CERN study finds early-universe quark-gluon plasma leaves a liquidlike wake

Deep beneath the French–Swiss border, inside a 27-kilometer ring of supercooled magnets, physicists say they have watched something that vanished from the universe more than 13 billion years ago slosh like water.

A liquid wake from the universe’s first moments

In a new analysis from the Large Hadron Collider (LHC) near Geneva, researchers report the clearest evidence yet that the “primordial soup” of quarks and gluons that filled the infant universe behaved as a true liquid—complete with wake-like patterns trailing behind fast-moving particles.

The result comes from the Compact Muon Solenoid (CMS), one of the LHC’s main detectors. It was highlighted Jan. 28 by the Massachusetts Institute of Technology and is based on lead-ion collision data recorded in 2018. The findings were published in Physics Letters B.

By smashing heavy lead nuclei together at nearly the speed of light, physicists briefly create tiny fireballs in which protons and neutrons “melt” into a state of matter called quark-gluon plasma (QGP): a fluid of free quarks and gluons at temperatures of several trillion degrees.

“It has been a long debate in our field on whether the plasma should respond to a quark,” said Yen-Jie Lee, a physics professor at MIT and a member of the CMS Collaboration. “What we see now is that this medium is incredibly dense and behaves like a liquid, producing splashes and swirls when a quark goes through. So quark-gluon plasma really is a primordial soup.”

Beyond global flow: isolating a single probe

For more than two decades, experiments at the Relativistic Heavy Ion Collider in the United States and at CERN’s LHC have shown that QGP behaves like a “nearly perfect” liquid, flowing with remarkably low viscosity. Much of that evidence has come from global flow patterns in heavy-ion debris and from “jet quenching,” in which energetic sprays of particles emerge from the plasma weakened and unbalanced.

What had been missing, researchers say, was a clean view of how the plasma responds to a single identified probe—analogous to spotting the wake from one boat rather than inferring ocean currents from a storm.

Z bosons as “silent partners”

CMS targeted a special class of events in which a hard scattering produces a Z boson along with a single high-energy quark or gluon recoiling in the opposite direction. The Z boson—electrically neutral and interacting only via the weak nuclear force—passes through the surrounding matter almost unaffected, making it a precise tag of the original recoil direction and energy.

“Z bosons are like silent partners,” Lee said in an MIT statement. “They carry information about the hard scattering, but they don’t disturb the medium on their way out. That allows us to use them as a pointer to where the quark went.”

From roughly 13 billion lead–lead collisions recorded in 2018 at a center-of-mass energy of 5.02 trillion electron volts per nucleon pair, the team identified about 2,000 events with a clearly reconstructed Z boson. In each, they measured how low-momentum charged particles—typically in the 1 to 2 billion electron volt transverse-momentum range—were distributed in angle relative to the Z.

To establish a baseline, the collaboration performed the same measurement in proton–proton collisions taken in 2017 at the same energy, where no QGP is expected to form.

A broadened pattern consistent with a wake

In proton–proton data, particles recoiling against the Z cluster in a relatively narrow pattern, tracking the jet of hadrons produced as the original quark or gluon fragments in vacuum.

In the lead–lead data, CMS observed a different signature: the correlations between the Z boson and soft hadrons were significantly modified compared with proton–proton events. Rather than a simple narrow peak, the distribution showed a broad, collective enhancement of low-momentum particles around the recoil direction, along with a depletion close to where the jet core would have been.

CMS reports this as “the first evidence of medium-recoil and medium-hole effects caused by a hard probe,” consistent with hydrodynamic calculations of a wake left when a fast parton plows through a dense liquid.

The result draws on theoretical work from the past decade, including a hybrid model pioneered by MIT physicist Krishna Rajagopal and collaborators. In that framework, a high-energy quark deposits energy and momentum into the QGP, which then evolves collectively according to relativistic hydrodynamics—producing a wake structure with both a “hole” and a surrounding swelling that should appear in the angular distribution of soft particles.

Daniel Pablos, a theorist at the University of Oviedo in Spain who helped develop wake models but was not part of the CMS experimental team, called the measurement “the first clean, clear, unambiguous evidence” of the predicted effect, according to MIT.

Why the wake matters

Seeing this wake is more than a compelling metaphor: it offers a new way to test the description of quark-gluon plasma as a fluid and to measure its fundamental properties.

By comparing the wake’s size and shape with hydrodynamic simulations, researchers aim to extract key “transport coefficients” of the plasma, including its shear viscosity, speed of sound, and how quickly disturbances dissipate. Those values feed into models of how the early universe expanded and cooled and may also inform understanding of matter at extreme densities, such as those inside neutron stars.

“This is essentially a snapshot of this primordial quark soup,” Lee said. “By watching how it responds when we kick it with a quark, we can start to measure how stiff it is, how fast sound travels through it, how quickly it relaxes. These are the basic properties of the first fluid in the universe.”

What comes next

The study also highlights the long timelines and international scale of modern particle physics. The data were collected in 2018 at a European laboratory supported by more than 20 member states, and the analysis drew on hundreds of scientists within the CMS Collaboration. U.S. groups—including MIT and Vanderbilt University—were supported in part by the Department of Energy.

CMS plans to repeat and extend the measurement with larger datasets from newer heavy-ion runs at the LHC, while theorists refine models to match the new level of detail. Future work may explore how wake features depend on the energy of the recoiling quark, the size of the QGP droplet, or the geometry of the collision.

For now, physicists say, the result marks a turning point in how concretely they can describe an exotic state of matter that no longer exists naturally.

We cannot travel back to the first microseconds after the Big Bang—but in a tunnel beneath the Alps, a single quark skimming through a droplet of man-made plasma has left ripples that let scientists see how that long-vanished fluid once flowed.