Forward jets and forward W -boson production at hadron colliders

In this talk we give a short review of forward jets and forward W -boson production at hadron colliders, in view of the extraction of footprints of BFKL physics. We argue that at Tevatron energies, dijet production at large rapidity intervals is still subasymptotic with respect to the BFKL regime, thus the cross section is strongly dependent on the various cuts applied in the experimental setup. In addition, the choice of equal transverse momentum cuts on the tagging jets makes the cross section dependent on large logarithms of non-BFKL origin, and thus may spoil the BFKL analysis. For vector boson production in association with two jets, we argue that the configurations that are kinematically favoured tend to have the vector boson forward in rapidity. Thus W + 2 jet production lends itself naturally to extensions to the high-energy limit. Rapporteur at EPS2001, Budapest, Hungary On leave from INFN, Sez. di Genova, Italy In strong-interaction processes characterised by two large and disparate energy scales, which are typically the squared parton center-of-mass energy ŝ and momentum transfer t̂, with ŝ ≫ t̂, the BFKL theory [1] resums the large logarithms ln(ŝ/|t̂|). Over the past years several attempts have been made to predict and detect footprints of emission of BFKL gluon radiation in strong-interaction processes, like in dijet production at hadron colliders at large rapidity intervals, in forward jet production in DIS and in γγ collisions in double-tag events, e e → e e+ hadrons. Here we shall review first dijet production at hadron colliders at large rapidity intervals, and then consider the production of a forward W -boson in association with two jets. 1. Dijet production at large rapidity intervals In dijet production at hadron colliders, at large enough rapidities, the rapidity interval is well approximated by the expression ∆y ≃ ln(ŝ/|t̂|), where ŝ = xaxbS and |t̂| ≃ ka⊥kb⊥, with ka,b⊥ being the moduli of the transverse momenta of the two jets, xa,b the momentum fractions of the incoming partons, and √ S the hadronic centre-of-mass energy. Once the transverse momenta are fixed, there are two ways of increasing ∆y: by increasing the x’s in a fixed energy collider; or viceversa, by fixing the x’s and letting S grow, in a ramping run collider experiment. The former set-up, the only feasible at a collider run at fixed energy, has been proven to be unpractical, since in the dijet production rate dσ/d∆y as a function of ∆y it is difficult to disentangle the BFKL-driven rise of the parton cross section from the steep fall-off of the parton densities [2]. The latter set-up, even though the first to be proposed [3], has been analysed only lately [4], because it required a collider running at different centre-of-mass energies. Here we review first the original Mueller-Navelet proposal [3], and then analyse its implementation. In the high-energy limit, ŝ ≫ |t̂|, any QCD scattering process is dominated by gluon exchange in the crossed channel. This constitutes the leading term of the BFKL resummation. The corresponding QCD amplitude factorizes into an effective amplitude formed by two scattering centres, the impact factors, connected by the gluon exchanged in the crossed channel. The BFKL equation then resums the leading logarithmic (LL) corrections, of O(αn S ln(ŝ/|t̂|)), to the gluon exchange in the crossed channel. In dijet production at large rapidity intervals, one can write the cross section in the following factorized form [3]