Scientists Found New Way to See Inside Atomic Nuclei with Tiny Explosions

A pair of tiny atomic nuclei slam together at nearly the speed of light, and for a fraction of a second, the building blocks of matter dissolve into a superhot fluid that flows like nothing else in the known universe. Scientists at CERN have now measured how that matter flows, and the patterns match a tiny droplet of that superhot fluid, its behavior shaped by the nuclei themselves.

Physicists working with the ALICE detector at CERN’s Large Hadron Collider published the first-ever measurements of how matter flows outward in oxygen-oxygen and neon-neon collisions, at a collision energy of 5.36 trillion electron volts per nucleon pair. Their results support an idea that has long been debated: even in these extremely small collisions, a superhot fluid of quarks and gluons, the particles that make up protons and neutrons, forms and flows in ways directly shaped by each nucleus’s physical structure.

Physicists have long known that slamming massive atomic nuclei like lead together creates a fleeting state of matter called a quark-gluon plasma, believed to have filled the entire universe in the first microseconds after the Big Bang. This plasma behaves like a nearly frictionless liquid. Whether the same thing could happen in collisions of far smaller nuclei had been an open question. These new results, published in Physical Review Letters and selected as an Editors’ Suggestion, offer strong evidence that it can.

Two Tiny Nuclei, Two Very Different Shapes

Part of what makes this experiment so interesting is what physicists already know about the internal architecture of these two nuclei. Oxygen-16, made of 16 protons and neutrons, has an internal arrangement resembling a three-dimensional triangular pyramid, with particles grouped into clusters of four. Neon-20 looks more like a bowling pin at the nuclear scale, with an extra cluster attached to an oxygen-like core. These shapes come from advanced nuclear-structure calculations rather than any literal snapshot of the nuclei themselves.

When two nuclei collide head-on, that shape gets imprinted on the blob of hot matter created at the moment of impact. As the fireball expands and cools, those initial geometric features drive matter to flow more strongly in some directions than others. By measuring which directions particles fly out of the collision, scientists can work backward, with the help of models, to infer what the original shape looked like, using the fireball as an indirect window into the nucleus’s structure.

Billions of Collisions, Carefully Counted

To get these measurements, the ALICE Collaboration used data from short “light-ion run” periods at the LHC in July 2025. About three billion oxygen-oxygen collisions and 400 million neon-neon collisions passed the quality checks required for analysis, a dataset large enough to reliably detect the subtle flow signals the team was hunting for.

Particles from each collision were tracked using ALICE’s inner tracking system and a large cylindrical detector that records particle paths through a gas-filled volume. Analysts measured two types of flow: elliptic flow, which reflects how oval-shaped the initial collision zone was, and triangular flow, which reflects more irregular, fluctuation-driven geometric features. Both were measured by looking for patterns in the direction in which groups of particles tend to travel together.

To rule out false signals, the team required that particle pairs come from well-separated regions of the detector. Potential errors from choices made in the analysis were estimated by varying selection criteria and found to be within a few percent.

Neon Flows Harder Than Oxygen, and Nuclear Shape Explains Why

One of the clearest results is a difference in elliptic flow between the two collision systems in their most head-on collisions. Neon-neon collisions show stronger elliptic flow than oxygen-oxygen collisions in these central events, with the ratio of neon-to-oxygen elliptic flow peaking at around 1.08 before decreasing. Researchers attribute that difference to neon’s stronger nuclear deformation, which the models describe as a bowling-pin-like shape that stamps a stronger oval-like geometry onto the initial fireball than oxygen’s more symmetric, pyramid-like structure.

Triangular flow tells a different story. It is associated mainly with random variations in where the protons and neutrons happen to sit inside the nucleus at the moment of collision, rather than with overall nuclear shape. Its behavior across collision types matches what has been seen in other small collision systems. Even so, the oxygen-neon comparison hints that nuclear structure enters here too: the authors link the smaller triangular-flow ratio in central collisions to oxygen’s tetrahedral, pyramid-like configuration.

Both types of flow also showed patterns never seen before in proton-proton or proton-lead collisions at the LHC. That includes a specific trend in the four-particle elliptic flow measurement, a signature of geometry-driven flow that is absent in smaller collision systems.

When the Models Know the Nuclear Shape, They Get It Right

When the ALICE team compared their measurements to theoretical predictions from a framework called Trajectum, which incorporates detailed nuclear structure from two independent, first-principles computational approaches, the agreement was strong across both collision systems and multiple flow measurements. Predictions built from first-principles simulations of nuclear structure reproduced both the trends and magnitudes of the flow measurements across a wide range of collision geometries. A second modeling approach, starting from energy-minimized nuclear configurations, also showed reasonable agreement, particularly in non-central collisions.

According to the paper, this level of agreement “equal or surpass the accuracy of full Bayesian parameter extraction in previous heavy-ion studies” involving much larger nuclei. The paper also notes that a particular problem that has long troubled theoretical descriptions of proton-proton collisions, where a specific four-particle flow measurement gives a contradictory result, is “not observed in light-ion collisions,” further supporting the interpretation that genuine collective fluid behavior is at work.

Comparisons between the two collision systems also placed new constraints on a subtle parameter: the effective size of the region within a single proton or neutron over which its internal quarks and gluons are spread. The measurements favored a smaller value for this parameter, roughly 0.1 to 0.2 femtometers (one femtometer is one quadrillionth of a meter), consistent with prior suggestions in the literature.

What emerges from this work is something physicists have been building toward for years: a way to use high-energy collisions not just to study the superhot plasma they create, but to extract information about the internal structure of nuclei that is difficult to get any other way. By colliding nuclei whose shapes are well understood theoretically, and comparing two similarly sized systems whose shapes differ in known ways, the team was able to isolate the effect of nuclear geometry with unusual clarity.

Smashing light nuclei together at extreme energies and watching how the wreckage flows is emerging as a powerful new way to probe the internal structure of atomic nuclei, and to confirm that even a tiny droplet of the universe’s earliest matter can behave like a nearly perfect liquid.

---