1 / 46

Tom Danielson University of Wisconsin – Madison

A Measurement of Multijet Production in Low- x Bj Deep Inelastic Scattering with ZEUS at HERA. Tom Danielson University of Wisconsin – Madison. Study of Partons. Particle Scattering Study charge & magnetic moment distributions Scattering via probe exchange Wavelength

merle
Download Presentation

Tom Danielson University of Wisconsin – Madison

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. A Measurement of Multijet Production in Low-xBjDeep Inelastic Scattering with ZEUS at HERA Tom Danielson University of Wisconsin – Madison

  2. Study of Partons • Particle Scattering • Study charge & magnetic moment distributions • Scattering via probe exchange • Wavelength • Special Case : Deep Inelastic Scattering • High energy lepton transfers momentum to a nucleon via probe h : Plank’s Constant Q: related to momentum of photon

  3. Naïve Quark Parton Model e(k) e(k’)  *(q)  Jet p(P) • Scattering on proton is sum of elastic scattering on all of the proton’s constituents (partons) • Point-like Partons • Structure Functions: quantify distribution of partons and their momentum • Parton Density Functions (PDF) • Must be extracted from experimental structure function measurements Bjorken Scaling: Only x dependence x = fraction of momentum carried by quark

  4. QCD Theory • Gluons: vector colored bosons carry strong force • Gluons produce quark and gluon pairs • Quarks gain transverse momentum • Gluon-driven increase in F2 Bjorken Scaling Violation: Fi(x) Fi(x,Q2) Observation of QCD effects  Small x

  5. e±p scattering mediated by g, Z0 (Neutral Current) W± (Charged Current) Deep Inelastic Scattering Squared center-of-mass energy of ep system Exchange boson virtuality (negative squared 4-mom. of exchange boson) Fractional proton momentum carried by struck quark (QPM) Fraction of electron energy transferred to proton in proton rest frame Kinematics relation

  6. Neutral Current (NC) DIS Cross Section • Express NC cross section in terms of DIS kinematics and proton structure • F2, FL, F3 → proton structure functions • F2 → interaction between transversely polarized photon and partons • FL → interaction between longitudinally polarized photon and partons • xBjF3 → parity violating term from Z0exchange • fi(x,Q2) → Parton density functions (PDFs) • Extracted from fits to structure function measurements

  7. Perturbative QCD and Splitting Functions      lepton S S S photon  gluon quark • DIS cross sections expressed as series expansion in as • Splitting functions give probability quark or gluon to split into parton pair • Leading Order quark and gluon splitting diagrams shown

  8. QCD evolution: Described by DGLAP equations Done by perturbative resummations over diagrams DGLAP Leading Log: Sum over diagrams contributing ln(Q2) terms Double Leading Log: Sum over ln(Q2) ln(1/x) terms BFKL: Sum over diagrams contributing ln(1/x) terms CCFM: Splitting functions with non-Sudakov form factors kt factorization interpolates between DGLAP, BFKL QCD Parton Evolution Splitting Functions

  9. Kinematic Coverage of Colliders LHC Tevatron HERA • HERA: e±p • 0.1 < Q2 < 20000 GeV2 • 10-6 < xBj < 0.9 • CM energy 300 – 318 GeV2 • Tevatron: pp • 2000 < Q2 < 250000 GeV2 • 10-2 –10-3 < xBj < 0.9 • CM Energy 1.8 – 1.96 TeV • LHC: pp • 100 < Q2 < 108 GeV2 • 10-6 < xBj < 0.9 • CM Energy 14 TeV (startup) • HERA coverage in x similar to LHC • ~2 orders of magnitude lower in x than Tevatron

  10. Evolving PDFs to LHC • Are extracted PDFs usable in the LHC kinematic range? • (10-6 < x < 1) • Account for low-x effects • Saturation • Multiple interactions • Does DGLAP evolution work sufficiently to extrapolate? • Gluon contribution dominant at low x

  11. Testing QCD Parton Evolution: Multijets Jets in DIS Boson-Gluon Fusion (BGF) QCD Compton • Partons not experimentally observed • Colorless hadrons form through hadronization • Collimated “spray” of particles → Jets • Dijets • Produced from Boson-Gluon Fusion, QCD Compton processes • Leading-Order diagrams O(a1s) • Direct coupling to gluon

  12. Testing Parton Evolution: Jets at low xBj increasing η • Study QCD evolution schemes with jets at low xBjin DIS • DGLAP • Leading Log: Strong ordering in kT, ordering in x • Double Leading Log: Strong ordering in kT, and x • well tested over large range of Q2 • BFKL: sums over ln(1/x) terms • Strong ordering in x, but not ordered in kT • More energetic forward jets • Jets from hard scatter less correlated in energies, angles • CCFM: kT factorization • Approaches BFKL for low xBj, DGLAP for high Q2 • angular ordering (instead of kTordering) • uses unintegrated parton densities

  13. Testing Parton Evolution: Jets at low xBj • Study QCD evolution schemes with jets at low xBjin DIS • DGLAP • Leading Log: Strong ordering in kT, ordering in x • Double Leading Log: Strong ordering in kT, and x • well tested over large range of Q2 • BFKL: sums over ln(1/x) terms • Strong ordering in x, but not ordered in kT • More energetic forward jets • Jets from hard scatter less correlated in energies, angles • CCFM: kT factorization • Approaches BFKL for low xBj, DGLAP for high Q2 • angular ordering (instead of kTordering) • uses unintegrated parton densities Applicability range of evolution approaches

  14. HERA Accelerator at DESY • Beam energies • Proton energies • 820 GeV (1992 – 1997) • 920 GeV (1998 – 2007) • 460, 575 GeV (2007) • Facilitated FL Measurement • 27.5 GeV e- or e+ • 300 – 318 GeV Center-of-Mass Energy • Equivalent to ~30 TeV fixed target • 220 bunches • Not all filled • Crossings every 96 ns • Typical beam currents • Proton: 100 mA • Electron: 40 mA • Instantaneous luminosity • ~5 x 1031cm-2s-1 HERMES H1 H1 ZEUS HERA-B DESY Hamburg, Germany

  15. ZEUS Detector x z ZEUS coordinate system p 920 GeV MUON CAL CTD y e+ 27.5 GeV Multi-purpose detector Characterizes energy, direction and type of particles in the final state

  16. ZEUS Calorimeter h = 0.0q = 90o h = -0.75q = 129.1o h = 1.1q = 36.7o h = 3.0q = 5.7o e p FCAL BCAL RCAL Hadronic Cells Electromagnetic Cells • Alternating layers of depleted uranium and plastic scintillator • Sampling Calorimeter • Most energy absorbed by DU • Compensating • Equal hadronic and electromagnetic response • Segmented • 3 Sections (FCAL, RCAL, BCAL) • Divided further into modules, cells • 2 PMTs/cell • Energy resolutions (test beam) • Electromagnetic: 18% / √E • Hadronic: 35% / √E • Timing Resolution • ~1 – 2 ns • 99.7% solid angle coverage • (-3.5 < h< 4.0) h = -ln(tan(q/2))

  17. Central Tracking Detector (CTD) • Drift chamber inside 1.43 T solenoid • Angular coverage: 15o < q < 164o (-1.96 < h < 2.04) • Organization • 16 azimuthal sectors • 9 concentric superlayers • 8 radial layers superlayer • 32 – 96 cells per superlayer • Resolutions • Transverse momentum s/pT = [(0.005pT)2 + (0.0016)2]1/2 • Vertex resolution • longitudinal (z): 4mm • transverse (x-y): 1mm Superlayers Sectors

  18. ZEUS Trigger • 10 MHz crossing rate, 100 kHz Background rate, 10Hz physics rate • First level: Use data subset: 10 MHz → 500 Hz • Pipelined without deadtime • Global and regional energy sums • Isolated m and e+ recognition • Basic tracking information • Second level: Use all data: 500 Hz → 100 Hz • Calorimeter timing cuts • Tracking, vertex information • Simple physics filters • Third level: Use full reconstruction information: • 100 Hz → 10 Hz • Full event information • Refined jet and electron finding • Complete tracking algorithms • Advanced physics filters

  19. ZEUS Event Reconstruction • Vertex: Use CTD tracks fit to 5-parameter Helix model • Calorimeter • Use cell position, magnitude of PMT pulse, timing of PMT pulse • Island formation: Cells merged based on location and size of energy deposits • e-/e+: SINISTRA95 Neural Network electron finder • Use electromagnetic islands as input • Require corresponding CTD track if island within CTD acceptance region • Output: list of electron candidates with associated probabilities • Use most probable candidate as DIS electron • Energy Flow Objects (EFOs) • Combine track and calorimeter information for hadrons • CTD has better angular resolution than CAL • CTD has better energy resolution at low energy than CAL

  20. Kinematic Reconstruction Four measured quantities: EH, gH, E‘e, qe. Only two independent quantities. Reconstruction methods used: Electron method, Jacquet-Blondel Methods have different resolutions over different kinematic regions gH qe

  21. Jet FindingHadronic Center of Mass Frame • Select a frame to optimize multijet finding • Single jet event • Struck quark rebounds parallel to frame’s z-axis • Zero transverse energy (ET) • Dijet event • Jets balanced in ET at leading order • Equivalent to Breit Frame, apart from longitudinal boost • Used in other ZEUS analysis • HCM frame used to compare to results from H1 collab.

  22. Jet Reconstruction – Energy Scale • Calorimeter response calibrated using Uranium background • Methods for finding jet scale uncertainty • 1: Use single jet DIS event • Scattered electron, jet balanced in momentum • Predict jet energy from scattered electron energy, compare rel. diff. • 2: Use MC to correct for energy loss from inactive material • Correct jets from MC and data to hadron level • Compare using tracking information in cone around jets • Jet energy scale uncertainty • ± 1% for ET > 10 GeV, ± 3% otherwise • ± 1% → ~5% uncertainty in s • 41% of dijets • ± 3% → ~15% uncertainty in s • 59% of dijets Jet energy scale uncertainty as function of hLAB

  23. Jet Trigger – Cone Algorithm R • Used in higher level trigger selection • Maximize ET of hadrons in cone of fixed size • Construct seeds from energy deposits in cells • Iterate until stable cone found • Merge overlapping cones according to scheme • For the jet: • Issues with older cone algorithms • Overlapping jets and soft radiation between jets • Seed – Energy threshold • Infrared unsafe – s → ∞ as seed threshold → 0

  24. Jet Reconstruction – kt Algorithm • kt: In ep transverse momentum with respect to beam line • For every object i and every pair of objects i, j calculate • di = (ET,i)2 → Distance to beam line in momentum space • dij = min{E2T,i,E2T,j}[(Dh)2 + (Df)2]→ Distance between objects • Combine all di, dijinto set • Calculate min{di,dij} for each object and pair of objects • If minimum of set corresponds to di → Object for ditaken as jet • If minimum of set corresponds to dij→ Combine objects i and j • Advantages • No seed required and no overlapping jets • Suitable for beyond-NLO pQCD calculations

  25. Leading Order Monte Carlo • MC events + detector simulation • to compare with the data • to correct data for detector effects • ARIADNE v4.10 • Color Dipole Model (CDM) • Gluons emitted from color field between quark-antiquark pairs • Gluons not necessarily kt ordered (BFKL-like) • Supplemented with boson-gluon fusion processes • LEPTO v6.5.1 • Matrix Element + Parton Shower (MEPS) • Parton cascade: approximate higher orders in LO calc • Decreasing virtuality (q2) as cascade progresses • Radiated gluons kt-ordered (DGLAP) • Both use Lund String Model to simulate hadronization CDM MEPS

  26. Next-to-Leading-Order (NLO) pQCD Calculations • NLOjet: Inclusion of single gluon emission in dijet, trijet final states • Terms of up to O(a2s) (O(a3s)) included for dijet (trijet) calculations • One-loop corrections for virtual particles • Correction for 3rd (4th) parton in final state (soft/collinear gluon emissions) • Calculations do not include parton showering, hadronization • Corrections taken from MC • Scale dependences • Renormalization scale: scale for evaluating as • Factorization scale: scale at which PDFs are evaluated Programs for DIS • DISENT • MEPJET • DISASTER++ • NLOJET++ (used)

  27. Application of NLOjet Calculations Scales: • Varied simultaneously by factors 4, ¼ for theoretical uncertainty • Renormalization scale uncertainty estimates contributions from higher-order terms PDF set: CTEQ6M Hadronization corrections from LEPTO For certain jet phase space, O(as3) calculations possible for dijets

  28. Low xBj DIS Multijet Sample • Data: 1998-2000 electron and positron: 82.2 pb-1

  29. Triggering Efficiency • Trigger selection → well-reconstructed event • Trigger rate, efficiency affected by bandwidth, storage limits • Low-Q2 trigger ~100% efficient, but prescaled • Enhance low-Q2 efficiency of trigger selection with medium-Q2, dijet and prescaled low-Q2 trigger • ~80% overall efficiency at low Q2 with combined trigger chain Low Q2 Low Q2 (prescaled) Medium Q2 Dijet Total

  30. Low-xBj Dijets vs. Ariadne ZEUS 82 pb-1 ARIADNE Discard Kinematic variables well-described by Ariadne Jet variables reasonably well-described 6% Excess jets at High ET included in systematics Level of agreement similar to other ZEUS multijet analyses

  31. Low-xBj Dijets vs. Lepto ZEUS 82 pb-1 LEPTO Kinematic variables reasonably well-described by Lepto High-ET excess reduced to ~2%

  32. Low-xBj Trijets vs. Ariadne ZEUS 82 pb-1 ARIADNE Trijet variables reasonably well-described by Ariadne Excess jets at High ET included in systematics

  33. Low-xBj Trijets vs. Lepto ZEUS 82 pb-1 LEPTO Trijet variables reasonably well-described by Lepto High-ET excess reduced

  34. Systematic Uncertainties • Kinematic and jet cuts: +/- Ariadne resolutions • Jet energy scale variation • Difference between Ariadne and Lepto acceptance correction • Largest uncertainties from • Lepto instead of Ariadne: 5 – 10 % • Sometimes larger • Jet energy scale: 5 – 15 % • Cut on E1,2T,HCM: 5 – 10 % for jet correlations • Cut on E3T,HCM for trijet sample: 5 – 10 %

  35. ZEUS Dijet, Trijets vs. xBj • Test DGLAP-based calculations from NLOjet for inclusive cross sections, ratios • Dijet, trijet cross sections both well-described. • Measure cross section ratios → cancel theoretical uncertainties • Ratios also well-described, esp. at low xBj

  36. Jet Correlations • Inclusive cross sections well-described by NLOjet calculations • Examine correlations between two highest ET,HCM jets for non-DGLAP effects Separation in h of two hardest jets Difference in ET,HCM of two hardest jets Magnitude of vector sum of jet pT Scaled magnitude of vector difference of jet pT Separation in f of two hardest jets

  37. Separation in h of two hardest jets Dijets Trijets • NLOjet calculations describe h correlations • Description of data independent of xBj • Higher-order terms not needed to describe dijet h correlations

  38. Dijets:Mag. of Vector Sum of pT from Jet 1,2 • |SpT| sensitive to parton evolution, gluon radiation • |SpT| = 0 without gluon radiation • NLOjet calculations at O(as2) do not describe dijet data at low xBj • NLOjet calculations at O(as3) describe data, even at low xBj • Higher order terms important at low xBj • Allows for more gluon emission

  39. Trijets: Mag. of Vector Sum of pT from Jet 1,2 • Expand study of |SpT| by examining trijets • Higher order measurement (O(as2) at LO) • NLOjet calculations at O(as3) describe the data • Better description at higher values of xBj • Higher-order NLO calculations not available • Effect much less pronounced than for dijets vs. O(as2) NLOjet calculations

  40. Dijets: Scaled Mag. of Vec. Difference of Jet pT • |DpT|/(2 ETjet1) sensitive to parton evolution, gluon radiation • |DpT|/(2 ETjet1) = 1 without gluon radiation • NLOjet calculations at O(as2) do not describe dijet data at low xBj • NLOjet calculations at O(as3) describe data, even at low xBj • Higher order terms important at low xBj • Allows for more gluon emission

  41. Dijets:Separation in f of Two Hardest Jets • |Df| sensitive to parton evolution, gluon radiation • |Df| = p without gluon radiation • NLOjet calculations at O(as2) do not describe dijet data at low xBj • NLOjet calculations at O(as3) describe data, even at low xBj • Higher order terms important at low xBj • Allows for more gluon emission

  42. Trijets: Separation in f of Two Hardest Jets • Expand study of |Df| by examining trijets • Combine first two bins in |Df| to reduce stat., systematic errors • NLOjet calculations at O(as3) describe the data • Slightly better description at higher values of xBj • Higher-order NLO calculations not available • Large theoretical uncertainties

  43. ZEUS 2,3-jet Azimuthal Correl. vs. Q2, xBj Dijets Trijets • |Df*| < 2p/3 → high-ET forward jet • Measurements sensitive both to angular correlations, forward jets • NLOjet calculations at O(as3) describe dijet, trijet data, even at low xBj • Dijets at low xBj: O(as3) calcs for dijets ~ 10 x O(as2) calcs

  44. H1 Trijet Results vs. NLOjet • Parton evolution for multijet events examined: • Angular correlations • Normalized jet energies • For • Inclusive trijet • Trijet with 1, 2 forward jets Inclusive Trijets • Inclusive trijet data generally well-described by NLOjet calculations at O(as3) • Data above calculations at low xBj

  45. Summary: Low-xBj Dynamics at HERA • Dijet, trijet correlations at ZEUS measured at small xBj(10-2 < xBj < 10-4) • Dijet, trijet pT and azimuthal correlations most sensitive to gluon radiation, parton evolution • Higher-order terms important at low Q2, xBj • Effects more pronounced for dijets • Higher-order calculations up to 10x larger at very small xBj • Correlations well-described by NLOjet calcs. • Cross sections in xBj, correlations in h well-described by NLOjet calcs. • Less sensitive to parton evolution scheme • DGLAP evolution with current PDFs safe to (x ~ 10-4) • O(as2) calcs not good enough for entire dijet phase space. Need O(as3).

  46. ZEUS inclusive forward jets: Expect more forward jets with BFKL, CCFM at low xBj LEPTO (DGLAP) consistently below measurements CASCDE (CCFM) inconsistent in description of data Better description from J2003 set 2 than set 1 ARIADNE (CDM, ~BFKL) gives best overall description H1 dijet azimuthal correlations: Expect broader Df* spectrum from BFKL, CCFM CASCADE with J2003 pdf set describes data at higher xBj CASCADE predictions dependent on unintegrated gluon PDF All models fail to describe Df* at low xBj Other ZEUS and H1 Results:Leading-Order Models

More Related