Physicists with the ATLAS experiment at CERN studied rare collisions of lead nuclei in which the nuclei do not break apart. The data come from lead–lead runs at a center‑of‑mass energy of 5.02 TeV per nucleon pair recorded in 2018, with an integrated luminosity of 1.72 nb⁻¹. In these “0n0n” events no forward neutrons are seen, so both nuclei appear to remain intact after the interaction.

The paper focuses on jets — narrow sprays of particles that come from quarks and gluons — produced by photons emitted from the fast moving nuclei. At large distances between the two nuclei, the strong force interactions are suppressed and the intense electromagnetic fields act like beams of quasi‑real photons. Those photons can scatter in several ways: by hitting the opposite nucleus and exchanging color (photon–nucleus, written γ+A), by a diffractive process where a colorless object called the pomeron is exchanged (photon–pomeron, written γ+IP), or by two photons colliding (photon–photon, written γ+γ). The 0n0n requirement tends to select diffractive and peripheral photon‑nucleus collisions, but it also admits photon–photon events.

To separate these different physics processes, the researchers used the pattern of particle production in rapidity — a coordinate related to angle and energy — and in particular the size of rapidity gaps, which are regions with no detected particles. They performed a template fit to the distribution of the minimum rapidity gap in each event. Jets were reconstructed with the standard anti‑kt algorithm using a radius parameter R = 0.4. With this statistical separation they extracted, for the first time in nuclear collisions at the LHC, cross sections for γ+IP→jets and measured how often electromagnetic dissociation (EMD) adds neutrons that would spoil the 0n0n signature.

The measurements matter because photon‑induced jet production probes the internal distribution of quarks and gluons in nuclei. Diffractive jet production (γ+IP) is especially interesting: it is sensitive to a colorless part of the interaction that tends to leave the nucleus intact and can reveal information not accessible in ordinary collisions. The study also compared the rate at which electromagnetic interactions add neutrons in events that would otherwise look like 0n0n to the analogous rate when one side already emits neutrons. The comparisons support the idea that the 0n0n selection favors more peripheral photon‑nucleus collisions, where the photon strikes the outer edge of the nucleus.