No abstract

We consider within QCD collinear factorization the process p + p → jet + jet + X, where two forward high-p T jets are produced with a large separation in rapidity ∆y (Mueller-Navelet jets). In this case the (calculable) hard part of the reaction receives large higher-order corrections ∼ α n s (∆y) n , which can be accounted for in the BFKL approach with next-to-leading logarithmic accuracy, including contributions ∼ α n s (∆y) n−1 . We calculate several observables related with this process, using the next-to-leading order jet vertices, recently calculated in the approximation of small aperture of the jet cone in the pseudorapidity-azimuthal angle plane. †

The Balitsky-Fadin-Kuraev-Lipatov (BFKL) approach for the investigation of semihard processes is plagued by large next-to-leading corrections, both in the kernel of the universal BFKL Green's function and in the process-dependent impact factors, as well as by large uncertainties in the renormalization scale setting. All that calls for an optimization procedure of the perturbative series. In this respect, one of the most common methods is the Brodsky-Lepage-Mackenzie (BLM) one, which eliminates the renormalization scale ambiguity by absorbing the nonconformal β 0 terms into the running coupling. In this paper, we apply the BLM scale setting procedure directly to the amplitudes (cross sections) of several semihard processes. We show that, due to the presence of β 0 terms in the next-to-leading expressions for the impact factors, the optimal renormalization scale is not universal but depends both on the energy and on the type of process in question.

We investigate the stability under variation of the renormalization, factorization and energy scales entering the calculation of the cross section, at next-to-leading order in the BFKL formalism, for the production of Mueller-Navelet jets at the Large Hadron Collider, following the experimental cuts on the tagged jets. To find optimal values for the scales involved in this observable it is possible to look for regions of minimal sensitivity to their variation. We show that the scales found with this logic are more natural, in the sense of being more similar to the squared transverse momenta of the tagged jets, when the BFKL kernel is improved with a resummation of collinear contributions than when the treatment is at a purely next-to-leading order. We also discuss the good perturbative convergence of the ratios of azimuthal angle correlations, which are quite insensitive to collinear resummations and well described by the original BFKL framework.