PProceedings of the PIC 2012, ˇStrbsk´e Pleso, Slovakia
JET PRODUCTION STUDIES AT COLLIDERS
ROBERT HIROSKY
University of VirginiaCharlottesville, VAE-mail: [email protected]
An overview of jet production, measurement techniques, and recent physics resultsfrom colliders is presented. Analyses utilizing jets and boson plus jets final statesare included and implications of the data are discussed. The results presentedhere are a snapshot of those available at the time of the PIC 2012 conference inSeptember 2012.
Hadronic jets are a key signature of strong interactions between constituent par-tons in high energy hadron collisions. The framework of perturbative quantumchromodynamics (pQCD) describes the partonic cross sections [1] for hard scat-tering at large momentum transfers and we have witnessed substantial progressin both experimental and theoretical understanding of such processes throughoutthe past two decades. The production amplitudes for various final states dependon the convolution of parton level cross sections with experimentally-determinedparton distribution functions (PDFs) [2]. Pictorially jet production in hadron col-lisions can be modeled as in Fig. 1. Matrix elements (MEs) for the hard interactionare available to (next-to-)next-to-leading-order, (N)NLO, (next-to-)next-to-leading-logarithm, (N)NLL, for many processes and phenomenological models tuned to datacan be used to account for hadronization effects.
Figure 1. Jet production in hadron collisions. Matrix elements for the hard scatter are available at(N)NLO/(N)NLL for many processes under study. Non-perturbative parton distribution functionsand final state hadronization are factorized from the hard scattering cross section in calculatingproduction rates at colliders. c (cid:13) Institute of Experimental Physics SAS, Koˇsice, Slovakia a r X i v : . [ h e p - e x ] M a y Robert Hirosky
Jet production processes include a wide range of phenomena. In addition topurely partonic final states, a mix of strong, weak, and electromagnetic verticesmay be present. These final states are used to study diverse topics such as, theperformance of pQCD, strong dynamics and couplings, hadron structure and com-positeness scales, hadronic decay modes of short-lived particles, signatures for newmassive states, etc. Furthermore, the precise knowledge of standard model produc-tion rates and kinematic properties in jet final states is a prerequisite for achievinghighest sensitivity in many searches for new physics and rare standard model (SM)processes. This paper reviews the status of recent QCD studies from collider exper-iments at the LHC and Tevatron. Discussions of jets in deep inelastic scattering,photoproduction, and e + e − collisions are included in the contributions by M. Wangand R. Kogler to these proceedings. An overview of the status of modern PDF mod-els is presented in the contribution by Johannes Bl¨umlein. Although hadronic jets arise from the hadronization of elementary quarks and glu-ons and are highly correlated with the four-vectors of their parent particle, it isimportant to note that, unlike elementary particles, jets are ultimately defined bythe algorithm used in their reconstruction. Therefore, different algorithms will ingeneral identify a different set of jets in the same data. A good jet algorithm willsatisfy the following properties: (1) deliver consistent results when applied to par-tons, particles from hadronization, or to detector-level information such as tracks orenergy clusters; (2) be relatively stable with respect to detector noise and additionalenergy deposits from hadron remnants, coincidental soft collisions, and soft partonradiation; (3) have good energy and angular resolution and be relatively straight-forward to calibrate. Typical approaches to jet reconstruction employ cone-basedor recombinant algorithms. An extensive review of modern jet algorithms can befound in Ref. [3] and references therein.In cone-based algorithms fixed cones with verticies set to the primary interactionpoint are used to select objects to form the jet. Typically the angular position ofthe cone is iterated until its geometric center (in azimuthal angle φ and rapidity y )matches the location of the combined four-vector or weighted average position ofthe enclosed objects (energy deposits, tracks, particle four-vectors, etc). The conealgorithms primarily used at the Tevatron are described in Refs. [4] and [5] for theD0 and CDF Experiments, respectively.Recombination or sequential clustering algorithms successively merge objectsbased on spatial and/or relative transverse momenta criteria. Clusters are thensuccessively merged until satisfying some well defined stopping criteria and thejet four-vector is calculated by summing all combined objects. Examples are the k T [6] algorithms and also the anti- k T algorithm [7] which is frequently employedin analyses from the LHC.While various experiments and individual analyses can employ different algo-rithms or choices of parameters to control their performance, each measurement isnecessarily compared to theory or Monte Carlo (MC) using consistent algorithmdefinitions. ets@Colliders Measures of inclusive jet production are sensitive to a combination of QCD matrixelements describing the hard parton scattering and initial state parton distributionfunctions (PDFs). Inclusive measures allow tests of perturbative QCD over wideranges of parton momentum exchange ( Q ) and their sensitivity to wide rangesof parton momentum fractions ( x ) can provide further constraints to PDF models.Furthermore, because the strong coupling ( α s ) depends on the momentum exchangescale, these measurements provide information about the running of the couplingto our largest accessible momentum scales.Measurements of the inclusive jet cross section plotted differentially in bins ofjet transverse momentum ( p T ) and rapidity ( y ) are shown in Figs. 2(a) and 2(b)for the D0 [8] and CMS [9] Experiments, respectively. These results illustrate theremarkable success of perturbative QCD and phenomenological understanding ofproton structure in high energy collisions, spanning 8–9 orders in jet p T . In eachcase experimental results are compared to theory calculated to next-to-leading-order in QCD. (GeV) T p50 60 100 200 300 400 d y ( pb / G e V ) T / dp σ d -6 -5 -4 -3 -2 -1
10 |y|<0.4 (x32)0.4<|y|<0.8 (x16)0.8<|y|<1.2 (x8)1.2<|y|<1.6 (x4)1.6<|y|<2.0 (x2)2.0<|y|<2.4s = 1.96 TeV= 0.7 cone
R NLO pQCD +non-perturbative corrections
CTEQ6.5M 600DØ, 0.70 fb -1 μ R = μ F = p T (a) Inclusive jet cross section measurements bythe D0 Experiment as a function of jet p T insix | y | bins. (GeV) T Jet p
200 300 1000 2000 d | y | ( pb / G e V ) T / dp s d -5 -1 ) · |y| < 0.5 ( ) · · · · CMS = 7 TeVs -1 L = 5.0 fb R = 0.7 T anti-k T = p F m = R m NP Corr. ˜ NNPDF2.1 (b) Inclusive jet cross section measurements bythe CMS Experiment as a function of jet p T infive | y | bins.Figure 2. Measures of the inclusive jet cross section at the Tevatron and LHC. Measurements of the dijet mass spectra in LHC collisions at 7 TeV presentedby the ATLAS [10] and CMS [9] Experiments are shown in Fig. 3. Preliminaryresults at 8 TeV are in agreement with these [11]. Again, detailed comparisons topredictions from pQCD show good agreement between data and theory.Examples of the relative precision of the inclusive jet measurements are shownin Figs. 4(a) and 4(b). In these results, illustrating D0 and CMS measurements,the experimental uncertainties are similar to or significantly smaller than thoseassociated with PDF models, showing that the data can provide significant con-
Robert Hirosky [TeV] m -1 × * [ pb / T e V ] y d m / d σ d -3 -1 ) × * < 0.5 (y ) × * < 1.0 (y ≤ × * < 1.5 (y ≤ × * < 2.0 (y ≤ × * < 2.5 (y ≤ × )) y* exp(0.3 T p = µ (CT10, NLOJET++ ATLAS
Preliminary = 0.6 R jets, t anti-k -1 dt = 4.8 fb L ∫ = 7 TeV, s (a) ATLAS measurement of dijet double-differential cross section plotted versus invari-ant mass of the leading p T jets. Measurementsare binned in terms of the average rapidity sep-aration between the jets y ∗ = | y − y | / (GeV) jj M
200 1000 2000 ( pb / G e V ) m a x d | y | jj / d M s d -6 -3
10 1 ) · < 0.5 ( max |y| ) · < 1.0 ( max · < 1.5 ( max · < 2.0 ( max · < 2.5 ( max CMS = 7 TeVs -1 L = 5.0 fb R = 0.7 T anti-k aveT = p F m = R m NP Corr. ˜ NNPDF2.1 (b) CMS measurement of the dijet double-dif-ferential cross section plotted versus invariantmass of the leading p T jets. Measurements arebinned in terms of the maximum rapidity ofthe two jets | y max | = max( | y | , | y | ).Figure 3. Measurements of the dijet mass cross sections at the LHC. straints on the PDFs. A similar comparison from the ATLAS Experiment is shownin Fig 6(a), but considering the dijet mass spectrum. This also illustrates the inter-est of pushing experimental measurements to more extreme regions of phase spacefor jet production. For dijet pairs at the maximum measured rapidity separation,the agreement of theory and experiment is clearly observed to break down at largeinvariant mass, which may indicate the need for higher order terms to accuratelydescribe the data. (a) D0 data divided by theory for the inclusive jet crosssection as a function of jet p T in six | y | bins. (GeV) T Jet p
200 300 400 1000 2000 R a t i o t o NN P D F . |y| < 0.5 DataCT10HERA1.5MSTW2008ABKM09
CMS = 7 TeVs -1 L = 5.0 fb R = 0.7 T anti-kExp. UncertaintyTheo. Uncertainty (b) CMS ratio of inclusive jet cross sec-tion ( | y | < .
5) to the theory predictionwith NNPDF2.1 [12] PDFs.Figure 4. Examples of relative precision of experimental measures and theory calculations oninclusive jet cross sections. ets@Colliders Measurements using multijet final states are powerful tools to test the applicabilityof pQCD calculations and to examine the running of the strong coupling constant.The D0 Experiment reported an analysis of multijet final states using a new ob-servable [13], R ∆ R , which is computed as a ratio of cross sections as illustrated inFig. 5. This observable measures the number of neighboring jets that accompany ajet of given p T within an angular region ∆ R defined in ( y, φ ) space. Because PDFdependencies largely cancel in the ratio, these results can be used to extract a mea-sure of α s , almost independent of initial assumptions on the renormalization groupequation (RGE). The measured results are displayed in Fig. 7(a) in bins of ∆ R ,the search region for neighboring jets, and p nbrT min , the minimum p T for inclusion ofneighboring jets as a function of inclusive jet p T . In each ∆ R interval the value of R ∆ R increases with jet p T until approaching the kinematic limit.Event shape studies [14] by the CMS Experiment include a measure of the cen-tral transverse thrust, τ ⊥ , C which probes QCD radiative processes and is mostlysensitive to the modeling of two- and three-jet topologies. The central transversethrust defined in Refs. [14,15] is a measure of radiation along an event’s transversethrust axis. The CMS measurement is shown in Fig. 6(b) compared to predictionsfrom five MC simulations. The parton shower (PS) MC generators pythia [16] and herwig [17] show generally good agreement with the data, except for pythia ver-sion 8 which shows some discrepancy at extreme values of τ ⊥ , C . The disagreementin calculations by alpgen [18] and MadGraph [19] implies that the regime of highjet p T and multiplicity where the explicit higher final state parton multiplicity MEcalculations of these generators significantly improves upon a pure PS approach hasnot been reached in these data. R D R = average number of neighboring jets per jet here: for D R < p /2in this exampleall jets havesame (p T , y)xy no neighborswithin D R: two jets haveone neighbor each: each of the four jetshas one neighbor: if all eventswere like this R D R = 0 R D R = 2/3 R D R = 1 Figure 5. Illustration of the variable R ∆ R ( p T , ∆ R, p nbrTmin ) = (cid:80) N jet ( p T ) i =1 N ( i ) nbr (∆ R, p nbrTmin ) /N jet ( p T ), which measures the number of neighboring jets accom-panying a jet satisfying a given requirement on transverse momentum. Robert Hirosky [TeV] m -1 × R a t i o w r t C T m -1 × R a t i o w r t C T ≤ m -1 × R a t i o w r t C T ≤ m -1 × R a t i o w r t C T ≤ m R a t i o w r t C T ≤ statistical errorData withuncertaintiesSystematicNLOJET++ × )) y* exp(0.3 T p = µ (Non-pert. corr.CT10NNPDF 2.1HERAPDF 1.5MSTW2008 -1 dt = 4.8 fb L ∫ s = 0.6 R jets, t anti-k ATLAS
Preliminary (a) ATLAS ratios of dijet double-differential crosssection, shown as a function of dijet invariant mass,to the theoretical prediction obtained using
NLO-JET++ [20] with the CT10 [21] PDF set. ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C Pythia6 ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C Pythia8 ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C Herwig++ ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C MadGraph+Pythia6 ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C ,C t ln -12 -10 -8 -6 -4 -2 D a t a / M C Alpgen+Pythia6 (b) CMS Distributions of the logarithmof the central transverse thrust for eventswith leading jet between 125 and 250 GeV.Results in data are compared to predictionsfrom five MC simulations. Shaded bandsrepresent the quadratic sum of the statisti-cal and systematic uncertainties.Figure 6. Measures of dijet and multijet distributions at the LHC.
Precise measures of jet production can be used to test the running of the strongcoupling constant α s , reaching significantly higher scales than are accessible byother methods. While the running of α s is predicted by the RGE, it remains theleast precisely known of the fundamental couplings. Furthermore, its running canbe modified at large Q by the presence of new physics, such as the presence of extraspatial dimensions.Because jet cross sections depend on a combination the strong coupling andparton distributions, the extraction of α s requires theory predictions as a continuousfunction of α s used in both the matrix elements and PDFs. The dependence of thePDFs on α s is parameterized by interpolating theory results based on PDF setsdetermined using incremental steps in α s . Correlated systematic experimental andtheoretical uncertainties are treated via the Hessian approach [22] and α s values aredetermined by minimizing a χ function with respect to α s and nuisance parameters ets@Colliders -3 -2 -1 R D R ( p T , D R , p T m i nnb r )
50 100 200 400 50 100 200 400 50 100 200 400 50 100 200 400 p T (GeV) p Tminnbr = 30 GeV p
Tminnbr = 50 GeV p
Tminnbr = 70 GeV p
Tminnbr = 90 GeV2.2 < D R < 2.61.8 < D R < 2.21.4 < D R < 1.8NLO pQCD +non-perturb. correct. m R = m F = p T MSTW2008 PDFs √ s = 1.96 TeVL = 0.7 fb -1 DØ (a) Measurement of R ∆ R for different intervals in the ∆ R searchregion for neighboring jets and requirements on p nbrTmin .
10 20 50 100 200 500
Q (GeV) a s ( Q ) H1 incl. jetsZEUS incl. jetsJADE evt. shapesALEPH evt. shapesDØ incl. jetsDØ R D R RGE for a s (M Z ) = 0.1184 ± a s (Q) from jet and event shape data (b) Measurements of the strongcoupling constant α s as a func-tion of momentum transfer Q .Figure 7. D0 measurements of (a) multijet distributions and (b) extraction of α s , determinedfrom D0 inclusive jets and event shapes. for the uncertainties. Results are show in Fig. 7(b) for this measurement andan earlier determination [23] using the D0 inclusive jet cross section. Additionaldeterminations of the strong coupling constant using inclusive and multijet crosssections in deep inelastic scattering data are presented in the contribution by R.Kogler to these proceedings. A great deal of progress has been made in measurements of jets in association withvector bosons ( V ) in recent years. These processes provide valuable tests of fixedorder and ME+PS predictions and their precise measurements provide importantconstraints on backgrounds for rare SM physics and searches for evidence of newphysics processes. The production of V + n -jet final states provides a handle to studymultiscale QCD processes and plays a significant role as background in searches forbeyond the standard model phenomena and many Higgs boson measurement chan-nels. Theoretical uncertainties on their production rates and kinematics introducelarge uncertainties and limit our ability to identify new physics processes.A large variety of Monte Carlo programs are compared to data to test theapplicability of perturbative calculations and phenomenologically tuned models indiverse regions of phase space for V + n -jet final states. Examples of MC calcu-lations compared to these data include: LO parton shower MC such as pythia ,leading-order matrix element plus parton shower matched MC such as alpgen and sherpa [24], calculations employing all-order resummation of wide angle emis-sions ( hej [25]), and next-to-leading order pQCD predictions for the productionof a vector boson plus multiple parton final states: blackhat+sherpa [26] and rocket+mcfm [27,28].Figure 8 shows examples of measurements performed at the D0 [29] and AT-LAS [30] Experiments. The D0 measurement of the W boson transverse momentumfor inclusive jet multiplicity bins is shown in Fig. 8(a). Predictions from blackhat- Robert Hirosky sherpa and hej show good agreement for all jet multiplicities. At jet p T below thethreshold of 20 GeV, nonperturbative effects dominate and the fixed order calcula-tions are expected to be unreliable.The measured H T distribution (scalar p T sum of reconstructed physics objects)for W events with one or more jets, measured at ATLAS, is shown in Fig. 8(b).The prediction, calculated inclusively at NLO by blackhat-sherpa , does notdescribe the data, due to the limited order of the matrix element calculation, whichdoes not include three or more real emissions of final-state partons. The LO MEcalculation of alpgen with up to five final-state partons, describes the data well.A study described in Ref. [30] considers improvements with a modified treatmentof blackhat-sherpa predictions introducing higher-order NLO terms describinghigher real emission multiplicities. A matching scheme is developed and requiredto reduce double-counting of cross sections, illustrating the challenges of comparingNLO calculations to complex inclusive jet variables like H T .The D0 Experiment reported a first measurement of the probability of emissionof a third jet in inclusive W + 2-jet events as a function of the dijet rapidity sepa-ration of the two leading jets. This measurement, shown in Fig. 9(a), is presentedin both pT- and rapidity-ordered scenarios, and a hybrid result considering theprobability of additional jet emission in the rapidity interval defined by the twoleading p T jets. Resummation predictions from hej are best able to describe boththe rate and shape across the full rapidity range.Studies involving differential properties in Z +jet production are similarly ofinterest to those involving W +jets. Using leptonic decay modes of the Z bosonallows for selection of low background samples and provides precise constraints onthe scale of hadronic recoil in the event. CDF reported measured Z +jet crosssections compared to NLO QCD combined with NLO EW calculations [31]. Ameasurement of the Z +jet cross section versus p T of the leading lepton from the Z decay [32] is shown to be well modeled in Fig. 9(b).Measurements of W/Z plus heavy flavor jets were also presented by experimentsat the Tevatron and LHC. These processes are important backgrounds to measure-ment of associated Higgs boson production and suffer from significant theoreticaluncertainty on their production. An earlier measurement [33] of W + b -jet produc-tion using 1.9 fb − of Tevatron Run 2 data by the CDF Experiment is shown inFig. 10(a), where the exclusive W + b -jet cross section is determined after removal of c -jet and light-flavor jet background using templates based on the secondary vertexmass in b -tagged jets. This result yielded a cross section of about three standard de-viations in excess of SM predictions at NLO accuracy for events with exactly one ortwo jets with E T >
20 GeV and lepton from W decay with similar requirement. Amore recent measurement [34] performed by ATLAS considers exclusive productionof W + b -jet events for one, two, and one or two identified b jets. The results shownin Fig. 10(b) are in good agreement with NLO predictions for the one jet exclusivecase, and agree to within 1.5 s.d. for the other categories. A measurement [35] pro-vided by the D0 Experiment following the conference finds the inclusive W + b -jetproduction cross section to be in agreement with NLO predictions within scale andPDF uncertainties.The CMS Experiment provided a recent study [36] of the Z/γ ∗ + b -jet cross ets@Colliders ( / G e V ) T W / dp σ d ⋅ W σ / -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 j e t ≥ W + - × j e t s ≥ W + - × j e t s ≥ W + - × j e t s ≥ W + >20 GeV T p>40 GeV, TW |<1.1, M e η >15 GeV, | Te p |<3.2 jet >20 GeV, |y Tjet =0.5, p cone
R )+jets+X ν e → , W( -1 DØ, 3.7 fbNLO Blackhat+SherpaHEJ Alpgen+PythiaAlpgen+HerwigPythiaHerwigSherpa (GeV) T W boson p0 50 100 150 200 250 300 T heo r y / D a t a ≥ )+ ν e → W( (GeV) T W boson p0 50 100 150 200 250 300 T heo r y / D a t a ≥ )+ ν e → W( (GeV) T W boson p0 50 100 150 200 250 300 T heo r y / D a t a ≥ )+ ν e → W( (GeV) T W boson p0 50 100 150 200 250 300 T heo r y / D a t a ≥ )+ ν e → W( (a) D0 measurement of the W boson transversemomentum distributions in inclusive W + n -jet events for n = 1 − [ pb / G e V ] T / d H σ d -7 -6 -5 -4 -3 -2 -1 + jets ν l → W =7 TeVsData 2010, ALPGENSHERPABLACKHAT-SHERPA -1 Ldt=36 pb ∫ jets, R=0.4 T anti-k |<4.4 jet y>30 GeV, | Tjet p ATLAS j e t s ≥ W + - j e t s , x ≥ W + -2 ≥ W + -3 ≥ W + [ pb / G e V ] T / d H σ d -7 -6 -5 -4 -3 -2 -1
200 400 600 T heo r y / D a t a ≥ W +
200 400 600 T heo r y / D a t a [GeV] T H
200 400 600 T heo r y / D a t a ≥ W + [GeV] T H
200 400 600 T heo r y / D a t a (b) ATLAS W+jets cross section as a func-tion of H T for separate jet multiplicities. Re-sults are compared to predictions from alpgen , sherpa , and blackhat-sherpa. Figure 8. W + n -jet measurements from the D0 and ATLAS Experiments. sections using 2.1 fb − of pp collisions at √ s = 7 TeV and additional results wereprovided by the CDF Experiment [37] using the full CDF data set. Figure 11(a)shows the p T of the bb system in an event sample selected to identify Z + bb finalstates at CMS. In general the CMS results indicate that predictions based on the MadGraph event generator interfaced with pythia and normalized to the NNLOinclusive cross section provide a fair description of the data.Finally a measurement [38] of the differential cross section for photon+ b -jetproduction was provided by the D0 Experiment. The results are summarized inFig. 11(b) which shows the ratios of data to the NLO QCD calculations and dif-ferent MC simulations to the same NLO calculations. The observed variance withrespect to several PDF models is also shown. The data can only be described by0 Robert Hirosky y(j,j) ∆ P r obab ili t y o f t h i r d j e t e m i ss i on >20 GeV T p>40 GeV, TW |<1.1, M e η >15 GeV, | Te p |<3.2 jet >20 GeV, |y Tjet =0.5, p cone
R 2jets+X ≥ )+ ν e → , W( -1 DØ, 3.7 fb(most rapidity-separated jets) jets) T (leading p jets, rapidity gap emission) T (leading pNLO Blackhat+SherpaHEJSherpa (a) D0 measurement of the probability of emis-sion of a third jet in inclusive W + 2-jet eventsfor various definitions of rapidity interval as de-scribed in the text. [f b / ( G e V / c )] l T / dp σ d -1 CDF Run II Preliminary ≥ ) + - l + l → *( γ Z/ > 25 GeV/c lT | < 1.0; p l η ; | µ l = e, 2.1 ≤ | jet
30 GeV/c, |Y ≥ jetT p -1 CDF Data L = 9.64 fb Systematic uncertaintiesNLO LOOPSIM+MCFMn MSTW2008NNLO PDF Corrected to hadron level ) - lT + P + lT + P Tj P j Σ (21 = T H 21 = µ [GeV/c] lT p20 30 40 50 60 100 200 300 D a t a / L OO PS I M LO MCFMNLO LOOPSIM+MCFMn NLO MCFM NLO)n/2 ( µ = µ ; µ = 2 µ (b) CDF measured cross section in Z/γ ∗ + ≥ p T .Figure 9. Measurements of W and Z +jets differential cross sections from the Tevatron. including higher order corrections into the NLO QCD predictions, such as thosecurrently present as additional real emissions in the sherpa MC generator. Thisis an interesting measurement to compare with future results from the LHC, sincethe production of photon plus heavy flavor events at the Tevatron is dominated byfinal state gluon splitting q ¯ q → γg ( g → b ¯ b ), while at the LHC bq → bγ dominatesfor most of the experimentally accessible p T ranges. Another area of study at the LHC and Tevatron experiments involves nonpertur-bative processes. Reviews of measurements involving multiparton interactions andtotal cross sections are presented respectively by E. Dobson and N. Cartiglia inthese proceedings. Presented here is a recent measurement [39] by the CMS ex-periment of the contribution from diffractive dijet production to the inclusive dijetcross section. Events are selected to contain at least two jets with p T >
20 GeVwith the leading jet satisfying | η jet | < . a ξ = M X /s , whichapproximates the fractional momentum loss of the scattered proton in single diffrac-tive events is approximated ( ˜ ξ ) from the energies and longitudinal momenta of all a In this study of diffractively produced events M X represents the mass of the jet system which isseparated from the an opposing scattered proton by a large rapidity gap ets@Colliders (a) CDF fit to selected events in W + b -jet anal-ysis, based on a maximum likelihood of the ra-tio of vertex mass distributions for signal andbackground in b tagged jets in the selected datasample in 1.9 fb − of Tevatron Run 2 data. b - j e t ) [ pb ] ≥ + ν l → ( W σ Electron Chan.Electron and Muon Chan.Muon Chan.NLO 5FNSALPGEN + JIMMY(b-jet from ME and PS)ALPGEN + JIMMY(b-jet only from ME)PYTHIA
ATLAS =7 TeVsData 2010, -1 Ldt = 35 pb ∫ (b) ATLAS measurement of the W + b -jet crosssection in the 1, 2, and 1+2 jet exclusive bins.The measurements are compared with NLOpredictions in pQCD.Figure 10. CDF and ATLAS measurements of the W + b production cross section. (GeV) bbT p E v en t s / G e V DataZ+lZ+cZ+bttZZ -1 = 7 TeV, L = 2.1 fbsCMS Preliminary JES + B-Tag Uncertainties (GeV) bbT p D a t a / M C (a) CMS measurement of the p T of the b -jetsystem in events selected to identify Z + bb finalstates. (GeV) g T p R a t i o t o N L O -0.500.511.522.533.5 (GeV) g T p R a t i o t o N L O -0.500.511.522.533.5 Data / NLOSHERPA / NLOPYTHIA / NLO fact./ NLO T k Scale uncertaintyPDF uncertaintyMSTW/CTEQABKM/CTEQ -1 DØ, L = 8.7 fb |<1.0 g |y >15 GeV jetT |<1.5, p jet |y (a) (b) D0 ratio of γ + b differential cross sectionbetween data and NLO QCD predictions withuncertainties for the rapidity region | y γ | < . Z + b and γ + b cross sections from CMS and D0. Robert Hirosky particle flow [41] objects measured in the region | η | < .
9. Figure 12(a) shows the η distribution for the second leading jet after application of a requirement to enhancethe diffractive contribution to the measured cross section. This requirement [39]is equivalent to imposing a pseudorapidity gap of at least 1.9 units, enhancing thediffractive component in the data, and selecting events with the jets mainly in thehemisphere opposite to that of the gap. A combination of PYTHIA6 with tune Z2and single diffractive events modeled with
POMPYT [42] is found to agree withthe data reasonably well. Figure 12(b) shows the reconstructed ˜ ξ distribution com-pared to detector-level MC predictions with and without the inclusion of diffractivedijet production. jet h -5 -4 -3 -2 -1 0 1 2 3 4 5 h d N / d DATAPYTHIA6 Z2 + POMPYTPYTHIA6 Z2 >20 GeV
T1,2 =7 TeV, ps, jet CMS preliminary, p+p->jet<3 max h (a) CMS reconstructed pseudorapidity distri-butions of the second-leading jets, after im-posing a pseudorapidity gap requirement, andcompared to detector-level MC predictionswith and without diffractive dijet production. x~ -3 -2 -1 x~ d N / d DATAPYTHIA6 Z2 + POMPYTPYTHIA6 Z2 >20 GeV
T1,2 |<4.4, p h =7 TeV, |s, jet CMS preliminary, p+p->jet (b) CMS reconstructed ˜ ξ distribution com-pared to detector-level MC predictions withand without diffractive dijet production.Figure 12. CMS study of diffractive dijet production. Remarkable progress continues in measurements concerned with jet production.Unprecedented data sets, measurement precision, and improvements in theory haveenabled ever more rigorous tests of standard model predictions and the extractionof fundamental parameters such as the strong coupling constant at ever increasingmomentum scales. These data inform models for PDFs and constrain importantbackgrounds for rare SM and new physics searches. They further test the applica-bility of fixed order perturbative calculations, parton shower MC, and other modelsover an exceptionally large phase space of final states. As shown in this brief sam-pling of results, numerous areas require work understand the effects of higher orderand nonperturbative processes to improve our modeling of the data and furtherenhance sensitivities to new physics processes.I wish to thank the organizers of Physics in Collision 2012 for giving me theopportunity to present these results and the ATLAS, CDF, CMS, D0, HERA, andZEUS collaborations for providing the results included in this talk. ets@Colliders References
1. F. Aversa et al. , Phys. Rev. Lett , 401 (1990);W. T. Giele, E.W. N. Glover, and D. A. Kosower, Phys. Rev. Lett , 2019(1994);D. Ellis, Z. Kunszt, and D. E. Soper, Phys. Rev. Lett , 2121 (1990).2. See contribution from J. Bl¨umlein in these proceedings.3. M. Cacciari, G. P. Salam, G. Soyez, CERN-PH-TH/2011-297, arXiv:1111.6097.4. V. M. Abazov et al. (D0 Collaboration), Phys. Rev. D (2012) 052006.5. A. Abulenci et al. (CDF Collaboration), Phys. Rev. D , 071103 (2006).6. S. Catani, Y. L. Dokshitzer, M. H. Seymour and B. R. Webber, Nucl. Phys.B , 187 (1993);S. D. Ellis and D. E. Soper, Phys. Rev. D , 3160 (1993).7. M. Cacciari and G. P. Salam, JHEP
005 (2008).8. V. Abazov et al. (D0 Collaboration), Phys. Rev. D , 052006 (2012); V.Abazov et al. (D0 Collaboration), Phys. Rev. Lett. , 062001 (2008).9. S. Chatrchyan et al. (CMS Collaboration), [arXiv:1212.6660].10. G. Aad et al. (ATLAS Collaboration), Phys. Rev. D (2012) 014022.11. B. Chapleau, presentation at 36th International Conference for High EnergyPhysics, July 2012,http://indico.cern.ch/conferenceDisplay.py?confId=181298.12. R. D. Ball et al. , Nucl.Phys. B , 136 (2010).13. V. Abazov et al. (D0 Collaboration), Phys. Lett. B , 56 (2012).14. V. Khachatryan et al. (D0 Collaboration), Phys. Lett. B The Herwig Event Generator , Comp. Phys. Comm. (1992) 465.18. M. L. Mangano et al., J. High Energy Phys. 07, 1 (2003).19. J. Alwall, M. Herquet, F. Maltoni, O.Mattelaer, T. Stelzer, J. High EnergyPhys. (2011) 1029.20. Z. Nagy, Phys. Rev. Lett. (2002) 122003 [hep-ph/0110315]; Phys. Rev.D (2003) 094002.21. J. Gao et al. , arXiv:1302.6246.22. A. Cooper-Sarkar and C. Gwenlan, in Proceedings of the Workshop: HERAand the LHC, Part A, edited by A. De Roeck and H. Jung, Geneva, Switzerland(2005), CERN- 2005-014, DESY-PROC-2005-01, arXiv:hep-ph/0601012, seepart 2, section 3.23. V. Abazov et al. (D0 Collaboration), Phys. Rev. D , 111107 (2009).24. T. Gleisberg et al., J. High Energy Phys. 02, 7 (2009).25. J. R. Andersen and J. M. Smillie, J. High Energy Phys. 01, 39 (2010); J. R.Andersen and J. M. Smillie, Phys. Rev. D 81, 114021 (2010); J. R. Andersenand J. M. Smillie, Nucl. Phys. Proc. Suppl. 205, 205 (2010); J. R. Andersenand J. M. Smillie, J. High Energy Phys. 06, 10 (2011); J. R. Andersen, T.Hapola and J. M. Smillie, J. High Energy Phys. 09, 047 (2012).4 Robert Hirosky
26. C. F. Berger, et al., Phys. Rev. Lett. 102, 222001 (2009); C. F. Berger et al.,Multi-jet cross sections at NLO with BlackHat and Sherpa, arXiv:0905.2735[hep-ph]; C. F. Berger et al., Phys. Rev. D 80, 074036 (2009).27. J. M. Campbell and R. K. Ellis, Phys. Rev. D 65, 113007 (2002); J. M.Campbell, R. K. Ellis and D. L. Rainwater, Phys. Rev. D 68, 94021 (2003).28. R. K. Ellis et al., J. High Energy Phys. 01, 12 (2009); W. T. Giele and G.Zanderighi, J. High Energy Phys. 06, 38 (2008).29. V. Abazov et al. (D0 Collaboration), [arXiv:1302.6508].30. G. Aad et al. (ATLAS Collaboration), Phys. Rev. D (2012) 092002.31. A. Denner, S. Dittmaier, T. Kasprzik, A. M¨uck, JHEP et al. (CDF Collaboration), Phys. Rev. Lett. et al. (ATLAS Collaboration), Phys. Lett. B
418 (2012).35. V. Abazov et al. (D0 Collaboration), Phys. Lett. B et al. (CMS Collaboration), J. High Energy Phys. (2012)126.37. T. Aaltonen et al. et al. (D0 Collaboration), Phys. Lett. B
32 (2012).39. S. Chatrchyan et al. (CMS Collaboration), Phys. Rev. D (2013) 012006.40. The pseudorapidity is defined as η = − ln[tan( θ/ θθ