Muon measurements embolden the search for physics beyond the standard model

“Have we glimpsed ‘new physics’ in the muon’s anomalous magnetic moment?” asked the title of a news story published in the April 2001 issue of Physics Today. Working at Brookhaven National Laboratory, the Muon g − 2 collaboration had measured the muon’s anomalous magnetic moment, a_μ, with a precision that surpassed by a factor of 14 that of previous measurements made at CERN in the mid 1980s. The result had uncovered a discrepancy: The more precise Brookhaven measurement was larger than the standard-model prediction.

The uncertainty of the a_μ measurement was still too high to claim a discovery—the value was just 2.6 standard deviations above the prediction, meaning the chance that researchers could find that result when no difference exists is about 1%. For the community to accept the result as a manifestation of physics beyond the standard model, the certainty would have to be at least 5 standard deviations, which corresponds to a chance of a false positive of less than 0.0001%.

So in 2013 the researchers shipped the storage ring used in the Brookhaven experiments nearly 1000 miles to Fermilab, where they could improve the experiment’s statistics by using a higher-intensity proton beam. When it smashes into a fixed target, the beam produces pions that decay into muons, and the number of muons produced is proportional to the number of incident protons.

Now, 20 years after the release of those initial results, the Muon g – 2 collaboration has published the first measurements of a_μ from Fermilab. The new value, based on data acquired in 2018, agrees with the previous experimental value and has an uncertainty of 3.3 standard deviations. Together, the Brookhaven and Fermilab data give a value for a_μ that differs from the standard-model value by 4.2 standard deviations—tantalizingly close to the 5 needed to claim a discovery.

Whether new physics lies in a_μ remains a question. The papers published by the Muon g – 2 collaboration include data from only the first of at least five planned Fermilab runs; results from the second and third runs should be published in the next year or two, and a fourth run is already underway. Improvements to the facility, magnet ring, and detectors should yield even higher precision in subsequent data.

Another experiment is in the works at J-PARC (the Japan Proton Accelerator Research Complex) to independently measure a_μ. Because J-PARC uses a different beam-generation technique and storage ring, its data will have different sources of systematic error. Agreement between the experiments would shore up confidence in their accuracy. (Muon g – 2 collaboration, Phys. Rev. Lett. 126, 141801, 2021 ; Phys. Rev. A 103, 042208, 2021 ; Phys. Rev. D 103, 072002, 2021 ; Phys. Rev. Accel. Beams, in press.)