Résumé

Le LHC dépasse la luminosité nominale pour les ions lourds

Le LHC a terminé l’année 2015 par une exploitation avec des ions lourds. Pour la première fois, la moyenne de l’énergie dans le centre de masse obtenue lors des collisions de noyaux de plomb a atteint 5,02 TeV par paire de nucléons. Cette énergie, inédite pour les collisions plomb-plomb, est presque deux fois supérieure à celle de l’exploitation plomb-plomb précédente, en 2011, et représente quelque 25 fois celle atteinte au RHIC, à Brookhaven ; cette prouesse permet d’élargir l’étude du plasma quarks-gluons à des densités et des températures encore plus élevées.

The extensive modifications made to the LHC during its first long shutdown allowed the energy of the proton beams to be increased from 4 TeV in 2012 to 6.5 TeV, enabling proton–proton collisions at a centre-of-mass energy of 13 TeV, in 2015. As usual, a one-month heavy-ion run was scheduled at the end of the year. With lead nuclei colliding, the same fields in the LHC’s bending magnets would have allowed 5.13 TeV per colliding nucleon pair. However, it was decided to forego the last whisker of this increase to match the equivalent energy of the proton–lead collisions that took place in 2013, namely 5.02 TeV. Furthermore, the first week of the run was devoted to colliding protons at 2.51 TeV per beam. This will allow the LHC experiments to make precise comparisons of three different combinations of colliding particles, p–p, p–Pb and Pb–Pb, at the same effective energy of 5.02 TeV. This is crucial to disentangling the ascending complexity of the observed phenomena (CERN Courier March 2014 p17).

The first (and last, until 2018) Pb–Pb operation close to the full energy of the LHC was also the opportunity to finally assess some of its ultimate performance limits as a heavy-ion collider. A carefully targeted set of accelerator-physics studies also had to be scheduled within the tight time frame.

Delivering luminosity

The chain of specialised heavy-ion injectors, comprising the electron cyclotron resonance ion source, Linac3 and the LEIR ring, with its elaborate bunch-forming and cooling, were recommissioned to provide intense and dense lead bunches in the weeks preceding the run. Through a series of elaborate RF gymnastics, the PS and SPS assemble these into 24-bunch trains for injection into the LHC. The beam intensity delivered by the injectors is a crucial determinant of the luminosity of the collider.

Planning for the recommissioning of the LHC to run in two different operational conditions after the November technical stop resembled a temporal jigsaw puzzle, with alternating phases of proton and heavy-ion set-up (the latter using proton beams at first) continually readapted to the manifold constraints imposed by other activities in the injector complex, the strictures of machine protection, and the unexpected. For Pb–Pb operation, a new heavy-ion magnetic cycle was implemented in the LHC, including a squeeze to β* = 0.8 m, together with manipulations of the crossing angle and interaction-point position at the ALICE experiment. First test collisions occurred early in the morning of 17 November, some 10 hours after first injection of lead.

The new Pb–Pb energy was almost twice that of the previous Pb–Pb run in 2011, and some 25 times that of RHIC at Brookhaven, extending the study of the quark–gluon plasma to still-higher energy density and temperature. Although the energy per colliding nucleon pair characterises the physical processes, it is worth noting that the total energy packed into a volume on the few-fm scale exceeded 1 PeV for the first time in the laboratory.

After the successful collection of the required number of p–p reference collisions, the Pb–Pb configuration was validated through an extensive series of aperture measurements and collimation-loss maps. Only then could "stable beams" for physics be declared at 10.59 a.m. on 25 November, and spectacular event displays started to flow from the experiments.

In the next few days, the number of colliding bunches in each beam was stepped up to the anticipated value of 426 and the intensity delivered by the injectors was boosted to its highest-ever values. The LHC passed a historic milestone by exceeding the luminosity of 1027 cm–2 s–1, the value advertised in its official design report in 2004.

This allowed the ALICE experiment to run in its long-awaited saturated mode with the luminosity levelled at this value for the first few hours of each fill.

Soon afterwards, an unexpected bonus came from the SPS injection team, who pulled off the feat of shortening the rise time of the SPS injection kicker array, first to 175 ns then to 150 ns, allowing 474, then 518, bunches to be stored in the LHC. The ATLAS and CMS experiments were able to benefit from luminosities over three times the design value. A small fraction of the luminosity in this run was delivered to the LHCb experiment, a newcomer to Pb–Pb collisions.

Nuclear beam physics

The electromagnetic fields surrounding highly charged ultrarelativistic nuclei are strongly Lorentz-contracted into a flat "pancake". According to the original insight of Fermi, Weizsäcker and Williams, these fields can be represented as a flash of quasi-real photons. At LHC energies, their spectrum extends up to hundreds of GeV. In a very real sense, the LHC is a photon–photon and photon–nucleus collider (CERN Courier November 2012 p9). The study of such ultraperipheral (or "near-miss") interactions, in which the two nuclei do not overlap, is an important subfield of the LHC experimental programme, alongside its main focus on the study of truly nuclear collisions.

From the point of view of accelerator physics, the ultraperipheral interactions with their much higher cross-sections loom still larger in importance. They dominate the luminosity "burn-off", or rate at which particles are removed from colliding beams, leading to short beam and luminosity lifetimes. Furthermore, they do so in a way that is qualitatively different from the spray of a few watts of "luminosity debris" by hadronic interactions. Rather, the removed nuclei are slightly modified in charge and/or mass, and emerge as new, well-focussed, secondary beams. These travel along the interaction region just like the main beam but, as soon as they encounter the bending magnets of the dispersion-suppressor section, their trajectories deviate, as in a spectrometer.

The largest contribution to the burn-off cross-section comes from the so-called bound-free pair-production (BFPP) in which the colliding photons create electron–positron pairs with the electron in a bound-state of one nucleus. A beam of these one-electron ions, carrying a power of some tens of watts, emerges from the interaction point and is eventually lost on the outer side of the beam pipe.

Controlled quench

The LHC operators have become used to holding their breath as the BFPP loss peaks on the beam-loss monitors rise towards the threshold for dumping the beams (figure). There has long been a concern that the energy deposited into superconducting magnet coils may cause them to quench, bringing the run to an immediate halt and imposing a limit on luminosity. In line with recent re-evaluations of the magnet-quench limits, this did not happen during physics operation in 2015 but may happen in future operation at still-higher luminosity. During this run, mitigation strategies to move the losses out of the magnets were successfully implemented. Later, in a special experiment, one of these bumps was removed and the luminosity slowly increased. This led to the first controlled steady-state quench of an LHC dipole magnet with beam, providing long-sought data on their propensity to quench. On the last night of the run, another magnet quench was deliberately induced by exciting the beam to create losses on the primary collimators.

Photonuclear interactions also occur at comparable rates in the collisions and in the interactions with the graphite of the LHC collimator jaws. Nuclei of 207Pb, created by the electromagnetic dissociation of a neutron from the original 208Pb at the primary collimators, were identified as a source of background after traversing more than a quarter of the ring to the tertiary collimators near ALICE.

These, and other phenomena peculiar to heavy-ion operation, must be tackled in the quest for still-higher performance in future years.