Retired Brookhaven Collider Shows Jets Losing Energy in Tiny Oxygen-Collision Plasma Drops

A collider that has already been switched off has just handed physicists a result they had been waiting on. Researchers with the STAR experiment at Brookhaven National Laboratory's Relativistic Heavy Ion Collider, known as RHIC, report that head-on collisions of oxygen nuclei show about a 20% suppression of high-energy particles. The effect is called jet quenching, and the study was published Oct. 20 in Physical Review Letters.

RHIC sits in Upton, New York, on Long Island. It fired its final beams on Feb. 6, 2026, after roughly 25 years of operation, and its tunnel is slated to be reused for the planned Electron-Ion Collider. The oxygen data behind the new paper were recorded in 2021 at 200 billion electron volts, so the analysis arrived well after the machine itself went quiet.

The physics starts with what happens when two nuclei hit head-on at RHIC energies. Protons and neutrons can melt into their building blocks, quarks and gluons, forming quark-gluon plasma, the matter believed to have filled the universe in its first microseconds. Brookhaven says a drop made this way is smaller than a single atomic nucleus and lasts less than a billionth of a billionth of a second. Some collisions also throw out jets, sprays of particles from a hard hit between quarks or gluons. A jet that must push through plasma on its way out loses energy.

Li Yi, a physics professor at Shandong University who oversaw the analysis, offered an analogy: it is like shooting a bullet through water, where resistance slows the bullet. To look for that slowing, the team compared head-on oxygen collisions, which produce the most particles, with glancing ones in which the nuclei barely touch and no plasma is expected. The head-on events had fewer high-energy particles, consistent with jets losing energy.

The result matters because of where it was found. Oxygen has 16 protons and neutrons, against 197 in gold. In gold collisions jet quenching has been seen for years, and by 2005 RHIC data showed the plasma behaving like a nearly perfect liquid, according to CERN Courier. Small systems have been more contentious. Flow patterns hinting at plasma have appeared in small collisions at both RHIC and Europe's Large Hadron Collider, but quenching had not been clearly observed. Rongrong Ma, a STAR physicist at Brookhaven, said no unambiguous signal had been seen, so the question remained under debate. The oxygen data add a clear quenching signal in a small system. The suppression is weaker than in gold collisions, as expected for a smaller drop of plasma.

The caveats are plain. The work is peer-reviewed and was posted earlier this year as an arXiv preprint, but it is one measurement, at one energy, from one experiment, and physicists still debate the minimum conditions needed to create quark-gluon plasma.

The next test is already visible. Oxygen is the only nucleus besides protons collided at both RHIC and the Large Hadron Collider, which runs at energies above 5,000 billion electron volts, so the two can be compared directly. Yi expects the quenching signal to be bigger at the LHC, since higher energy should yield hotter, denser plasma. The team also plans to study pairs of jets produced near the edge of the collision zone; Maowu Nie, an associate professor at Shandong University, said that would make it easier to look for asymmetric quenching of di-jets.

The work was supported by the Department of Energy's Office of Science, the National Science Foundation and international agencies, and used computing at Brookhaven and Lawrence Berkeley National Laboratory. It was also a training ground: Sijie Zhang did the analysis for her master's thesis at Shandong University and is now a Ph.D. student at Yale, while former Shandong student Yang He set up the analysis framework.

RHIC's data are expected to keep yielding results for at least another decade, CERN Courier reported. The sPHENIX detector alone recorded more than 200 petabytes of raw data, including 40 billion gold collisions, more than all earlier RHIC datasets combined. Much of the hardware will carry over to the Electron-Ion Collider, which will collide electrons with protons and nuclei to map how quarks and gluons are arranged inside matter. Comparisons with LHC oxygen collisions will be the next check on whether tiny drops of early-universe matter really form.