I discuss old and recent aspects of QCD jet-emission and describe how hard QCD results are used to construct Monte Carlo programs for generating hadron emission in hard collisions. I focus on the program HERWIG at LHC.
The statusLHC is a discovery machine, it is expected to tell us how to complete the unified theory of elementary interactions. New (heavy) particles are searched to indicate/confirm new symmetries. Events with heavy particles are expected to be accompanied by an intense emission of hadrons at short distances, and this is the domain of perturbative QCD. Therefore, to identify and understand non-standard events a quantitative knowledge of the characteristics of the hard radiation is strongly needed. In 1973 QCD was at the frontier of particle physics (discovery of asymptotic freedom [1] and beginning of quantitative QCD studies), now in 2007 QCD is at the center of particle studies. The Monte Carlo programs for jet emissions [2][3][4] are important instruments for analyzing standard and non-standard short distance events. They are the Summa of most QCD theoretical results and many present studies aim to improve their quantitative predictions. Thanks to the QCD factorization structure [5], Monte Carlo programs can be interfaced with hard cross sections involving also non-QCD processes (electroweak, supersymmetric, extra dimension, black holes, ...). In this way, Monte Carlo generators can describe both QCD and non-QCD events at short distances.In this paper I describe the main QCD results which enter the construction of a Monte Carlo generator. They are so many that most of the key points will be recalled in a schematic way, but I hope that this short description could provide an idea of the reliability range of the Monte Carlo generators. For a more detailed description see [6]. Here, aiming to be simple and synthetic, I follow a personal point of view and the focus will be on the Monte Carlo event generator HERWIG [2]. Its general structure is similar to other important Monte Carlo generators [3,4]. In section 2 I present the scheme of the operations performed by Monte Carlo codes for LHC. The fact that the generation of events can be subdivided into successive stages is physically based on QCD factorization properties. The theoretical basis are discusses/recalled in section 3. In section 4 I discuss 1