Charmonia, bound states of charm (c) and anti-charm (c) quarks, are probes for the formation of hot quark–gluon plasma (QGP) in heavy-ion collisions. The suppression of charmonium, already observed at the lower energies of CERN’s Super Proton Synchrotron (SPS) and the Relativistic Heavy Ion Collider at Brookhaven, has been attributed to the screening of the cc binding by the high density of colour charges present in the QGP. However, the modification of charmonium production in heavy-ion collisions can be induced not only by a hot deconfined medium, but also by effects of cold nuclear matter (CNM). The latter can be studied in proton–nucleus interactions, where the temperature and energy density necessary for QGP formation are not expected to be reached.

CNM affects the cc pair throughout its time evolution, from a pre-resonant state to the fully formed resonance, and it can be investigated by comparing the behaviour of the tightly bound J/ψ and the weakly bound ψ(2S) charmonium states. Effects present in the early stages of the cc evolution – such as nuclear-parton shadowing and initial-state energy loss – do not depend on the final charmonium quantum numbers, and should have similar effects on the J/ψ and ψ(2S). On the other hand, final-state mechanisms, such as the break-up of the bound state via interactions with nucleons or with the hadronic matter produced in the collision, will be sensitive to the binding energy of the resonance, and should have a stronger effect on the ψ(2S) than on the J/ψ.

ALICE has studied the production of J/ψ and ψ(2S) in proton–lead collisions at √s = 5.02 TeV, in both the proton-going direction (rapidity 2.03 < ycms < 3.53) and the lead-going direction (–4.46 < ycms < –2.96). The modification of the production yields induced by CNM, with respect to the corresponding proton–proton yield scaled by the number of nucleon–nucleon collisions, is quantified through the nuclear modification factor RpA, which is shown in the figure for J/ψ and ψ(2S). The ψ(2S) suppression is large, and stronger than for the J/ψ, in particular in the backward rapidity region, where the J/ψ is not suppressed at all. This observation implies that final-state effects play an important role, as initial-state mechanisms alone (see also the theory predictions in the figure relative to a pure initial-state scenario) would lead to the same behaviour for both charmonium states.

Such a result was also observed at lower energies (at the SPS, Fermilab and HERA at DESY), where it was related to break-up effects by the nucleons in the nucleus. However, at LHC energies, the resonance formation time (around 0.1 fm/c) is significantly smaller than the time spent by the cc pair in the nucleus, implying that CNM cannot affect the final-state charmonia. This suggests that the difference between the J/ψ and ψ(2S) suppression is due to the interaction with hadrons produced in the proton–lead collision. A detailed study of this effect, still in progress on the theory side, is expected to provide quantitative information on the density and characteristics of such a hadronic medium.