrogen would have formed a little sooner. That shifts the moment the light was released and changes the size of the patterns in the microwave background, which acts as a kind of cosmic ruler. Recalibrate the ruler, and the Hubble constant inferred from it moves upward, towards the supernova value.
Pogosian and Jedamzik showed this in 2020 using a simplified model. In their new paper, published in Nature Astronomy, the team says it used the first full three-dimensional simulations of the early plasma with magnetic fields embedded in it. They then tested the predictions against observations including Planck data, the DESI galaxy survey and supernovae.
A hint, not a discovery
The data favour present-day field strengths of about 5 to 10 pico-Gauss. A pico-Gauss is a trillionth of a gauss, a standard unit of magnetic strength. It is almost unimaginably faint, and SLAC likens it to around a trillionth of the strength of a fridge magnet.
How convincing is the evidence? Scientists measure this in "sigma", a yardstick of how unlikely a result would be if it were just chance. Physicists usually want 5 sigma before they speak of a discovery. According to the paper, the preference for magnetic fields is about 1.8 sigma using Planck and DESI data, rising to about 3 sigma when supernovae calibrated to the local expansion rate are added.
Writing in The Conversation, the authors put the range across different combinations of data at about 1.5 to 3 sigma, and they are plain about what that means: this is "not yet a discovery, but a meaningful hint". Importantly, the idea could have been ruled out by the microwave background data, and it was not.
There is a bonus. The favoured strength is close to what would be needed for the magnetic fields in galaxy clusters to originate entirely from primordial seeds, without extra amplification. "Turns out, the number that we need to fix the Hubble tension is the same one we need to explain galaxy clusters," Abel said in a SLAC statement. "It's a very simple and satisfying solution."
The team is allowing itself a little excitement. "As theoretical physicists, we often work on ideas that have a very small chance of being true," Pogosian told SLAC. "This is actually not that far from having a chance to be true, so there is a real thrill for us."
Confirming or ruling out the idea will take sharper measurements, and the paper says future high-resolution microwave background observations will be crucial. For now, the picture is of a puzzle that may, just possibly, have a whisper of an answer, written in magnetism roughly a trillion times fainter than the magnet on your fridge door.
