STAR Data Reveal 'Splash' of the Quark-Gluon Plasma

Reconstruction of jets produced in particle smashups reveals how 'lost' energy is dispersed sideways

June 11, 2025

UPTON, N.Y. — New data from particle collisions at the Relativistic Heavy Ion Collider (RHIC), an “atom smasher” at the U.S. Department of Energy’s (DOE) Brookhaven National Laboratory, reveal how the primordial soup generated in the most energetic particle collisions “splashes” sideways when it is hit by a jet of energetic particles. The evidence comes from the first measurement at RHIC of reconstructed jets produced in the collisions back-to-back with photons, particles of light. Scientists have long anticipated using measurements of photon-correlated jets to study the matter generated in these collisions. The findings, described in two papers just published in Physical Review Letters and Physical Review C, offer fresh insight into this primordial soup, which is known as a quark-gluon plasma (QGP) — and raise new questions about its extraordinary properties.

“Measuring reconstructed jets gives us unique views into how the strongly interacting plasma responds as the jets move through it,” said Peter Jacobs, a physicist at DOE’s Lawrence Berkeley National Laboratory and member of RHIC’s STAR Collaboration, which published these results. “Instead of focusing on what happens to the jet, we want to turn it around and see what the jet can tell us about the QGP.”

Big Bang soup

The entire universe was filled with a QGP a fraction of a second after the Big Bang, before the quark and gluon building blocks of matter became bound together in protons and neutrons — and eventually atomic nuclei. RHIC, a DOE Office of Science user facility for nuclear physics research, routinely creates a QGP by accelerating the nuclei of gold atoms close to the speed of light and smashing them together. The energetic collisions “melt” the nuclei, setting free their inner building blocks, so scientists can study the quarks and gluons as they existed at the dawn of time.

This is not easy. The QGP created at RHIC lasts for less than a trillionth of a trillionth of a second! There’s no time to put it under a microscope or bombard it with X-rays the way scientists who study other forms of matter do. But fortunately, the collisions that create the QGP sometimes also knock individual quarks or gluons out of the nuclei with enormous energy. These energetic scattered particles rapidly decay into cascades of correlated particles — jets — that carry information about the plasma to RHIC’s detectors.

“The scattered quarks and gluons come along with the nuclei; they are inside the matter,” Jacobs said. “We can use them like X-ray beams to learn about the plasma.”

Triggering photons and tracking jets

Many groups of scientists studying jets at RHIC have focused on a phenomenon known as jet quenching, an apparent suppression of energetic jets emerging from the QGP. The idea is that jets are losing energy through their interactions with the QGP.

RHIC’s measurements of jet quenching to date have focused primarily on the most energetic, leading jet particles, because they are straightforward to measure. However, such leading particles provide only limited insight into the process. The new results from STAR reconstruct a wider correlated spray of particles making up the jets, revealing much more detail about how the QGP is “excited” and responds to the jet — and where the “lost” energy goes.

The new analysis, for the first time, included the reconstruction of jets produced back-to-back with photons.

“Since photons don’t interact at all with the QGP, their energy, as measured by the detector, provides a gauge for comparing with the energy of the jet particles emerging in the opposite direction,” said co-author Saskia Mioduszewski, a STAR collaborator from Texas A&M University. Mioduszewski pioneered the direct photon identification technique with former postdoc Ahmed Hamed, now at the American University of Cairo. The effort was later joined by Nihar Sahoo, also a former postdoc, now at the Indian Institute of Science Education and Research-Tirupati, and Derek Anderson, one of Mioduszewski’s former graduate students who will be joining DOE's Thomas Jefferson National Accelerator Facility on June 16.

For the new papers, Sahoo proposed putting this photon analysis technique together with statistical methods developed by Jacobs and Alex Schmah, now at the GSI Helmholtz Centre for Heavy Ion Research in Germany, to pick up the subtle signals of even low-energy jets from the background of thousands of other particles produced in RHIC collisions.

“Using the fine-tuned photon identification algorithm and these jet-reconstruction techniques, we could find collisions with an energetic photon and then reconstruct all of the jets back-to-back with the photon,” said Anderson, who led the analysis with Sahoo.

“Identifiying the photons is tricky,” Anderson added. “We only want so-called ‘direct photons’ — the ones produced directly in the collision at the same time as the energetic jets. But RHIC collisions produce a lot more photons through other processes. The shape of the signal in the STAR detector helps us identify direct photon candidates. Then we use statistical techniques to remove the not-direct photons so we can reconstruct the jets that come out within a particular angular window in the opposite direction.”

Crucially, the team reconstructed jets with different angular “cone” sizes — some more narrow and some wider. Widening the cone through which they searched for correlated particles allowed them to observe the response of the QGP to jet excitation in a new way. 

Sideways splash

The team studied data from both proton-proton collisions, which do not generate a QGP, and head-on gold-gold collisions that routinely generate the primordial soup. In each type of collision, they looked for jets in a narrow cone, where only the most energetic jet particles would be observed, and a broader cone designed to catch any correlated particles on the periphery of the jet.

“When a jet emerges from a proton-proton collision with no QGP forming, there should be lots of particles inside the narrow cone and very few outside it,” Sahoo said. That’s the baseline the scientists use for comparison with the gold-gold collisions that create the QGP. “If the same particle goes through the QGP, there should be fewer energetic particles in the narrow cone. But because of all the extra interactions of the jet particles with the QGP, there should be more particles in the wider cone.”

This is exactly what the scientists found. The broader observation cone and fine-tuned analysis allowed them to pick up the signals of particles created through branching interactions of jet particles with the QGP. Adding up the energy of all those extra correlated particles accounted for the “missing” energy of the quenched jets.

“We found that energy within jets is distributed more broadly in collisions that produce the QGP compared to those that do not,” Jacobs said. “It’s like stuff is splashing sideways.”

For comparison, think about riding a bike and hitting a puddle of water. As you go through the puddle, the water splashes outward, and you slow down. The bike is like the jet going through the QGP, giving up bits of energy to the sideways interactions with the plasma’s free quarks and gluons.

“In both cases, the energy goes somewhere; it’s not ‘lost,’” Jacobs said. “What you learned in high school is still true; the energy is conserved.”

Open questions

The measurements revealed that a cone with a 30-degree opening angle is sufficient to recover much of the initial jet energy. This sets a limit on the distance over which the QGP excitation travels. That could have implications for understanding the viscosity of the QGP, which has been described as a nearly perfect fluid with frictionless flow.

In addition, Mioduszewski said, “To really characterize the energy loss and the QGP response, we need to understand how the energy loss depends on the path length, or the distance a jet travels through the plasma, and the strength of its interactions with the plasma.”

The data the scientists collected from different types of jets gave them clever ways to explore these more subtle characteristics.

“These are all new measurements, and we’re going to have to work with our theory colleagues to put all this data together to see if there’s a consistent picture,” Mioduszewski said.

This work was supported by the DOE Office of Science, the U.S. National Science Foundation (NSF), and a range of international agencies and organizations listed in the scientific paper. In addition to using the Open Science Grid, supported directly by NSF, the researchers made use of computing resources in the Scientific Data and Computing Center at Brookhaven Lab and the National Energy Research Scientific Computing Center (NERSC), which is another DOE Office of Science user facility at DOE’s Lawrence Berkeley National Laboratory.

Brookhaven National Laboratory is supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit science.energy.gov.

Follow @BrookhavenLab on social media. Find us on Instagram, LinkedIn, X, and Facebook.

2025-22311  |  INT/EXT  |  Newsroom