Physicists working with the Large Hadron Collider have found a way to study the strong nuclear force, the binding agent inside every atomic nucleus, by analyzing collisions that almost did not happen. When lead ions racing through the LHC pass close enough to interact but not close enough to smash apart, virtual photons from one nucleus strike the other, producing rare particles in remarkably clean conditions. Two new analyses from LHC Run 3, one from the ATLAS experiment and one from CMS, now show that these ultraperipheral collisions can reveal how quarks and gluons behave at scales and energies that head-on collisions obscure with debris.
What Ultraperipheral Collisions Actually Measure
Standard particle collisions at the LHC are violent events. Two beams of lead ions crash together, and the resulting fireball produces thousands of fragments that detectors must sort through. Ultraperipheral collisions, or UPCs, work differently. The ions pass at distances larger than the sum of their nuclear radii, so they never physically overlap. Instead, the intense electromagnetic field surrounding each ion generates virtual photons that can interact with the opposing nucleus. Because no nuclear breakup contaminates the signal, the resulting particle production is far easier to interpret.
That cleanliness matters because it lets physicists isolate specific processes tied to quantum chromodynamics, or QCD, the theory governing the strong force. In a direct collision, strong-force effects are tangled with electromagnetic and weak-force contributions. In a UPC, the photon acts as a known probe, and everything interesting comes from how the target nucleus responds. An MIT-led analysis of earlier LHC data emphasized that these photon-nucleus interactions can uncover new properties of matter and sharpen our understanding of the forces that hold nuclei together.
UPCs effectively turn the LHC into a photon–nucleus collider. Each fast-moving lead ion carries an enormous electromagnetic field; in quantum terms, that field can be treated as a swarm of photons. When one of those photons hits the other nucleus, it can briefly fluctuate into a quark-antiquark pair, interact via gluons, and then emerge as a bound state such as the J/psi or as open heavy-flavor particles like D mesons. Because the rest of the nucleus remains intact, the event leaves a characteristic pattern: very low additional particle activity, large empty regions in rapidity, and often intact ions that can be tagged in dedicated detectors.
ATLAS Maps J/psi Production at Record Energy
The ATLAS Collaboration has now pushed this technique to record energies with a new preprint, labeled CERN-EP-2025-180, that measures coherent exclusive J/psi meson production in ultraperipheral lead–lead collisions at a center-of-mass energy of 5.36 TeV per nucleon pair. The J/psi is a bound state of a charm quark and its antimatter partner, and “coherent exclusive” means the entire nucleus acted together to produce it, with no other particles created in the event. That coherence is the key signal: it implies the photon probed the whole nuclear gluon field at once rather than scattering from a single nucleon.
The 5.36 TeV collision energy corresponds to the LHC’s Run 3 operating conditions and is the highest yet used for this kind of study. At such energies, the photon can resolve gluons that carry extremely small fractions of the nucleus’s total momentum, a regime known as small-x. In this domain, gluon densities are predicted to grow rapidly, potentially triggering nonlinear QCD effects that slow further growth, a phenomenon often called gluon saturation. The ATLAS preprint, which supersedes an earlier conference note, incorporates full systematic uncertainties and compares the measured cross sections with several theoretical models that implement different saturation scenarios.
These model comparisons turn the measurement into a precision test of nuclear structure. If a model without saturation overshoots the data while a model with saturation agrees, that pattern would indicate that nonlinear gluon recombination is already important inside lead nuclei at LHC energies. Conversely, if saturation-based predictions undershoot the measurement, theorists would need to revisit where and how these effects set in. By covering a broad range of J/psi rapidities, ATLAS effectively scans different x values in the nucleus, mapping how the gluon distribution evolves toward the small-x frontier.
CMS Tracks Charm Quarks Through a Different Channel
While ATLAS focused on hidden charm in the form of J/psi mesons, the CMS Collaboration chose a complementary observable. Their study, presented in the arXiv preprint 2509.08626, measures D0 meson photoproduction in ultraperipheral heavy-ion collisions at the same 5.36 TeV energy. The D0 contains a charm quark paired with an up antiquark, and CMS reconstructs it via its characteristic decay into a charged kaon and a pion. This decay leaves a clean signature in the detector, allowing precise reconstruction of the D0 momentum.
To ensure they were truly studying photonuclear events rather than ordinary hadronic collisions, the CMS team applied nuclear-breakup tagging and rapidity-gap requirements. Nuclear-breakup tagging looks for neutrons emitted when one of the ions is slightly excited by the photon interaction, while the other remains intact. Rapidity gaps (extended regions in the detector with no produced particles) signal the absence of a messy hadronic overlap. Together, these criteria strongly suppress backgrounds from central heavy-ion collisions where the nuclei shatter and produce a quark–gluon plasma.
Charm quarks are heavy enough that they can only be created in very energetic interactions, making them incisive probes of the gluon content inside nuclei. The CMS paper reports D0 production cross sections as functions of both transverse momentum and rapidity, yielding a two-dimensional map of how often and at what angles these mesons emerge from near-miss events. Because D0 mesons contain “open” charm rather than a charm–anticharm bound state, they are sensitive to somewhat different aspects of the underlying gluon distributions than the J/psi data.
By comparing their measurements with perturbative QCD calculations that incorporate various nuclear parton distribution functions, CMS can test how well existing parametrizations describe charm production in photon–nucleus interactions. Discrepancies between theory and data, especially at forward rapidities where small-x gluons dominate, would point to missing physics in current descriptions of the nuclear wave function.
Why Near-Misses Beat Head-On Crashes for This Physics
The value of UPC measurements lies not only in their cleanliness but also in their ability to preserve initial-state information. In a head-on heavy-ion collision, the quark–gluon plasma that forms is so hot and dense that charm quarks and quarkonia produced inside it undergo multiple interactions before escaping. Those final-state effects (energy loss, recombination, and screening) blur the link between what is observed and the original gluon distributions.
In a UPC, by contrast, no quark–gluon plasma forms. The photon interacts with the nucleus, creates a charm quark pair or a J/psi, and the products fly out through essentially empty space. That environment makes UPCs ideal for disentangling the initial gluon structure from the complicated medium effects that dominate central collisions. As one researcher put it, a collision between a photon and a nucleus can act like a high-precision microscope, offering an unusually sharp picture of the interaction.
The new ATLAS and CMS results extend a line of work in which physicists have used near-miss events to probe fundamental forces. Earlier studies showed that near-miss collisions can reveal subtle properties of the strong force that are otherwise washed out in busier environments. The latest measurements deepen that program at higher energies and with more sophisticated analysis techniques, turning UPCs into a precision tool rather than a curiosity at the edges of the detector.
Toward a Tomographic Picture of Nuclei
Taken together, the J/psi and D0 measurements begin to outline a kind of tomography of the lead nucleus. Coherent J/psi production is especially sensitive to the spatial distribution of gluons: the coherence condition effectively samples the whole nuclear disk, tying the cross section to the transverse profile of the gluon field. Open-charm production, by contrast, is more directly linked to the momentum distribution of gluons at a given x. By combining both, physicists can start to disentangle where within the nucleus different gluons reside and how their densities evolve with energy.
These studies also serve as pathfinders for future facilities. Concepts for next-generation colliders often include dedicated electron–ion or photon–ion modes, inspired in part by the success of using the LHC’s heavy-ion beams as intense photon sources. The demonstration that UPCs can deliver detailed, model-discriminating data on small-x gluons strengthens the case for machines explicitly designed to map nuclear structure with light. As an earlier overview of this strategy noted, near misses can sometimes illuminate the strong force more clearly than the most energetic crashes.
For now, the ATLAS and CMS Run 3 analyses underscore how much physics hides in events that almost did not happen. By exploiting the LHC’s heavy-ion beams as both targets and photon sources, and by focusing on rare but exceptionally clean signatures, researchers are turning ultraperipheral collisions into a powerful lens on the gluons that bind nuclei together. As more data arrive and theoretical models sharpen, these near-miss measurements are poised to play a central role in charting the strong force at its most extreme scales.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
