Forward Drell-Yan plus backward jet as a test of BFKL evolution Martin Hentschinski1, Clara Salas2 1Physics Department, Brookhaven National Laboratory, Upton, 11973, USA 2Instituto de F´ısica Te´orica UAM/CSIC, C/ Nicol´as Cabrera 13-15, Universidad Aut´onoma de Madrid, Cantoblanco, E-28049 Madrid, Spain 3 1 0 DOI: will be assigned 2 n We study Drell-Yan plus jet events where the gauge boson is produced in the forward a direction of one of the colliding protons and a jet is produced in the forward direction of J the second proton. The resulting large rapidity difference between the final states then 7 opens up the phase space for BFKL evolution. First numerical results on partonic level are provided. ] h p - 1 Introduction p e h Due to its large center of mass energy the LHC allows for the study of forward physics using [ methods of perturbative QCD. Among them we find forward production of different systems 1 such as high p jets [1], heavy quark pairs [2] and Drell-Yan (DY) processes where a virtual T v photon or Z boson decays into a pair of leptons [3,4]. The study of these type of processes is 7 interesting as they allow to probe parton distribution functions at very small values of x which 2 have not been reached in so-far collider experiments. They therefore provide a possibility to 2 1 test formalisms which have been especially developed for the description of small x processes . and which go beyond the standard formulation in terms of collinear factorization by including 1 additionalsmallxenhancedcontributions. ThestartingpointofsuchstudiesisgivenbyBFKL 0 3 evolution which resums small x logarithms on the level of partonic scattering amplitudes at 1 leading logarithmic (LL) [5] and next-to-leading-logarithmic (NLL) [6] accuracy. In addition, : since small x evolution ultimatively leads to high parton densities, such processes may further v i allowfortheobservationofsaturationeffectswhichrequireanextensionoftheBFKLformalism, X see e.g. [7]. r While studies of inclusive observables provide strong hints for the presence of BFKL evo- a lution in small x data, see e.g. [9], a proper identification of relevant effects at small x is hard to achieve. Cancellations between different final states minimize the sensitivity to the partic- ular feature of the employed method and deviations from inclusive evolution equations due to smallxeffectsmaybepartlyhiddenintothechoseninitialconditions. Itisthereforenecessary to turn to the study of more exclusive observables to distinguish different effects at small x. Among these exclusive observables there is a class of events where the entire dependence of the processonthenon-perturbativedynamicsistreatedwithinconventionalcollinearfactorization. DIS 2012 1 Theseobservablestypicallyinvolvehardeventsintheforwardregionofbothscatteringprotons, while the large difference in rapidity between the hard final states opens up the phase space for BFKL evolution. Among the best explored processes of this type are ‘Mueller-Navelet’ jets which consist of a high p -jets in the forward regions of each proton. Currently this process is T oneofthefewexampleswhereacompleteNLLBFKLdescriptionexists[8]. Incontrasttona¨ıve expectations, theresultrevealsastrongdependenceonthenext-to-leadingordercorrectionsto thejetimpactfactors. Atthesametime,thenumericaldifferencesbetweentheNLLresummed result and a pure collinear NLO result remain rather small for a large class of observables. This observation motivates the study of a new type of forward-backward observable, where a DY pair is produced in the forward direction of one of the particles instead of a jet. The hopeisthatthisobservableisabletobetterdistinguishbetweenstandardNLOresultsandNLL BFKLresummedpredictionsandallowstoidentifyuniversalfeaturesofBFKLevolution. Even thoughthelargevirtualityofthephotonand/orthemassoftheZ diminishesatfirstthevalue ofthestrongcouplingconstantα ,therapiditydifferencebetweenleptonpairandbackwardjet s remains large at the Large Hadron Collider, ∆Y < 7 and a study of BFKL evolution appears meaningful. Inaddition,studyofnewfinalstatesmayalsotriggernewtheoreticaleffortsforan improved definition of impact factors and lead to the identification of new BFKL observables. In the following we present some partial results of our study, which are currently restricted to the partonic level. For details we refer to our paper in preparation [10]. 2 The leading-order DY impact factor = + (a) (b) Figure 1: a) A large difference in rapidity between the forward gauge boson (γ∗,Z) and the backward jet opens up the phase space for BFKL evolution. b) The leading order DY impact factorisobtainedasthesumoftwoeffectivediagramswherethet-channelgluoncarrieseikonal polarizations. In the current study we restrict to the LO impact factor, where relevant diagrams can be found in Fig. 1.b. A complete NLO study seems possible using Lipatov’s effective action [11] 2 DIS 2012 which is currently explored at NLO [12]. The leading order impact factor reads c α (cid:112)N2−1(cid:34)zk2(cid:0)(1−z)2+1(cid:1)+2M2(1−z)z M2z(1−z) M2(1−z)z(cid:35) Φ = f s c − − Zq πk2N D D D2 D2 c 1 2 1 2 D =(q−zk)2+(1−z)M2 D =q2+(1−z)M2 (1) 1 2 Here M denotes the mass of the Z boson and the virtuality of the photon respectively, while c yields the coupling of the gauge boson to the quark. q and k are the transverse momenta f of the final state gauge boson and initial gluon, while z is the momentum fraction of the initial quark momentum carried on by the gauge boson. 3 Preliminary numerical results at partonic level Theaboveimpactfactorcarriesalogarithmicsingularityifthefinalstatequarkturnstobesoft. After convolution of the impact factor with the BFKL Green’s function, this corresponds to the limit z →1. To avoid this singularity we study ratios of angular coefficients C =(cid:104)cosnφ(cid:105), n where φ denotes the azimuthal angle between the jet and the gauge boson. Inapreliminarystudy,seeFig.2,whichrestrictstothepartoniclevelandconformalpartof the NLO BFKL kernel, we use a fixed value z =0.9 while the QCD coupling is taken at the Z- scale. Asinthejet-jetcasewefindthatthebestconvergenceoftheBFKLpredictionisachieved for ratios which do not contain the angular coefficient C . A complete study at hadronic level 0 facesadditionalcomplications. Eventhoughformallyfinite, thelimitz →1, whichisnaturally encountered once the convolution with parton distribution functions is included, leads to large contributions which can even cause negative results for some of the angular coefficients. The study of these effects and their appropiate treatment as well as the inclusion of the running coupling corrections of the NLO BFKL Green’s function is currently in progress. Acknowledgments We are grateful for financial support from the MICINN under grant FPA2010-17747, the Re- search Executive Agency (REA) of the European Union under the Grant Agreement number PITN-GA-2010-264564 (LHCPhenoNet) and Comunidad de Madrid (HEPHACOS ESP-1473). M.H. further acknowledges support from the German Academic Exchange Service (DAAD) and the U.S. Department of Energy under contract number DE-AC02-98CH10886 and a BNL “Laboratory Directed Research and Development” grant (LDRD 12-034). References [1] M.Deak,F.Hautmann,H.JungandK.Kutak,Eur.Phys.J.C72(2012)1982[arXiv:1112.6354 [hep-ph]]. [2] C.Salas,J.Stirling,inpreparation. [3] S.MarzaniandR.D.Ball,Nucl.Phys.B814(2009)246[arXiv:0812.3602[hep-ph]]. [4] F.Hautmann,M.HentschinskiandH.Jung,Nucl.Phys.B865(2012)54[arXiv:1205.1759[hep-ph]], arXiv:1205.6358[hep-ph]. [5] V.S.Fadin,E.A.KuraevandL.N.Lipatov,Phys.Lett.B60(1975)50,I.I.BalitskyandL.N.Lipatov, Sov.J.Nucl.Phys.28(1978)822[Yad.Fiz.28(1978)1597]. DIS 2012 3 0C.71(cid:144)C0 C2(cid:144)C0 0.006 0.6 0.005 0.5 0.004 0.4 0.003 0.3 0.002 0.2 0.001 0.1 (cid:68)y 1 2 3 4 5 6 7 (cid:68)y 1 2 3 4 5 6 7 (a) (b) C2(cid:144)C1 0.009 0.008 0.007 0.006 0.005 0.004 (cid:68)y 1 2 3 4 5 6 7 (c) Figure 2: Ratios of angular coefficients C /C , C /C and C /C versus ∆y = |y −y | on 1 0 2 0 2 1 Z jet partonic level with α (M2) and z =0.9. The transverse momentum of the Z-boson is taken to s Z beq =2GeVwhilejetswithtransversemomentap>20GeVareconsidered. Thedependence on ∆y is described by the LL (purple), NLL (blue) and NLL RG improved (see [13] for details) (red) BFKL Green’s function [6] V.S.FadinandL.N.Lipatov,Phys.Lett.B429(1998)127[hep-ph/9802290],M.Ciafaloniand G.Camici,Phys.Lett.B430(1998)349[hep-ph/9803389]. [7] M.A.Betemps,M.B.GayDucatiandM.V.T.Machado,Phys.Rev.D66(2002)014018 [hep-ph/0111473],F.GelisandJ.Jalilian-Marian,Phys.Rev.D66(2002)094014[hep-ph/0208141], F.GelisandJ.Jalilian-Marian,Phys.Rev.D76(2007)074015[hep-ph/0609066],K.Golec-Biernat, E.LewandowskaandA.M.Stasto,Phys.Rev.D82,094010(2010)[arXiv:1008.2652[hep-ph]]. [8] D.Colferai,F.Schwennsen,L.SzymanowskiandS.Wallon,JHEP1012(2010)026[arXiv:1002.1365 [hep-ph]],F.Caporale,D.Y..Ivanov,B.MurdacaandA.Papa,arXiv:1211.7225[hep-ph]. [9] M.Hentschinski,A.SabioVeraandC.Salas,arXiv:1209.1353[hep-ph]. [10] M.Hentschinski,C.Salas,inpreparation. [11] L.N.Lipatov,Nucl.Phys.B452(1995)369[hep-ph/9502308]. [12] M.HentschinskiandA.S.Vera,Phys.Rev.D85(2012)056006[arXiv:1110.6741[hep-ph] M.Hentschinski,Nucl.Phys.B859(2012)129[arXiv:1112.4509[hep-ph]],G.Chachamis, M.Hentschinski,J.D.MadrigalMartinezandA.SabioVera,Nucl.Phys.B861(2012)133 [arXiv:1202.0649[hep-ph]],arXiv:1211.2050[hep-ph],arXiv:1212.4992[hep-ph]. [13] A.SabioVera,Nucl.Phys.B722(2005)65[hep-ph/0505128],A.SabioVeraandF.Schwennsen,Nucl. Phys.B776(2007)170[hep-ph/0702158[HEP-PH]]. 4 DIS 2012