Credit: © CERN

CERN's Large Hadron Collider (LHC) will fire up again this month, with its three detectors, ATLAS, CMS and LHCb, poised to begin close monitoring of the proton–proton collisions in pursuit of such novelties as supersymmetry and the Higgs boson. Also part of the programme are 'heavy ion' collisions — using the LHC to collide lead ions instead of protons, in the hope of creating quark–gluon plasma — for which a dedicated detector called ALICE has been built.

Heavy-ion physics poses particular challenges, owing to the terrific density of quarks and gluons involved in each collision. ALICE will have to cope with a multitude of tracks from the particles produced, as pictured in this simulation. For theorists, modelling the process is also tricky, but Korinna Zapp and colleagues have a proposal that may help (Phys. Rev. Lett. 103, 152302; 2009).

Quarks and gluons (known collectively as partons) emerging from any type of collision may radiate a gluon. This can happen repeatedly, to create what is known as a parton shower; eventually the partons will group together (or 'hadronize') to form the composite particles that are observed in detectors. The process can be quite effectively modelled when the colliding particles are electrons or protons, but in the heavy-ion case the parton shower develops in a 'dense QCD-matter' environment, packed with other quarks and gluons that may also be radiating.

Zapp et al. have devised an algorithm to take account of the non-Abelian Landau–Pomeranchuk–Migdal effect — or rather, the quantum interference between spatially separated incidences of gluon radiation — which occurs in such an environment. They define a 'gluon formation time', based on the gluon energy and the transverse momentum of the gluon radiated previously. The Landau–Pomeranchuk–Migdal effect is then accounted for by requiring that, for a gluon radiated within the formation time, the momentum transfer is coherent; and for a gluon radiated later than the formation time, it is incoherent.

The algorithm can be incorporated straightforwardly in so-called Monte Carlo simulations of heavy-ion collisions, ahead of the first heavy-ion run at the LHC, scheduled for late 2010.