News Release

Elusive particle may point to undiscovered physics

Peer-Reviewed Publication

Cornell University

ITHACA, N.Y. - The muon is a tiny particle, but it has the giant potential to upend our understanding of the subatomic world and reveal an undiscovered type of fundamental physics.

That possibility is looking more and more likely, according to the initial results of an international collaboration - hosted by the U.S. Department of Energy's Fermi National Accelerator Laboratory - that involved key contributions by a Cornell team led by Lawrence Gibbons, professor of physics in the College of Arts and Sciences.

The collaboration, which brought together 200 scientists from 35 institutions in seven countries, set out to confirm the findings of a 1998 experiment that startled physicists by indicating that muons' magnetic field deviates significantly from the Standard Model, which is used to explain the laws that govern fundamental particles.

Digitizer modules undergo testing in the lab of Lawrence Gibbons, professor of physics, before being shipped to the Fermi National Accelerator Laboratory. Twenty-eight crates of these modules were installed around the muon g-2 ring.

"The question was, what's going on? Was the experiment wrong? Or is the theory incomplete?" Gibbons said. "And if the theory is incomplete, then confirming what's going on becomes the first terrestrial evidence of a totally new kind of fundamental particle or force that we don't know about. It would be the first experiment on Earth that is sort of the equivalent of the discovery of dark matter in space."

On April 7, the team confirmed that the original findings were correct, which means there must be more to the physics surrounding the muon than previously known.

Muons are like electrons but are more than 200 times more massive. Both are essentially tiny magnets with their own magnetic field. Muons are far more unstable, though, and decay in a few millionths of a second. They are also notoriously difficult to observe at the quantum mechanical level because the vacuum in which they exist is not a big empty cavity, but rather a bubbling, frothing, dynamic environment.

"It's your cappuccino foam version of the vacuum, where there's virtual particles winking in and out of existence all the time," Gibbons said. "And that turns out to affect the strength of the magnetic field of a muon."

To figure out why, researchers at Brookhaven National Laboratory 20 years ago set out to measure the absolute strength of muon's magnetic field. They did this by firing a beam of muons into a 14-meter-diameter magnetic ring at nearly the speed of light while a series of detectors captured data. The scientists discovered a major discrepancy in the muon's magnetic field: It was more than 3.5 standard deviations from the Standard Model predicted by theoretical physicists.

A plan was eventually hatched to repeat the Brookhaven experiment with higher precision. In 2013, the Brookhaven magnetic ring was transported to the Fermilab facility in Batavia, Illinois, where it was coupled with an even stronger particle accelerator that could produce more than 20 times the amount of muons. In 2018, the first of several experiment runs was launched.

This muon g-2 experiment - "g" refers to the value of the magnet's strength caused by its intrinsic spin, which is slightly larger than two - was successful thanks to a system of detectors developed through a joint partnership between Cornell and the University of Washington.

The University of Washington group built a set of 24 calorimeters out of lead fluoride crystals and silicon photomultipliers that measure a blue light, known as Cherenkov radiation, that results when the positrons from muon decay strike the crystals. By measuring the time and amount of light for each of about 8 billion positrons, the researchers can pinpoint the muon's precession rate, which is the frequency of its rotational wobble. The rate is directly related to the value of g-2.

The Cornell team built the digitizers that could look at the electronic signal coming out of the detectors and create a digitized version of the waveform that could be analyzed offline. The researchers were supported in the effort by the Laboratory for Elementary-Particle Physics (LEPP), and their digitizers incorporated $200,000 worth of specialized analog-to-digital converter chips donated by Texas Instruments.

Gibbons' group also built one of the pair of reconstruction packages that helped their collaborators parse and analyze the collected data, and they were assisted in getting the most precise measurements by David Rubin, the Boyce D. McDaniel Emeritus Professor of Physics (A&S), who helped correct for the spread of muon momenta in the stored beam and for the small vertical motion as the beam speeds around the magnetic ring. Two other Cornell faculty, Toichiro "Tom" Kinoshita, professor emeritus of physics, and G. Peter Lepage, the Goldwin Smith Professor of Physics, both in A&S, contributed to the Standard Model prediction of g-2, to which the project compared its results.

As a fitting final touch, Gibbons chose to make the digitizer faceplate Cornell red.

With so much subatomic information to be sifted through, six different groups worked to separately confirm the muon's precession frequency. Gibbons helped design blinding software that would ensure the groups made their calculations independently.

Then the time came to compare results.

"I have to say, it was nerve-racking. You go into the room, and there's all these points scattered all over the place from all the offsets, and you have to decide, OK, are we going to compare results now? And will they agree?" Gibbons said. "We were trying to measure something to 500 parts per billion. The range that we had was plus or minus 25 parts per million on the frequencies that we're trying to measure. There was a huge sigh of relief when we found everything agreed beautifully."

And when all the international collaborators came together online for the final unblinding of the magnetic field measurement and checked it against the original Brookhaven result?

"Oh man. It was like hats flying in the air," Gibbons said. "It was a combination of elation and relief."

The results from this first experimental run represent only 6% of the data the researchers hope to eventually collect. Additional analysis has already begun on a second and third run, which will generate three to four times as much data. It will be 10 years before all the analysis is complete.

"We landed right on top of this result that really could indicate that there's something totally new going on. We really want to push the uncertainty, the precision, to make the strongest possible statement that we can experimentally," said Gibbons, who began work on the project in 2011. "We may be onto something really profound, something we don't understand. And we still have to figure out what it is."

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.