1 / 52

Boson + jets production and the search for New Physics

Boson + jets production and the search for New Physics. Monica D’Onofrio IFAE-Barcelona Results from D0 and CDF collaborations IFAE Seminar, Barcelona 8 th May 2006. Going beyond the Standard Model: Supersymmetry

ronli
Download Presentation

Boson + jets production and the search for New Physics

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. Boson + jets production and the search for New Physics Monica D’Onofrio IFAE-Barcelona Results from D0 and CDF collaborations IFAE Seminar, Barcelona 8th May 2006

  2. Going beyond the Standard Model: Supersymmetry Squarks/gluinos production as Golden signature for SUSY at the LHC, but also at the Tevatron Key points: Jet Reconstruction Background estimations Use Tevatron Data on Boson + jets production: Important by itself as QCD measurement Z and photon + jet used to define jet energy correction Test ground of Monte Carlo generators Latest results from the Tevatron on squarks/gluinos searches Extension to possible non-SUSY searches Summary and future plans Outline IFAE Seminar, Barcelona 8/05/2006

  3. Thanks to R.Kolb The Standard Model Higgs • Matter is made out of fermions: • 3 generations of quarks and leptons • Forces are carried by Bosons: • Electroweak: ,W,Z • Strong: gluons • Higgs boson: • Gives mass to particles  Not found yet • Remarkably successful description of known phenomena but ... The Standard Model is theoretically incomplete • Mass hierarchy problem • radiative correction in Higgs sector • Unification • Dark Matter • Matter-antimatter asymmetry .. Many good reasons to believe there is unknown physics beyond SM IFAE Seminar, Barcelona 8/05/2006

  4. The Hierarchy problem The SM requires a non-vanishing VEV for the Higgs at the minimum of the potential V if m2H < 0, VEV results in: Experimentally, <H> = 174 GeV and m2H ~ –(100 GeV)2 + quantum corrections from virtual effects of particles coupling to Higgs field Fermion loop loop of scalar particles UV ultraviolet cutoff Mass of Higgs scalar with quantum corrections is kept small only with fine tuning of the parameters! Possible solution: introduce a symmetry to cancel all dangerous contributions IFAE Seminar, Barcelona 8/05/2006

  5. Supersymmetry • New symmetry relating fermions and bosons to cancel out contributions to Dm2H : Supersymmetry • Minimal SuperSymmetric SM (MSSM): • Mirror spectrum of particles • Enlarged Higgs sector (two doublets with 5 physical states) • Define R-parity = (-1)3(B-L)+2s • R = 1 for SM particles, R = -1 for MSSM partners  if R-parity is conserved, sparticles produced in pair, Lightest Supersymmetric Particle (LSP) is stable Q|Boson> = Fermion Q|Fermion> = Boson IFAE Seminar, Barcelona 8/05/2006

  6. What’s Nice about SUSY? SM With SUSY • Unifications of forces possible • Dark matter candidate exists: • LSP stable if R-parity is conserved • Typically LSP is the lightest neutralino • Current mass limit > 43 GeV • Abundance of neutralino matches Dark Matter density in the Universe • Naturally solve the hierarchy problem • No fine-tuning required • Changes relationship between mW, mtop and mH IFAE Seminar, Barcelona 8/05/2006

  7. EWK GUT Symmetry breaking • No SUSY particles found as yet: • SUSY must be broken: breaking mechanism determines phenomenology • More than 100 parameters even in minimal (MSSM) models! choose a model ! • mSUGRA • New superfields in “hidden” sector • Interact gravitationally with MSSM • Soft SUSY breaking • 5 parameters at GUT scale • 1. Unified gaugino mass m1/2 • 2. Unified scalar mass m0 • 3. Ratio of H1, H2 vevs tanβ • 4. Trilinear coupling A0 • 5. Higgs mass term sgn() IFAE Seminar, Barcelona 8/05/2006

  8. Sparticles cross sections T. Plehn, PROSPINO (pb) • At LHC: pp collisions, √s = 14 TeV • Strongly interacting particles • High cross sections for gluinos andsquarksproduction •  Golden signature! T. Plehn, PROSPINO (pb) m (GeV) At Tevatron: pp collisions, √s = 1.96 TeV • Smaller cross sections • Similar environment (background ..) •  already a good place to look for new physics! m (GeV) IFAE Seminar, Barcelona 8/05/2006

  9. The Tevatron Highest-energy accelerator currently operational Peak luminosity 1.8 *1032 cm-2 s-1 Integrated luminosity/week  about 25 pb-1 CDF and D0: ~1.2 fb-1on tape Analyses shown here use 0.3 – 1.0 fb -1 IFAE Seminar, Barcelona 8/05/2006

  10. CDF and D0 in RunII CDF • Both detectors • Silicon microvertex tracker • Solenoid • High rate trigger/DAQ • Calorimeters and muons D0 L2 trigger on displaced vertices Excellent tracking resolution Excellent muon ID and acceptance Excellent tracking acceptance || < 2-3 IFAE Seminar, Barcelona 8/05/2006

  11. Missing ET 103 s (pb) Multiple jets 1 10-3 Missing ET 10-6 10-9 Phys.Rev.D59:074024,1999 300 500 700 Squarks and Gluinos at Tevatron • Squarks / gluinos pair production Signature: energetic jets + Missing transverse Energy (from the undetected LSP) • Consider mSUGRA scenario: • A0 =0, m < 0, tan b = 3 or 5 • 5 flavors degenerate IFAE Seminar, Barcelona 8/05/2006

  12. Background Background processes dominate  Need to be specifically rejected: W/Z+jets with Wl or Z, DiBoson and tt production Signatures very similar to SUSY QCDmultijets production dominates:  Missing ET due to jet energy mismeasurements • Key factors • Good understanding of Jet reconstruction and Energy calibration tools • Careful study of Monte Carlo generators (usually at LO), to be tested on data in a regime not sensitive to the signal. IFAE Seminar, Barcelona 8/05/2006

  13. QCD Background QCD multijets present Missing ET from mismeasurement of jet transverse energy • A jet is a composite object: • complex underlying physics • depends on detector properties • Corrections for different effects: • calorimeter response to hadrons (non-linear and non-compensating calorimeter) • Multiple parton interactions • Underlying event Time For Calibrations: use Ze+e- and Minimum Ionizing Particles (as J/) For Corrections: use MC simulations tuned using tracking detector  model single particle response (E/p) IFAE Seminar, Barcelona 8/05/2006

  14. CDF Jet Energy Scale Method Different correction factors: • (frel)Relative Corrections  Make response uniform in h • (MPI)Multiple Particle Interactions  Energy from different ppbar interaction • (fabs)Absolute Corrections  Calorimeter non-linear and non-compensating • (UE)Underlying Event  Energy associated with spectator partons in a hard collision CDF Run II Absolute correction factor PT jet(R) = [ PT jetraw(R)  frel (R) – MPI(R)]  fabs(R) - UE(R) Total systematic uncertainties for JES  between 2% and 3% IFAE Seminar, Barcelona 8/05/2006

  15. Data • Pythia • Herwig g (Z) + jet pT balance • g/Z + jet used for many cross checks • Also: to define JES uncertainties •  difference between data and MC • ET leading jet > 25 GeV • ET (second jet) < 3 GeV • Df (Jet-g) > 3 Sensitive to radiation effects when allow second jet: Herwig farther away from jet cone pT balance: Agreement Data/MC within 3% IFAE Seminar, Barcelona 8/05/2006

  16. Missing Transverse Energy (MET) • Calorimeter based MET: • Clean-up cuts: to remove cosmics and beam-halo bkds • must be corrected for primary vertex, additional interactions, jets (calorimeter resolution), muons (here rejected) Also: MET resolution depends on event SET  MET reconstruction more challenging in multiple-collision environment (Pile-up) Correction ~ 12% at low MET IFAE Seminar, Barcelona 8/05/2006

  17. Good shape ZONE 3 ZONE 4 Good shape ZONE 1 ZONE 2 QCD Background Estimation Define HT = E T jet2 + E T jet3 + MET Identify region in MET vs HT, dominated by jet events • Use PYTHIA in different pT bins • Compared distributions MC events to data to find NLO factor and obtained scale factor to the MC ~1.0 QCD Rejection Df (MET, jeti)> 0.7 for three leading jets IFAE Seminar, Barcelona 8/05/2006

  18. Example of one generated signal point in [mq,mg] plane ~ ~ Non-QCD Backgrounds • W en + 2/3 jets • W t n + 2 jets • Z  nn + 3 jets • Top • DiBosons • OTHER BACKGROUNDS: • Partial rejection, i.e.: • lepton-veto • Cut on M mm/ee for Z+jets • Need to estimateremaining irreducable backgrounds Missing Transverse energy From Leading Order MC  Test MC-LO tools  Perform a dedicate study to define NLO/LO scale (“k-factor”) IFAE Seminar, Barcelona 8/05/2006

  19. Top cross section Top cross section in good agreement with SM prediction:  use theory Measurements in all channels using different methods consistent. Most precise single measurement ~14%! Combined CDF measurements ~14%! IFAE Seminar, Barcelona 8/05/2006

  20. W+jets production • Background also to top and Higgs Physics • Testing ground for pQCD in multijet environment • Sample to test LO and NLO ME+PS predictions + PS LO predictions: ALPGEN+Pythia normalized to data integrated cross sections Differential cross section w.r.t. ET jet spectrum in W+n jets inclusive sample Differential cross section w.r.t. di-jet DR in the W+2 jet inclusive sample IFAE Seminar, Barcelona 8/05/2006

  21. jet parton beam remnants anti-proton proton Underlying Event (UE) Z + jets production • Same good features of W+jets: • Presence of a boson ensures high Q2 • Large BR into leptons • No New Physics expected in Z+jets • s(Z) ~ s(W) / 10, but Ze+e- cleaner  to study all aspects of hadronic collisions and relative simulation. • LO andNLOcalculations • Pythia, Herwig: •  shower, ME (Z+1 parton) • Alpgen, Sherpa, Madgraph: • ME with shower (Z + multi-parton) • MCFM: NLO ME (Z +1, 2 or 3 partons) IFAE Seminar, Barcelona 8/05/2006

  22. f=0 Z transverse plane calorimeter towers |Y| < 0.7 Z+jets event topology O.Salto et al. Integrated jet shape • Event selection • Z  e+e- with |he|< 2.8 • PTjet > 25 GeV/c, |Yjet|< 2.1 PT profile of cal towers Pythia with the Tune A is the simulation that better reproduce Jet fragmentation and Underlying Event Underlying event Jet fragmentation IFAE Seminar, Barcelona 8/05/2006

  23. PTin and PTout Sum of the components of the momentum of the track parallel and transversal to the plane defined by the Z particle direction and the beam axis. PTout Some “clean” events (very low pTtrk in the event) in the data. Well simulated by Alpgen+Herwig. Pythia puts always a certain amount of extra activity (UE tuning) • Using tracks with: • pT > 0.5 GeV/c • |z0| < 1.5 cm • d0 < 2.0 cm Z PTin PTout beam PTin G.Marchesini et al., JHEP 0108(2001) 047 hep-ph/0106278 IFAE Seminar, Barcelona 8/05/2006

  24. Inclusive pTjet distribution PT distribution of the jets in events with Z + ≥n jets Pythia Alpgen+Herwig Every MC distribution is normalized to the number of entries in the data While Pythia reproduces well the shape of the pT distributions, Alpgen+Herwig underestimates jets with low pT due to the lack of UE • For Alpgen, check impact of interfaced shower • Use Alpgen + Pythia Tune A IFAE Seminar, Barcelona 8/05/2006

  25. Alpgen+Pythia PT of the first leading jet Interfacing Alpgen ME with Pythia gives slightly better agreement in shape but further tuning could be necessary (still 10% difference) MC distribution is normalized to the number of entries in the data Issue: Alpgen + PS not have the correct absolute normalization PT of the first leading jet Data/MC ~ 1.4 Data and MC distribution normalized to relative integrated luminosity • Parton-level checks: same cross sections • for Pythia and Alpgen+Pythia (Z+1p) • Z+n partons processes simulated with • Alpgen depends on many parameter: • Shower evolution • parton-jet matching • Q2 scale  effects? IFAE Seminar, Barcelona 8/05/2006

  26. For Alpgen ME use: Q2 = mZ2 + S pT2(partons) Use 2 different scales for Pythia: Default: ~ Q2 = mZ2 + S pT2/2 New Scale: ~ Q2 = mZ2 + S pT2 Pythia default/Pythia New Scale Def/NScale PT leading jet Q2 scale effects PYTHIA (Tune A) • No negligible effects (~10%), important to evaluate NLO/LO k-factors  Use MCFM: same LO scale required at NLO • Does not explain the 1.4 factor for Alpgen + Parton Shower • Data suggest Pythia default Q2 • Tests on-going: discussions with M.Mangano foreseen next week Absolute normalization IFAE Seminar, Barcelona 8/05/2006

  27. MCFM • Parton-Level Monte Carlo: • NLO event integrator • In the Matrix Element Calculation |M|2: • Complementary approach to LO showering event generators • Give aprediction of total cross section and distribution at parton level but is not a fully implemented event generator IFAE Seminar, Barcelona 8/05/2006

  28. Jet multiplicity LO calculations → expected less events with high jet multiplicity. Pythia: simulation of the parton shower leads to an accurate number of jets. Distributions normalized to the first bin ME + PS: MADGRAPH + Pythia tree level process up to 3 partons  reproduce shape of Njet MCFM good description of the measured cross sections IFAE Seminar, Barcelona 8/05/2006

  29. Interesting to cross check other showering generators Sherpa LO MC seems to give good description in term of Jet Multiplicity although a bit high. Z+jets with Sherpa IFAE Seminar, Barcelona 8/05/2006

  30. Summary of Backgrounds • For Boson + jets and DiBoson (negligible here) • Use Alpgen + Herwig: in the process of implementing Pythia Parton Shower and tune it on Z(e+e-)+jets. • Also: many developments on going for Alpgen ME • MCFM used to define NLO/LO For top production: theoretical calculations For QCD multijets: Data Further background reduction can be achieved optimizing Signal/Bkg IFAE Seminar, Barcelona 8/05/2006

  31. mSUGRA signal X.Portell et al. Scan squark-gluino mass plane PYTHIA Tune A: - No stop and no sbottom - tan(β) = 5, Sgn μ = -1, A0 = 0 - Normalize to PROSPINO NLO σ calculation (consider each production component separately) - Compared with ISAJET: same mq,g harder distributions due to low ISR ~~ Squarks and Gluino production: • squark-antisquark • squark-squark (and c.c.) • gluino-gluino • squark-gluino (and c.c.) 4 cases IFAE Seminar, Barcelona 8/05/2006

  32. Analysis strategy In the squark-gluino mass plane, define three zones as signal regions in terms of jet ET, MET, HT  HT = ET jet1 + ET jet2 + ET jet3 • Signal region determined by optimizing S/√B • Compare number of expected events from SM background with observed events in 371 pb-1 of data Zone A: HT > 230 GeV MET > 75 GeV ET1 > 95 GeV ET2 > 55 GeV Zone B: HT > 280 GeV MET > 90 GeV ET1 > 120 GeV ET2 > 70 GeV Zone C: HT > 330 GeV MET > 120 GeV ET1 > 140 GeV ET2 > 100 GeV IFAE Seminar, Barcelona 8/05/2006

  33. Data VS MC: Zone A as example HT distribution after applying all the cuts except HT >230 GeV MET distribution after applying all the cuts except MET >75 GeV Background compositions after all cuts being applied IFAE Seminar, Barcelona 8/05/2006

  34. Systematic uncertainties • Energy scale:  1s variation • Luminosity: 6% • Renormalization scale • ISR/FSR: < 10% • PDF: largest effect (30%) on signal, 20% for QCD • Signal  PROSPINO default: • gluino-gluino  m = Mgl • gluino-squark  m= 1/2 * [Mgl + Msq] • squark-squark  m= Msq • squark-antisquark  m = Msq • Background  MCFM default: • Z  m = MZ • W  m= MW Same coupling.. Susy QCD ..as example Considered variations of twice and half the scale to calculate the uncertainty:  effect of 20% (10% on QCD) IFAE Seminar, Barcelona 8/05/2006

  35. PDF uncertainties O.Norniella et al. Used Hessian method for PDF in PROSPINO and MCFM: at LO CTEQ6.1L, at NLO CTEQ6.1M  important for signal due to high-x gluon contribution Forward jets measurements help to distinguish between new physics and PDF if any excess in the central region Big uncertainty for high-x gluon PDF Uncertainty on gluon PDF (from CTEQ6) Five regions in jet rapidity explored IFAE Seminar, Barcelona 8/05/2006

  36. Results Good consistency between Data and SM Background XY view of event with large MET ET(1st) = 172 GeV HT = ET(1st) + ET(2nd) + ET(3rd) = 404 GeV ET(2nd) = 153 GeV ET (3rd) = 80 GeV ET(4th) = 65 GeV Missing ET = 223 GeV IFAE Seminar, Barcelona 8/05/2006

  37. Exclusion limits • No excess found w.r.t. Standard Model background  estabilish exclusion limits on gluino and squark production • Bayesian technique applied to exclude a range of mq, mg to a 95% C.L. • Correlated (JES, PDF, Luminosity) and uncorrelated (statistic, renormalization scale, ISR/FSR) uncertainties also included in the limit calculation • For each point of the three different zones, expected and observed s95 is found. Cross section along squark masses (gluino mass = constant) Cross section along the diagonal (gluino mass ~ squark mass) IFAE Seminar, Barcelona 8/05/2006

  38. Limits: CDF and D0 • On going development: • Use >1 fb-1 of CDF data • Extend analysis to 4 jets • to further constrain high • squark mass region. IFAE Seminar, Barcelona 8/05/2006

  39. Large Extra –Dimensions:  n extra dimensions (≥ 2) compactified  Effective Planck scale: Non Susy searches: an example M2Planck ~ Rn(MD)n+2 with MD ~ 1 TeV • Possible signatures: • Virtual Graviton exchange: • Excess in Dilepton mass • Direct production: • MET + recoiling jet (monojet) • 265 ±30 events predicted • 263 events observed No excess observed IFAE Seminar, Barcelona 8/05/2006

  40. Conclusions • In 2005, Tevatron achieved the 1 fb-1 goal • 1.2 fb-1 on tape ready for data analyses! • Very rich searches physics program ongoing at CDF • Tevatron is currently one of the best places to search for new physics and test MC tools Very important for the LHC • Squarks and gluinos pair production constitutes the Golden Signature for the LHC • Parallel efforts on Di-Leptons, Z+jets and SUSY are mandatory. • A lot of work on different (but correlated!) topics is necessary. Thanks to the whole CDF-IFAE team, especially to Olga Norniella, Xavier Portell, Oriol Salto! IFAE Seminar, Barcelona 8/05/2006

  41. Back up

  42. Calorimeter calibration Use Ze+e- mass peak stability • Use Minimum Ionizing Particles: • J/ and W • Peak HAD and EM calorimeter • Time dependence IFAE Seminar, Barcelona 8/05/2006

  43. Tuning of CDF simulation Measure p of particles using tracking, E from HAD and EM calorimeter  Use isolated tracks from Minimum Bias data  E/p used to tune simulation (GFLASH parametrization) Plug calorimeter Central calorimeter IFAE Seminar, Barcelona 8/05/2006

  44. Z-jet pT balance • Selection • two e(m) with ET>18 GeV (pT>20 GeV) • 76 < M ee(mm) < 106 GeV • ET leading jet > 25 GeV • ET (second jet) < 3 GeV • Df (Jet-Z) > 3 Similar Herwig behaviour for Z+jet w.r.t. g+jet but less visible IFAE Seminar, Barcelona 8/05/2006

  45. JES Systematic uncertainties Total systematic uncertainties for JES  between 2 and 3% as a function of corrected transverse jet momentum IFAE Seminar, Barcelona 8/05/2006

  46. Possible effects of low ISR … Preliminary MC studies (1999) at the LHC suggested that SUSY could be discovered via the jet+MET channel within weeks after LHC started 1999 TDR 2005 evaluation Signal uses now Pythia (IsaSUGRA for ME) Additional changes in tools Used on the background  Very dependent from MC!! IFAE Seminar, Barcelona 8/05/2006

  47. S/√B Optimization: Example Example of HT optimization: s56 is a low squark mass point and s61 is a high squark mass point. A common threshold is taken for all the points in the zone: 280 GeV. Mgl ~ 295 GeV/c2 IFAE Seminar, Barcelona 8/05/2006

  48. Exclusion limits: more details • The program needs Nobs events, Nexp backgrounds and the Lum.*Efficiency for the signal point. • Include as uncorrelated uncertainties: statistical uncert. and ISR/FSR • Include as correlated uncertainties all the rest: JES, PDF, Luminosity, Renormalization Scale. • The theoretical uncertainties for the sigma cross section (PDF and Renorm.) are also introduced in the program. Rescale the PDFs to 1 sigma uncertainty (Hessian method gives 1.64 sigma). • For the expected limits, perform pseudo-experiments with a Poisson fluctuation using the formula: For large number of expected events (>10) we can use the approximation Nobs=Nexp (rounding up) • Calculate the expected sigma95 per each point for the three different set of cuts. • Assign to each point the set of cuts that maximizes the exclusion using the expected sigma95 (in this way, each point belongs only to a zone) • With this definition, the observed sigma95 is calculated. Knowing sigma95 and the masses of the points, we can perform a linear interpolation between them. IFAE Seminar, Barcelona 8/05/2006

  49. Linear interpolation: Example s95 = 0.163 s = 0.225 s95 = 0.124 s = 0.120 s86 s80 In this situation, s80 is excluded and s86 it is not. To find exactly where the exclusion line is, we define: Note that r*=1 marks the position of the excl. line Assuming one dimension only (take Msq) we interpolated linearly: When r*=1 we can find the mass of the point in the line, m*. Same procedure to find gluino mass. IFAE Seminar, Barcelona 8/05/2006

  50. Proposed as a solution to the hierarchy problem Extra dimensions are compactified or inaccessible to some part of SM Model of Arkani-Hamed, Dimopoulos, and Dvali (ADD) n extra dimensions (≥ 2) compactified at radius R SM constrained to a 4-d brane in higher dimensional space  Gravity exists in (4+n)-d “bulk” Effective Planck scale: Large Extra Dimensions (ADD scenario) M2Planck ~ Rn(MD)n+2 ,MD ~ 1 TeV • Gravitons appear to have mass, m2=m02+p2 • Tower of KK modes with splittings m ~ 1/R • Each mode couples with strength, MPl-1, but there are many: cross section summed over all modes IFAE Seminar, Barcelona 8/05/2006

More Related