The Flow Is Not Uniform express
When a proton slams into a lead nucleus at the Large Hadron Collider, a swarm of particles is born. They fly off in all directions, but not entirely randomly: their motion resembles ripples on water from a thrown stone. Physicists call this orderly dispersal a "flow."
In the new study, scientists tested how stable this flow is. It turns out that from collision to collision it varies slightly, like ripples on water depend on exactly how the stone fell. Previously, such fluctuations were only seen in collisions of two large lead nuclei, but now they have also been observed in the small system of "proton + lead nucleus."
To ensure the effect was not due to background noise (for example, from individual particle jets), physicists applied a special data-cleaning method. The background could not explain the observed fluctuations. So the flow is indeed not the same in every event.
This discovery helps better understand how coordinated motion emerges from many individual particles. Such studies at the Large Hadron Collider provide clues about the earliest stage of collisions, when matter behaves like a liquid.
Original abstract
as written by the authors arXiv:2602.10645
Measurements of transverse momentum ($$p_{\rm T}$$) and pseudorapidity ($$\eta$$) dependent flow vector fluctuations in p$$-$$Pb collisions at $$\sqrt{s_{_{\rm NN}}} = 5.02$$ TeV at the CERN Large Hadron Collider are presented. By studying long-range two-particle correlations with a template fit method, potential biases from non-flow effects such as jets and resonance decays are effectively suppressed. Significant $$p_{\rm T}$$- and $$\eta$$-dependent fluctuations of the second-harmonic flow vector are observed with more than 5$$\sigma$$ confidence in p$$-$$Pb collisions, similar to the observations in Pb$$-$$Pb collisions. The influence of residual non-flow effects has been evaluated and cannot account for the observed fluctuations, thereby confirming the observation of flow vector fluctuations in small collision systems at the LHC. Comparisons to model calculations from 3DGlauber+MUSIC+UrQMD and the parton transport model from AMPT are also presented. The measurements provide constraints on the theoretical modelling of the three-dimensional initial geometry and its event-by-event fluctuations, offering critical insights into the origin of collective flow in small collision systems at the LHC.
In a proton–lead collision, several thousand particles are produced, but their overall motion can be described by just a few numbers, like the shape of a wave.