- Email: [email protected]
Search for pair production of the scalar top quark in the electron + muon final state
Physics Letters B 696 (2011) 321–327
Contents lists available at ScienceDirect
Physics Letters B www.elsevier.com/locate/physletb
Search for pair production of the scalar top quark in the electron + muon final state D0 Collaboration 1 V.M. Abazov ai , B. Abbott bu , M. Abolins bj , B.S. Acharya ac , M. Adams av , T. Adams at , G.D. Alexeev ai , G. Alkhazov am , A. Alton bi,2 , G. Alverson bh , G.A. Alves b , L.S. Ancu ah , M. Aoki au , Y. Arnoud n , M. Arov be , A. Askew at , B. Åsman an , O. Atramentov bm , C. Avila h , J. BackusMayes cb , F. Badaud m , L. Bagby au , B. Baldin au , D.V. Bandurin at , S. Banerjee ac , E. Barberis bh , P. Baringer bc , J. Barreto b , J.F. Bartlett au , U. Bassler r , V. Bazterra av , S. Beale f , A. Bean bc , M. Begalli c , M. Begel bs , C. Belanger-Champagne an , L. Bellantoni au , S.B. Beri aa , G. Bernardi q , R. Bernhard v , I. Bertram ao , M. Besançon r , R. Beuselinck ap , V.A. Bezzubov al , P.C. Bhat au , V. Bhatnagar aa , G. Blazey aw , S. Blessing at , K. Bloom bl , A. Boehnlein au , D. Boline br , T.A. Bolton bd , E.E. Boos ak , G. Borissov ao , T. Bose bg , A. Brandt bx , O. Brandt w , R. Brock bj , G. Brooijmans bp , A. Bross au , D. Brown q , J. Brown q , X.B. Bu g , D. Buchholz ax , M. Buehler ca , V. Buescher x , V. Bunichev ak , S. Burdin ao,3 , T.H. Burnett cb , C.P. Buszello ap , B. Calpas o , E. Camacho-Pérez af , M.A. Carrasco-Lizarraga af , B.C.K. Casey au , H. Castilla-Valdez af , S. Chakrabarti br , D. Chakraborty aw , K.M. Chan ba , A. Chandra bz , G. Chen bc , S. Chevalier-Théry r , D.K. Cho bw , S.W. Cho ae , S. Choi ae , B. Choudhary ab , T. Christoudias ap , S. Cihangir au , D. Claes bl , J. Clutter bc , M. Cooke au , W.E. Cooper au , ad ´ M. Corcoran bz , F. Couderc r , M.-C. Cousinou o , A. Croc r , D. Cutts bw , M. Cwiok , A. Das ar , G. Davies ap , bx ah af r au K. De , S.J. de Jong , E. De La Cruz-Burelo , F. Déliot , M. Demarteau , R. Demina bq , D. Denisov au , S.P. Denisov al , S. Desai au , K. DeVaughan bl , H.T. Diehl au , M. Diesburg au , A. Dominguez bl , T. Dorland cb , A. Dubey ab , L.V. Dudko ak , D. Duggan bm , A. Duperrin o , S. Dutt aa , A. Dyshkant aw , M. Eads bl , D. Edmunds bj , J. Ellison as , V.D. Elvira au , Y. Enari q , S. Eno bf , H. Evans ay , A. Evdokimov bs , V.N. Evdokimov al , G. Facini bh , T. Ferbel bf,bq , F. Fiedler x , F. Filthaut ah , W. Fisher bj , H.E. Fisk au , M. Fortner aw , H. Fox ao , S. Fuess au , T. Gadfort bs , A. Garcia-Bellido bq , V. Gavrilov aj , P. Gay m , W. Geist s , W. Geng o,bj , D. Gerbaudo bn , C.E. Gerber av , Y. Gershtein bm , G. Ginther au,bq , G. Golovanov ai , A. Goussiou cb , P.D. Grannis br , S. Greder s , H. Greenlee au , Z.D. Greenwood be , E.M. Gregores d , G. Grenier t , Ph. Gris m , J.-F. Grivaz p , A. Grohsjean r , S. Grünendahl au , M.W. Grünewald ad , F. Guo br , J. Guo br , G. Gutierrez au , P. Gutierrez bu , A. Haas bp,4 , S. Hagopian at , J. Haley bh , L. Han g , K. Harder aq , A. Harel bq , J.M. Hauptman bb , J. Hays ap , T. Head aq , T. Hebbeker u , D. Hedin aw , H. Hegab bv , A.P. Heinson as , U. Heintz bw , C. Hensel w , I. Heredia-De La Cruz af , K. Herner bi , G. Hesketh bh , M.D. Hildreth ba , R. Hirosky ca , T. Hoang at , J.D. Hobbs br , B. Hoeneisen l , M. Hohlfeld x , S. Hossain bu , Z. Hubacek j , N. Huske q , V. Hynek j , I. Iashvili bo , R. Illingworth au , A.S. Ito au , S. Jabeen bw , M. Jaffré p , S. Jain bo , D. Jamin o , R. Jesik ap , K. Johns ar , M. Johnson au , D. Johnston bl , A. Jonckheere au , P. Jonsson ap , J. Joshi aa , A. Juste au,5 , K. Kaadze bd , E. Kajfasz o , D. Karmanov ak , P.A. Kasper au , I. Katsanos bl , R. Kehoe by , S. Kermiche o , N. Khalatyan au , A. Khanov bv , A. Kharchilava bo , Y.N. Kharzheev ai , D. Khatidze bw , M.H. Kirby ax , J.M. Kohli aa , A.V. Kozelov al , J. Kraus bj , A. Kumar bo , A. Kupco k , T. Kurˇca t , V.A. Kuzmin ak ,
1 With visitors from Augustana College, Sioux Falls, SD, USA, The University of Liverpool, Liverpool, UK, SLAC, Menlo Park, CA, USA, ICREA/IFAE, Barcelona, Spain, Centro de Investigacion en Computacion – IPN, Mexico City, Mexico, ECFM, Universidad Autonoma de Sinaloa, Culiacán, Mexico, and Universität Bern, Bern, Switzerland. 2 Augustana College, Sioux Falls, SD, USA. 3 The University of Liverpool, Liverpool, United Kingdom.
0370-2693/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.physletb.2010.12.052
322
D0 Collaboration / Physics Letters B 696 (2011) 321–327
J. Kvita i , S. Lammers ay , G. Landsberg bw , P. Lebrun t , H.S. Lee ae , S.W. Lee bb , W.M. Lee au , J. Lellouch q , L. Li as , Q.Z. Li au , S.M. Lietti e , J.K. Lim ae , D. Lincoln au , J. Linnemann bj , V.V. Lipaev al , R. Lipton au , Y. Liu g , Z. Liu f , A. Lobodenko am , M. Lokajicek k , P. Love ao , H.J. Lubatti cb , R. Luna-Garcia af,6 , A.L. Lyon au , A.K.A. Maciel b , D. Mackin bz , R. Madar r , R. Magaña-Villalba af , S. Malik bl , V.L. Malyshev ai , Y. Maravin bd , J. Martínez-Ortega af , R. McCarthy br , C.L. McGivern bc , M.M. Meijer ah , A. Melnitchouk bk , D. Menezes aw , P.G. Mercadante d , M. Merkin ak , A. Meyer u , J. Meyer w , N.K. Mondal ac , G.S. Muanza o , M. Mulhearn ca , E. Nagy o , M. Naimuddin ab , M. Narain bw , R. Nayyar ab , H.A. Neal bi , J.P. Negret h , P. Neustroev am , S.F. Novaes e , T. Nunnemann y , G. Obrant am , J. Orduna af , N. Osman ap , J. Osta ba , G.J. Otero y Garzón a , M. Owen aq , M. Padilla as , M. Pangilinan bw , N. Parashar az , V. Parihar bw , S.K. Park ae , J. Parsons bp , R. Partridge bw,4 , N. Parua ay , A. Patwa bs , B. Penning au , M. Perfilov ak , K. Peters aq , Y. Peters aq , G. Petrillo bq , P. Pétroff p , R. Piegaia a , J. Piper bj , M.-A. Pleier bs , P.L.M. Podesta-Lerma af,7 , V.M. Podstavkov au , M.-E. Pol b , P. Polozov aj , A.V. Popov al , M. Prewitt bz , D. Price ay , S. Protopopescu bs , J. Qian bi , A. Quadt w , B. Quinn bk , M.S. Rangel b , K. Ranjan ab , P.N. Ratoff ao , I. Razumov al , P. Renkel by , P. Rich aq , M. Rijssenbeek br , I. Ripp-Baudot s , F. Rizatdinova bv , M. Rominsky au , C. Royon r , P. Rubinov au , R. Ruchti ba , G. Safronov aj , G. Sajot n , A. Sánchez-Hernández af , M.P. Sanders y , B. Sanghi au , A.S. Santos e , G. Savage au , L. Sawyer be , T. Scanlon ap , R.D. Schamberger br , Y. Scheglov am , H. Schellman ax , T. Schliephake z , S. Schlobohm cb , C. Schwanenberger aq , R. Schwienhorst bj , J. Sekaric bc , H. Severini bu , E. Shabalina w , V. Shary r , A.A. Shchukin al , R.K. Shivpuri ab , V. Simak j , V. Sirotenko au , P. Skubic bu , P. Slattery bq , D. Smirnov ba , K.J. Smith bo , G.R. Snow bl , J. Snow bt , S. Snyder bs , S. Söldner-Rembold aq , L. Sonnenschein u , A. Sopczak ao , M. Sosebee bx , K. Soustruznik i , B. Spurlock bx , J. Stark n , V. Stolin aj , D.A. Stoyanova al , E. Strauss br , M. Strauss bu , D. Strom av , L. Stutte au , P. Svoisky bu , M. Takahashi aq , A. Tanasijczuk a , W. Taylor f , M. Titov r , V.V. Tokmenin ai , D. Tsybychev br , B. Tuchming r , C. Tully bn , P.M. Tuts bp , L. Uvarov am , S. Uvarov am , S. Uzunyan aw , R. Van Kooten ay , W.M. van Leeuwen ag , N. Varelas av , E.W. Varnes ar , I.A. Vasilyev al , P. Verdier t , L.S. Vertogradov ai , M. Verzocchi au , M. Vesterinen aq , D. Vilanova r , P. Vint ap , P. Vokac j , H.D. Wahl at , M.H.L.S. Wang bq , J. Warchol ba , G. Watts cb , M. Wayne ba , M. Weber au,8 , L. Welty-Rieger ax , M. Wetstein bf , A. White bx , D. Wicke x , M.R.J. Williams ao , G.W. Wilson bc , S.J. Wimpenny as , M. Wobisch be , D.R. Wood bh , T.R. Wyatt aq , Y. Xie au , C. Xu bi , S. Yacoob ax , R. Yamada au , W.-C. Yang aq , T. Yasuda au , Y.A. Yatsunenko ai , Z. Ye au , H. Yin g , K. Yip bs , H.D. Yoo bw , S.W. Youn au , J. Yu bx , S. Zelitch ca , T. Zhao cb , B. Zhou bi , J. Zhu bi , M. Zielinski bq , D. Zieminska ay , L. Zivkovic bp a
Universidad de Buenos Aires, Buenos Aires, Argentina LAFEX, Centro Brasileiro de Pesquisas Físicas, Rio de Janeiro, Brazil c Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil d Universidade Federal do ABC, Santo André, Brazil e Instituto de Física Teórica, Universidade Estadual Paulista, São Paulo, Brazil f Simon Fraser University, Vancouver, British Columbia, and York University, Toronto, Ontario, Canada g University of Science and Technology of China, Hefei, People’s Republic of China h Universidad de los Andes, Bogotá, Colombia i Charles University, Faculty of Mathematics and Physics, Center for Particle Physics, Prague, Czech Republic j Czech Technical University in Prague, Prague, Czech Republic k Center for Particle Physics, Institute of Physics, Academy of Sciences of the Czech Republic, Prague, Czech Republic l Universidad San Francisco de Quito, Quito, Ecuador m LPC, Université Blaise Pascal, CNRS/IN2P3, Clermont, France n LPSC, Université Joseph Fourier Grenoble 1, CNRS/IN2P3, Institut National Polytechnique de Grenoble, Grenoble, France o CPPM, Aix-Marseille Université, CNRS/IN2P3, Marseille, France p LAL, Université Paris-Sud, CNRS/IN2P3, Orsay, France q LPNHE, Universités Paris VI and VII, CNRS/IN2P3, Paris, France r CEA, Irfu, SPP, Saclay, France s IPHC, Université de Strasbourg, CNRS/IN2P3, Strasbourg, France t IPNL, Université Lyon 1, CNRS/IN2P3, Villeurbanne, France and Université de Lyon, Lyon, France u III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany v Physikalisches Institut, Universität Freiburg, Freiburg, Germany w II. Physikalisches Institut, Georg-August-Universität Göttingen, Göttingen, Germany x Institut für Physik, Universität Mainz, Mainz, Germany y Ludwig-Maximilians-Universität München, München, Germany z Fachbereich Physik, Bergische Universität Wuppertal, Wuppertal, Germany aa Panjab University, Chandigarh, India ab Delhi University, Delhi, India ac Tata Institute of Fundamental Research, Mumbai, India ad University College Dublin, Dublin, Ireland ae Korea Detector Laboratory, Korea University, Seoul, Republic of Korea af CINVESTAV, Mexico City, Mexico ag FOM – Institute NIKHEF and University of Amsterdam/NIKHEF, Amsterdam, The Netherlands b
D0 Collaboration / Physics Letters B 696 (2011) 321–327
323
ah
Radboud University Nijmegen/NIKHEF, Nijmegen, The Netherlands Joint Institute for Nuclear Research, Dubna, Russia aj Institute for Theoretical and Experimental Physics, Moscow, Russia ak Moscow State University, Moscow, Russia al Institute for High Energy Physics, Protvino, Russia am Petersburg Nuclear Physics Institute, St. Petersburg, Russia an Stockholm University, Stockholm and Uppsala University, Uppsala, Sweden ao Lancaster University, Lancaster LA1 4YB, United Kingdom ap Imperial College London, London SW7 2AZ, United Kingdom aq The University of Manchester, Manchester M13 9PL, United Kingdom ar University of Arizona, Tucson, AZ 85721, USA as University of California Riverside, Riverside, CA 92521, USA at Florida State University, Tallahassee, FL 32306, USA au Fermi National Accelerator Laboratory, Batavia, IL 60510, USA av University of Illinois at Chicago, Chicago, IL 60607, USA aw Northern Illinois University, DeKalb, IL 60115, USA ax Northwestern University, Evanston, IL 60208, USA ay Indiana University, Bloomington, IN 47405, USA az Purdue University Calumet, Hammond, IN 46323, USA ba University of Notre Dame, Notre Dame, IN 46556, USA bb Iowa State University, Ames, IA 50011, USA bc University of Kansas, Lawrence, KS 66045, USA bd Kansas State University, Manhattan, KS 66506, USA be Louisiana Tech University, Ruston, LA 71272, USA bf University of Maryland, College Park, MD 20742, USA bg Boston University, Boston, MA 02215, USA bh Northeastern University, Boston, MA 02115, USA bi University of Michigan, Ann Arbor, MI 48109, USA bj Michigan State University, East Lansing, MI 48824, USA bk University of Mississippi, University, MS 38677, USA bl University of Nebraska, Lincoln, NE 68588, USA bm Rutgers University, Piscataway, NJ 08855, USA bn Princeton University, Princeton, NJ 08544, USA bo State University of New York, Buffalo, NY 14260, USA bp Columbia University, New York, NY 10027, USA bq University of Rochester, Rochester, NY 14627, USA br State University of New York, Stony Brook, NY 11794, USA bs Brookhaven National Laboratory, Upton, NY 11973, USA bt Langston University, Langston, OK 73050, USA bu University of Oklahoma, Norman, OK 73019, USA bv Oklahoma State University, Stillwater, OK 74078, USA bw Brown University, Providence, RI 02912, USA bx University of Texas, Arlington, TX 76019, USA by Southern Methodist University, Dallas, TX 75275, USA bz Rice University, Houston, TX 77005, USA ca University of Virginia, Charlottesville, VA 22901, USA cb University of Washington, Seattle, WA 98195, USA ai
a r t i c l e
i n f o
Article history: Received 27 September 2010 Received in revised form 21 November 2010 Accepted 26 December 2010 Available online 31 December 2010 Editor: H. Weerts Keywords: Supersymmetry Squarks Supersymmetric top
a b s t r a c t We report the result of a search for the pair production of the lightest supersymmetric partner of the top quark (t˜1 ) in p p¯ collisions at a center-of-mass energy of 1.96 TeV at the Fermilab Tevatron collider corresponding to an integrated luminosity of 5.4 fb−1 . The scalar top quarks are assumed to decay into a b quark, a charged lepton, and a scalar neutrino (ν˜ ), and the search is performed in the electron plus muon final state. No significant excess of events above the standard model prediction is detected, and substantially improved exclusion limits at the 95% C.L. are set in the the (M t˜1 , M ν˜ ) mass plane. © 2010 Elsevier B.V. All rights reserved.
Supersymmetric theories [1] predict the existence of scalar partners for each of the standard model (SM) fermions. In the minimal supersymmetric standard model (MSSM) [2], the mixing between the chiral states of the scalar partners of the SM fermions
4 5 6 7 8
SLAC, Menlo Park, CA, USA. ICREA/IFAE, Barcelona, Spain. Centro de Investigacion en Computacion – IPN, Mexico City, Mexico. ECFM, Universidad Autonoma de Sinaloa, Culiacán, Mexico. Universität Bern, Bern, Switzerland.
is greatest for the partners of the top quark due to its large Yukawa coupling [3]. Thus, it is possible that the scalar top quark (t˜1 ) is the lightest squark and has the largest production cross section. If R-parity [1] is conserved, then scalar top quarks would be produced by p p¯ collisions in pairs with the dominant processes being quark–antiquark annihilation and gluon fusion [3]. In this Letter we report on a search for the production of t˜1 t¯˜1 ¯ ± μ∓ ν˜ ν˜ final state, where ν˜ is the sneutrino, the pairs in the bbe supersymmetric partner of the neutrino. In this search, we assume that the two-body decays of t˜1 to t χ˜ 10 , t g˜ , and bχ˜ 1+ are kine-
324
D0 Collaboration / Physics Letters B 696 (2011) 321–327
matically forbidden, where χ˜ 10 is the neutralino, g˜ is the gluino, and χ˜ 1+ is the chargino. For this scenario, the three-body decay t˜1 → blν˜ is predicted to dominate over the flavor changing twobody decay t˜1 → c χ˜ 10 [4]. We assume that the t˜1 has a 100% branching fraction to blν˜ , with equal fraction to each lepton type, that R-parity is conserved, and that the sneutrino is the lightest supersymmetric particle or decays invisibly into a neutrino and a neutralino. This analysis uses data corresponding to an integrated luminosity of 5.4 fb−1 collected using √ the D0 detector operating at the Fermilab Tevatron collider at s = 1.96 TeV. The data were collected from April 2002 through June 2009. The D0 Collaboration has previously searched [5–7] for top squark pair production in the final states bb¯ ± ∓ ν˜ ν˜ where the lepton pair is ee, μμ, or e μ. Two of these earlier searches used subsets of this data set corresponding to integrated luminosities of 0.43 fb−1 and 1.1 fb−1 , while the earliest search used data from the Tevatron Run I, corresponding to an integrated luminosity of 108 pb−1 . Searches for top squark pair production in the bb¯ ± ∓ ν˜ ν˜ final states have also been reported by the CDF Collaboration [8] and by the ALEPH, L3, and OPAL Collaborations [9]. The main components of the D0 detector [10] include a central tracking system located inside a 2 T superconducting solenoid. The inner-most tracking element is the silicon microstrip tracker (SMT), followed by a scintillating fiber tracker. These two detectors together measure the momenta of charged particles. The tracking system provides full coverage in the azimuthal (φ ) direction for |η| < 2, where the pseudorapidity η is defined as η = − ln(tan θ/2) and θ is the polar angle with respect to the proton beam direction. Outside the solenoid is the uranium/liquid argon calorimeter which is divided into a central calorimeter and two end-cap calorimeters. Each of these three calorimeters have electromagnetic layers followed by hadronic layers. The outermost component of the detector is the muon system, which consists of proportional drift tubes and scintillator trigger counters, followed by 1.8 T iron toroids and two additional layers of drift tubes and scintillators. Events are selected for offline analysis by a three-level trigger system. All events are required to pass one of a suite of single-electron triggers or single-muon triggers using information from the tracking system, the calorimeter, and the muon system. For each event, a primary p p¯ interaction vertex is defined. If more than one vertex is reconstructed, the primary vertex is taken to be the vertex least consistent with originating from a soft collision. The location of the primary vertex along the beam direction is required to be within ±60 cm of the detector center. Jets are reconstructedusing the D0 Run II cone algorithm [11] with cone of radius R ≡ (φ)2 + ( y )2 < 0.5, where y is the rapidity. Jet energies are calibrated using the standard D0 procedure [12]. A jet is retained in an event if it has transverse energy E T > 20 GeV, |η| < 2.5, and if R(jet, electron) = (φ)2 + (η)2 > 0.5. No requirement on the number of jets is applied. Electrons are required to have transverse momentum p eT > 15 GeV and |η| < 1.1. They are required to be isolated, defined as having [ E tot (0.4) − E EM (0.2)]/ E EM (0.2) < 0.15 where E tot (0.4) is the total calorimeter energy in a cone of radius R = 0.4 and E EM (0.2) is the electromagnetic energy in a cone of radius R = 0.2. In addition, the shower development in the calorimeter is required to be consistent with that of an electromagnetic shower both transversely and longitudinally. An eight-variable likelihood function is constructed to further distinguish between electromagnetic and hadronic showers. The output of this function ranges from 0 to 1, and electrons are required to have a likelihood value greater than 0.85. Electromagnetic showers associated with electrons are also required to match a central track within η < 0.05 and φ < 0.05 of the electromagnetic cluster.
μ
Muons are required to have transverse momentum p T > 10 GeV and |η| < 2. They are required to have both drift tube and scintillator hits in the muon system and to match a track in the central tracker. If the central track includes hits in the SMT, the distance of closest approach (DCA) between the muon track and the primary vertex is required to be less than 0.02 cm. If there are no SMT hits, then the DCA is required to be less than 0.2 cm. Muons are also required to satisfy isolation requirements in both the calorimeter and the central tracker. For the calorimeter isolation, the transverse energy in the cone R < 0.5 around the muon μ track divided by p T must be less than 0.15. For the central tracker isolation, the sum of the transverse energy of the tracks in the μ hollow cone 0.1 < R < 0.5 divided by p T must be less than 0.15. Events are required to have exactly one electron and one muon with opposite charge and to have a minimum separation between the electron and the muon R(e , μ) > 0.5. The missing transverse energy (/ E T ) is calculated from the calorimeter energy corrected for the jet and electron calibrations. It is then adjusted to account for the transverse momentum of the muon. All retained events are required to have / E T > 7 GeV. We refer to this preliminary set of selection requirements as the preselection. Signal Monte Carlo (MC) events are generated in a 2-D grid, i.e., for t˜1 masses ranging from 100 GeV to 240 GeV, and for ν˜ masses ranging from 40 GeV to 140 GeV, each in 10 GeV steps. For each point, the MSSM decay parameters are calculated with suspect [13] and sdecay [14]. madgraph/madevent [15] is used to generate four vectors for the signal events with pythia [16] providing the showering and hadronization. The next-to-leading order (NLO) cross section for t˜1 pair production is calculated by prospino 2.0 [17] with the CTEQ6.1M [18] parton distribution functions (PDFs). The calculations are performed with the factorization and renormalization scales set to one, one half, and two times the t˜1 mass to determine the nominal value and the negative and positive uncertainties. The scale factor uncertainties are combined quadratically with the PDF uncertainties [18,19] to give the total theoretical uncertainties for the signal cross sections. The dominant SM backgrounds for this decay are Z /γ ∗ → τ τ with τ → lν ; diboson production including W W , W Z , and Z Z ; top quark pairs; W + jets; and instrumental background coming from multijet (MJ) processes where jets are misidentified as electrons or contain muons that pass the isolation criterion and with / E T arising from energy mismeasurement. All the background processes in this analysis except for MJ are modeled using MC simulation. Vector boson pair production is simulated with pythia, while all other backgrounds are simulated at the parton level with alpgen [20], with pythia used for hadronization and showering. The detector geometry and response are fully simulated for each MC event using geant [21] based software, and in order to simulate detector noise and multiple p p¯ interaction effects, each MC event is overlayed with a data event from a randomly chosen p p¯ crossing. MC correction factors determined from data are applied to make distributions consistent between data and MC. These corrections include factors for the luminosity profile, beam spot position, muon and electron identification efficiencies, boson transverse momentum, and jet, electron, and muon energy resolutions. The MJ background is estimated from a selection of data events not overlapping with the search sample which is selected by inverting the electron likelihood and muon isolation requirements. Jets fake isolated leptons due to detector or jet fragmentation effects. It is therefore expected that inverting the electron likelihood or the muon isolation requirements does not significantly change the event kinematics and that this background sample is a good representation of the MJ events in the search sample. Because most same-sign di-lepton events come from MJ processes, we obtain the
D0 Collaboration / Physics Letters B 696 (2011) 321–327
325
Fig. 1. (Color online.) E T versus φ(e , μ) for Z /γ ∗ → τ τ MC events. Events inside the black box in the lower right corner are removed. The sizes of the boxes are proportional to the number of events in the underlying cell.
/
Fig. 3. (Color online.) Distribution of δ Z versus δ W W for (a) the small- M signal benchmark, ( M t˜1 , M ν˜ ) = (110 GeV, 90 GeV), MC events, (b) Z /γ ∗ → τ τ MC events, and (c) W W MC events. The top right quadrant is used in the limit-setting procedure.
/
Fig. 2. (Color online.) Distribution of (a) E T , (b) electron p T , and (c) muon p T comparing data and all background processes after Selection 1. The thick dashed and thick solid lines represent the small- M and large- M signal benchmarks, respectively.
normalization factor for the MJ background by taking the ratio of the number of same-sign events that pass the likelihood and isolation requirements to the number of same-sign events that fail these requirements. To remove W + jet events from the MJ sameE T < 20 GeV, sign sample, we make the additional requirement / E T . We also correct this since W + jets events tend to have large /
ratio for non-MJ SM processes that produce like-sign leptons, using the MC samples. Data events are required to satisfy at least one of a suite of single-electron or single-muon triggers. The efficiency of the combination of the single-electron triggers is measured using a subset of the search sample in which at least one of the single-muon triggers fired, and vice-versa for the single muon triggers. The combination of these two efficiencies, taken to be the overall trigger efficiency, is then applied as a correction to the MC samples. The trigger efficiency averaged over all background MC events that pass the preselection is 95%. The mass difference, M = M t˜1 − M ν˜ , determines the kinematics of the final state. A larger M will lead, on average, to larger /E T , larger jet energy, and higher p T charged leptons. We divide the range of M into a “large- M” region ( M > 60 GeV) and “small M” region ( M < 60 GeV). To illustrate these regions, we have chosen two benchmark points, ( M t˜1 , M ν˜ ) = (200 GeV, 100 GeV) and (110 GeV, 90 GeV), which will be referred to as the large M and small- M benchmarks, respectively. Since there are many
326
D0 Collaboration / Physics Letters B 696 (2011) 321–327
Table 1 Expected numbers of background and signal events, and the number of events observed in the data at each stage of the analysis. The errors include statistical and systematic uncertainties. For M < 60 GeV, Selection 2 is δt t¯( M = 20 GeV) > 0, and for M 60 GeV, Selection 2 is δ Z ( M = 100 GeV) > 0. Sample
Preselection events
Selection 1 events
Selection 2
δt t¯ > 0
δZ > 0
events
events
Z → ττ Z → μμ Z → ee WZ WW ZZ t t¯ W MJ
1516 ± 150 33.1 ± 4.7 23.2 ± 3.9 12.7 ± 1.6 295 ± 32 2.2 ± 0.3 25 206+ −28 70 ± 9.2 33 ± 9.2
582 ± 61 22.9 ± 3.7 16.6 ± 3.0 12.0 ± 1.5 268 ± 30 2.0 ± 0.3 25 204+ −28 67.5 ± 9.0 19.7 ± 5.5
515 ± 54 16.3 ± 2.9 10.2 ± 2.2 6.3 ± 0.8 157 ± 18 1.0 ± 0.15 0 .8 6.6+ −0.9 55 ± 7.7 18.4 ± 5.1
17.3 ± 2.0 5.5 ± 1.2 0.1 ± 2.3 9.3 ± 1.2 237 ± 26 1.1 ± 0.16 22 179+ −24 53 ± 7.4 1.3 ± 0.35
Background total Data
160 2191+ −159 2168
1195 ± 73 1147
785 ± 57 776
513 ± 37 472
35 ± 5.6
25.5 ± 4.2
23.8 ± 3.9
–
55 ± 9.3
53.4 ± 9.0
–
51.8 ± 8.7
Small- M benchmark (110 GeV, 90 GeV) Large- M benchmark (200 GeV, 100 GeV)
signal points and their characteristics differ significantly, the analysis strategy is to optimize the signal selection as a function of M. For all values of M, the largest background after preselection is Z /γ ∗ → τ τ . A two-dimensional plot of the azimuthal angle between the electron and muon, φ(e , μ), vs. / E T for Z /γ ∗ → τ τ MC events is shown in Fig. 1. The two leptons from Z /γ ∗ → τ τ tend to be back-to-back in φ , and tend to have low / E T . We therefore reject events in which φ(e , μ) > 2.8 and / E T < 20 GeV and label this as “Selection 1”. Fig. 2 compares / E T , electron p T , and muon p T of the data and the sum of all backgrounds at this stage of the analysis. The agreement confirms our understanding of the SM backgrounds, of the trigger efficiency, and of other MC corrections. After Selection 1, the three largest backgrounds are Z /γ ∗ → τ τ , W W , and t t¯ production. To discriminate these backgrounds from signal we create for each of them a composite discriminant variable from a linear combination of kinematic quantities. We use the R software package [22] to calculate the maximum likelihood coefficients β for a generalized linear model (GLM) [23] of the form
δ A = ln
μ 1−μ
= β0 + β · X
(1)
to discriminate between signal and a specific background source A. Here, μ is the probability that an event is signal, β0 is a con is the vector of event stant, β is the vector of coefficients, and X kinematic variables. By construction, δ A = 0 when μ = 0.5, and signal-like events have positive δ A. The discriminant δ Z is constructed to separate signal from Z /γ ∗ → τ τ background, using an equal number of signal and Z /γ ∗ → τ τ MC events to de we use the following termine the coefficients β0 and β. For X μ variables: ln(/ E T ), ln( p T ), ln( p eT ), φ(e , μ), φ(e , / E T ), φ(μ, / E T ), and φ(e , / E T ) × φ(μ, / E T ). For each value of M, ranging from 20 to 200 GeV, we use the same variables with re-optimized coefficients. We use a similar method for creating the discriminants δ W W and δt t¯ to separate signal from W W and t t¯ backgrounds. μ For δ W W we use the variables ln(/ E T ), ln( p T ), ln( p eT ), number of jets, φ(e , μ), and ln( W W tag ). Here W W tag is the magnitude μ of the vector sum of p eT , p T , and / E T , which should be close to μ E T ), ln( p T ), zero for W W events. For δt t¯, we use the variables ln(/ ln( p eT ), ln(1 + H T ), the energy of the second most energetic jet,
and W W tag . The variable H T is the scalar sum of the transverse energies of all jets in an event. We first apply a requirement using the most effective discriminator of the three. For M < 60 GeV, we require δt t¯ > 0. The efficiency of this requirement is 0.95 for the small- M signal benchmark and 0.03 for t t¯. For M 60 GeV, we require δ Z > 0. The efficiency of this requirement is 0.96 for the large- M signal benchmark and 0.01 for Z /γ ∗ → τ τ . After making these requirements on one variable, we build 2-D distributions of the two remaining discriminants. Fig. 3 shows these distributions for the small- M benchmark signal and the two most significant remaining backgrounds. In calculating the signal exclusion confidence limits, we use only the bins in the upper right quadrant where the signal is concentrated. Table 1 summarizes the observed data events and the expected number of signal and background events. The overall selection efficiencies for the small and large M benchmarks are 0.16% and 12% respectively. The theoretical uncertainties on the signal cross section are approximately 20% as discussed above and are the dominant uncertainties in this analysis. The uncertainty on the integrated luminosity is 6.1%. Other systematic uncertainties included in the limit setting calculations are the lepton identification and track matching efficiencies (5%), the MJ background (27%) scale factor, the jet energy calibration (1–2)%, and the production cross section uncertainties on all the SM background processes (3–10)%. All uncertainties except for those on the MJ background and the SM production cross sections are treated as fully correlated. All systematic uncertainties are included in the limit calculations with Gaussian errors [24]. For M < 60 GeV, we use the two-dimensional histograms of the positive values of δ Z and δ W W in the limit setting procedure. For M 60 GeV, we use the positive values of δt t¯ and δ W W . A modified frequentist approach [25] is used to determine the 95% C.L. exclusion limits on scalar top quark production as a function of the ν˜ and t˜1 masses, as shown in Fig. 4. Also shown are the exclusion regions from the CERN LEP experiments [9] and a previous D0 search [6]. In conclusion, we set 95% C.L. exclusion limits on the cross section for scalar top quark pair production assuming a 100% branch¯ ±l∓ ν˜ ν¯˜ using 5.4 fb−1 of integrated luminosity ing fraction to bbl from the D0 experiment at the Fermilab Tevatron collider. We
D0 Collaboration / Physics Letters B 696 (2011) 321–327
327
FAPERJ, FAPESP and FUNDUNESP (Brazil); DAE and DST (India); Colciencias (Colombia); CONACyT (Mexico); KRF and KOSEF (Korea); CONICET and UBACyT (Argentina); FOM (The Netherlands); STFC and the Royal Society (United Kingdom); MSMT and GACR (Czech Republic); CRC Program and NSERC (Canada); BMBF and DFG (Germany); SFI (Ireland); The Swedish Research Council (Sweden); and CAS and CNSF (China). References [1] [2] [3] [4] [5] [6] [7] [8] [9] Fig. 4. (Color online.) The observed (expected) 95% C.L. exclusion region includes all mass points below the solid (dashed) line. The shaded band around the expected limit shows the effects of the scalar top quark pair production cross section uncertainty. The kinematically forbidden region is shown in the upper left, and the regions excluded by LEP I and LEP II [9], by previous D0 searches [6,7], and by a previous CDF search [8] are also indicated.
have excluded stop pair production for M t˜1 < 210 GeV when M ν˜ < 110 GeV and the difference M t˜1 − M ν˜ > 30 GeV. This extends the previous limits on the top squark mass by more than 40 GeV for sneutrino masses less than 90 GeV and the limits on the sneutrino mass by more than 30 GeV for top squark mass equal to 150 GeV.
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
Acknowledgements
[22]
We thank the staffs at Fermilab and collaborating institutions, and acknowledge support from the DOE and NSF (USA); CEA and CNRS/IN2P3 (France); FASI, Rosatom and RFBR (Russia); CNPq,
[23] [24] [25]
P. Fayet, S. Ferrara, Phys. Rep. 32 (1977) 249. S. Dimopoulos, Nucl. Phys. B 193 (1981) 150. W. Beenakker, et al., Nucl. Phys. B 515 (1998) 3. W. Porod, Phys. Rev. D 59 (1999) 095009. V.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 659 (2008) 500. V.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 675 (2009) 289. W.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 581 (2004) 147. T. Aaltonen, et al., CDF Collaboration, Phys. Rev. D 82 (2010) 092001. LEP SUSY Working Group, ALEPH, DELPHI, L3, OPAL Collaborations, LEPSUSYWG/01-02.1, http://lepsusy.web.cern.ch/lepsusy/, 2001. W.M. Abazov, et al., D0 Collaboration, Nucl. Instr. Methods Phys. Res. A 565 (2006) 463. G. Blazey, et al., arXiv:hep-ex/0005012, 2000. J. Hegeman, J. Phys.: Conf. Ser. 160 (2009) 012024. A. Djouadi, J.-L. Kneur, G. Moultaka, Comput. Phys. Comm. 176 (2007) 426. M. Mühlleitner, Acta Phys. Polon. B 35 (2004) 2753 (we use version 1.1a). J. Alwall, et al., J. High Energy Phys. 0709 (2007) 028 (we use version 4.4.13). T. Sjöstrand, et al., Comput. Phys. Comm. 135 (2001) (we use version 6.409). W. Beenakker, R. Höpker, M. Spira, arXiv:hep-ph/9611232, 1996 (we use version 2.0). J. Pumplin, et al., J. High Energy Phys. 0207 (2002) 012. D. Stump, et al., J. High Energy Phys. 0310 (2003) 046. M. Mangano, et al., J. High Energy Phys. 0307 (2003) 001 (we use version 2.11). R. Brun, F. Carminati, CERN Program Library Long Writeup W5013, 1993, unpublished. R Development Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2009, http://www.R-project.org (we use version 2.9.2). J.A. Nelder, R.W.M. Wedderburn, J. Roy. Statist. Soc. Ser. A 135 (1972) 370. W. Fisher, Fermilab-TM-2386-E, 2007. T. Junk, Nucl. Instr. Methods Phys. Res. A 434 (1999) 435.
Contents lists available at ScienceDirect
Physics Letters B www.elsevier.com/locate/physletb
Search for pair production of the scalar top quark in the electron + muon final state D0 Collaboration 1 V.M. Abazov ai , B. Abbott bu , M. Abolins bj , B.S. Acharya ac , M. Adams av , T. Adams at , G.D. Alexeev ai , G. Alkhazov am , A. Alton bi,2 , G. Alverson bh , G.A. Alves b , L.S. Ancu ah , M. Aoki au , Y. Arnoud n , M. Arov be , A. Askew at , B. Åsman an , O. Atramentov bm , C. Avila h , J. BackusMayes cb , F. Badaud m , L. Bagby au , B. Baldin au , D.V. Bandurin at , S. Banerjee ac , E. Barberis bh , P. Baringer bc , J. Barreto b , J.F. Bartlett au , U. Bassler r , V. Bazterra av , S. Beale f , A. Bean bc , M. Begalli c , M. Begel bs , C. Belanger-Champagne an , L. Bellantoni au , S.B. Beri aa , G. Bernardi q , R. Bernhard v , I. Bertram ao , M. Besançon r , R. Beuselinck ap , V.A. Bezzubov al , P.C. Bhat au , V. Bhatnagar aa , G. Blazey aw , S. Blessing at , K. Bloom bl , A. Boehnlein au , D. Boline br , T.A. Bolton bd , E.E. Boos ak , G. Borissov ao , T. Bose bg , A. Brandt bx , O. Brandt w , R. Brock bj , G. Brooijmans bp , A. Bross au , D. Brown q , J. Brown q , X.B. Bu g , D. Buchholz ax , M. Buehler ca , V. Buescher x , V. Bunichev ak , S. Burdin ao,3 , T.H. Burnett cb , C.P. Buszello ap , B. Calpas o , E. Camacho-Pérez af , M.A. Carrasco-Lizarraga af , B.C.K. Casey au , H. Castilla-Valdez af , S. Chakrabarti br , D. Chakraborty aw , K.M. Chan ba , A. Chandra bz , G. Chen bc , S. Chevalier-Théry r , D.K. Cho bw , S.W. Cho ae , S. Choi ae , B. Choudhary ab , T. Christoudias ap , S. Cihangir au , D. Claes bl , J. Clutter bc , M. Cooke au , W.E. Cooper au , ad ´ M. Corcoran bz , F. Couderc r , M.-C. Cousinou o , A. Croc r , D. Cutts bw , M. Cwiok , A. Das ar , G. Davies ap , bx ah af r au K. De , S.J. de Jong , E. De La Cruz-Burelo , F. Déliot , M. Demarteau , R. Demina bq , D. Denisov au , S.P. Denisov al , S. Desai au , K. DeVaughan bl , H.T. Diehl au , M. Diesburg au , A. Dominguez bl , T. Dorland cb , A. Dubey ab , L.V. Dudko ak , D. Duggan bm , A. Duperrin o , S. Dutt aa , A. Dyshkant aw , M. Eads bl , D. Edmunds bj , J. Ellison as , V.D. Elvira au , Y. Enari q , S. Eno bf , H. Evans ay , A. Evdokimov bs , V.N. Evdokimov al , G. Facini bh , T. Ferbel bf,bq , F. Fiedler x , F. Filthaut ah , W. Fisher bj , H.E. Fisk au , M. Fortner aw , H. Fox ao , S. Fuess au , T. Gadfort bs , A. Garcia-Bellido bq , V. Gavrilov aj , P. Gay m , W. Geist s , W. Geng o,bj , D. Gerbaudo bn , C.E. Gerber av , Y. Gershtein bm , G. Ginther au,bq , G. Golovanov ai , A. Goussiou cb , P.D. Grannis br , S. Greder s , H. Greenlee au , Z.D. Greenwood be , E.M. Gregores d , G. Grenier t , Ph. Gris m , J.-F. Grivaz p , A. Grohsjean r , S. Grünendahl au , M.W. Grünewald ad , F. Guo br , J. Guo br , G. Gutierrez au , P. Gutierrez bu , A. Haas bp,4 , S. Hagopian at , J. Haley bh , L. Han g , K. Harder aq , A. Harel bq , J.M. Hauptman bb , J. Hays ap , T. Head aq , T. Hebbeker u , D. Hedin aw , H. Hegab bv , A.P. Heinson as , U. Heintz bw , C. Hensel w , I. Heredia-De La Cruz af , K. Herner bi , G. Hesketh bh , M.D. Hildreth ba , R. Hirosky ca , T. Hoang at , J.D. Hobbs br , B. Hoeneisen l , M. Hohlfeld x , S. Hossain bu , Z. Hubacek j , N. Huske q , V. Hynek j , I. Iashvili bo , R. Illingworth au , A.S. Ito au , S. Jabeen bw , M. Jaffré p , S. Jain bo , D. Jamin o , R. Jesik ap , K. Johns ar , M. Johnson au , D. Johnston bl , A. Jonckheere au , P. Jonsson ap , J. Joshi aa , A. Juste au,5 , K. Kaadze bd , E. Kajfasz o , D. Karmanov ak , P.A. Kasper au , I. Katsanos bl , R. Kehoe by , S. Kermiche o , N. Khalatyan au , A. Khanov bv , A. Kharchilava bo , Y.N. Kharzheev ai , D. Khatidze bw , M.H. Kirby ax , J.M. Kohli aa , A.V. Kozelov al , J. Kraus bj , A. Kumar bo , A. Kupco k , T. Kurˇca t , V.A. Kuzmin ak ,
1 With visitors from Augustana College, Sioux Falls, SD, USA, The University of Liverpool, Liverpool, UK, SLAC, Menlo Park, CA, USA, ICREA/IFAE, Barcelona, Spain, Centro de Investigacion en Computacion – IPN, Mexico City, Mexico, ECFM, Universidad Autonoma de Sinaloa, Culiacán, Mexico, and Universität Bern, Bern, Switzerland. 2 Augustana College, Sioux Falls, SD, USA. 3 The University of Liverpool, Liverpool, United Kingdom.
0370-2693/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.physletb.2010.12.052
322
D0 Collaboration / Physics Letters B 696 (2011) 321–327
J. Kvita i , S. Lammers ay , G. Landsberg bw , P. Lebrun t , H.S. Lee ae , S.W. Lee bb , W.M. Lee au , J. Lellouch q , L. Li as , Q.Z. Li au , S.M. Lietti e , J.K. Lim ae , D. Lincoln au , J. Linnemann bj , V.V. Lipaev al , R. Lipton au , Y. Liu g , Z. Liu f , A. Lobodenko am , M. Lokajicek k , P. Love ao , H.J. Lubatti cb , R. Luna-Garcia af,6 , A.L. Lyon au , A.K.A. Maciel b , D. Mackin bz , R. Madar r , R. Magaña-Villalba af , S. Malik bl , V.L. Malyshev ai , Y. Maravin bd , J. Martínez-Ortega af , R. McCarthy br , C.L. McGivern bc , M.M. Meijer ah , A. Melnitchouk bk , D. Menezes aw , P.G. Mercadante d , M. Merkin ak , A. Meyer u , J. Meyer w , N.K. Mondal ac , G.S. Muanza o , M. Mulhearn ca , E. Nagy o , M. Naimuddin ab , M. Narain bw , R. Nayyar ab , H.A. Neal bi , J.P. Negret h , P. Neustroev am , S.F. Novaes e , T. Nunnemann y , G. Obrant am , J. Orduna af , N. Osman ap , J. Osta ba , G.J. Otero y Garzón a , M. Owen aq , M. Padilla as , M. Pangilinan bw , N. Parashar az , V. Parihar bw , S.K. Park ae , J. Parsons bp , R. Partridge bw,4 , N. Parua ay , A. Patwa bs , B. Penning au , M. Perfilov ak , K. Peters aq , Y. Peters aq , G. Petrillo bq , P. Pétroff p , R. Piegaia a , J. Piper bj , M.-A. Pleier bs , P.L.M. Podesta-Lerma af,7 , V.M. Podstavkov au , M.-E. Pol b , P. Polozov aj , A.V. Popov al , M. Prewitt bz , D. Price ay , S. Protopopescu bs , J. Qian bi , A. Quadt w , B. Quinn bk , M.S. Rangel b , K. Ranjan ab , P.N. Ratoff ao , I. Razumov al , P. Renkel by , P. Rich aq , M. Rijssenbeek br , I. Ripp-Baudot s , F. Rizatdinova bv , M. Rominsky au , C. Royon r , P. Rubinov au , R. Ruchti ba , G. Safronov aj , G. Sajot n , A. Sánchez-Hernández af , M.P. Sanders y , B. Sanghi au , A.S. Santos e , G. Savage au , L. Sawyer be , T. Scanlon ap , R.D. Schamberger br , Y. Scheglov am , H. Schellman ax , T. Schliephake z , S. Schlobohm cb , C. Schwanenberger aq , R. Schwienhorst bj , J. Sekaric bc , H. Severini bu , E. Shabalina w , V. Shary r , A.A. Shchukin al , R.K. Shivpuri ab , V. Simak j , V. Sirotenko au , P. Skubic bu , P. Slattery bq , D. Smirnov ba , K.J. Smith bo , G.R. Snow bl , J. Snow bt , S. Snyder bs , S. Söldner-Rembold aq , L. Sonnenschein u , A. Sopczak ao , M. Sosebee bx , K. Soustruznik i , B. Spurlock bx , J. Stark n , V. Stolin aj , D.A. Stoyanova al , E. Strauss br , M. Strauss bu , D. Strom av , L. Stutte au , P. Svoisky bu , M. Takahashi aq , A. Tanasijczuk a , W. Taylor f , M. Titov r , V.V. Tokmenin ai , D. Tsybychev br , B. Tuchming r , C. Tully bn , P.M. Tuts bp , L. Uvarov am , S. Uvarov am , S. Uzunyan aw , R. Van Kooten ay , W.M. van Leeuwen ag , N. Varelas av , E.W. Varnes ar , I.A. Vasilyev al , P. Verdier t , L.S. Vertogradov ai , M. Verzocchi au , M. Vesterinen aq , D. Vilanova r , P. Vint ap , P. Vokac j , H.D. Wahl at , M.H.L.S. Wang bq , J. Warchol ba , G. Watts cb , M. Wayne ba , M. Weber au,8 , L. Welty-Rieger ax , M. Wetstein bf , A. White bx , D. Wicke x , M.R.J. Williams ao , G.W. Wilson bc , S.J. Wimpenny as , M. Wobisch be , D.R. Wood bh , T.R. Wyatt aq , Y. Xie au , C. Xu bi , S. Yacoob ax , R. Yamada au , W.-C. Yang aq , T. Yasuda au , Y.A. Yatsunenko ai , Z. Ye au , H. Yin g , K. Yip bs , H.D. Yoo bw , S.W. Youn au , J. Yu bx , S. Zelitch ca , T. Zhao cb , B. Zhou bi , J. Zhu bi , M. Zielinski bq , D. Zieminska ay , L. Zivkovic bp a
Universidad de Buenos Aires, Buenos Aires, Argentina LAFEX, Centro Brasileiro de Pesquisas Físicas, Rio de Janeiro, Brazil c Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil d Universidade Federal do ABC, Santo André, Brazil e Instituto de Física Teórica, Universidade Estadual Paulista, São Paulo, Brazil f Simon Fraser University, Vancouver, British Columbia, and York University, Toronto, Ontario, Canada g University of Science and Technology of China, Hefei, People’s Republic of China h Universidad de los Andes, Bogotá, Colombia i Charles University, Faculty of Mathematics and Physics, Center for Particle Physics, Prague, Czech Republic j Czech Technical University in Prague, Prague, Czech Republic k Center for Particle Physics, Institute of Physics, Academy of Sciences of the Czech Republic, Prague, Czech Republic l Universidad San Francisco de Quito, Quito, Ecuador m LPC, Université Blaise Pascal, CNRS/IN2P3, Clermont, France n LPSC, Université Joseph Fourier Grenoble 1, CNRS/IN2P3, Institut National Polytechnique de Grenoble, Grenoble, France o CPPM, Aix-Marseille Université, CNRS/IN2P3, Marseille, France p LAL, Université Paris-Sud, CNRS/IN2P3, Orsay, France q LPNHE, Universités Paris VI and VII, CNRS/IN2P3, Paris, France r CEA, Irfu, SPP, Saclay, France s IPHC, Université de Strasbourg, CNRS/IN2P3, Strasbourg, France t IPNL, Université Lyon 1, CNRS/IN2P3, Villeurbanne, France and Université de Lyon, Lyon, France u III. Physikalisches Institut A, RWTH Aachen University, Aachen, Germany v Physikalisches Institut, Universität Freiburg, Freiburg, Germany w II. Physikalisches Institut, Georg-August-Universität Göttingen, Göttingen, Germany x Institut für Physik, Universität Mainz, Mainz, Germany y Ludwig-Maximilians-Universität München, München, Germany z Fachbereich Physik, Bergische Universität Wuppertal, Wuppertal, Germany aa Panjab University, Chandigarh, India ab Delhi University, Delhi, India ac Tata Institute of Fundamental Research, Mumbai, India ad University College Dublin, Dublin, Ireland ae Korea Detector Laboratory, Korea University, Seoul, Republic of Korea af CINVESTAV, Mexico City, Mexico ag FOM – Institute NIKHEF and University of Amsterdam/NIKHEF, Amsterdam, The Netherlands b
D0 Collaboration / Physics Letters B 696 (2011) 321–327
323
ah
Radboud University Nijmegen/NIKHEF, Nijmegen, The Netherlands Joint Institute for Nuclear Research, Dubna, Russia aj Institute for Theoretical and Experimental Physics, Moscow, Russia ak Moscow State University, Moscow, Russia al Institute for High Energy Physics, Protvino, Russia am Petersburg Nuclear Physics Institute, St. Petersburg, Russia an Stockholm University, Stockholm and Uppsala University, Uppsala, Sweden ao Lancaster University, Lancaster LA1 4YB, United Kingdom ap Imperial College London, London SW7 2AZ, United Kingdom aq The University of Manchester, Manchester M13 9PL, United Kingdom ar University of Arizona, Tucson, AZ 85721, USA as University of California Riverside, Riverside, CA 92521, USA at Florida State University, Tallahassee, FL 32306, USA au Fermi National Accelerator Laboratory, Batavia, IL 60510, USA av University of Illinois at Chicago, Chicago, IL 60607, USA aw Northern Illinois University, DeKalb, IL 60115, USA ax Northwestern University, Evanston, IL 60208, USA ay Indiana University, Bloomington, IN 47405, USA az Purdue University Calumet, Hammond, IN 46323, USA ba University of Notre Dame, Notre Dame, IN 46556, USA bb Iowa State University, Ames, IA 50011, USA bc University of Kansas, Lawrence, KS 66045, USA bd Kansas State University, Manhattan, KS 66506, USA be Louisiana Tech University, Ruston, LA 71272, USA bf University of Maryland, College Park, MD 20742, USA bg Boston University, Boston, MA 02215, USA bh Northeastern University, Boston, MA 02115, USA bi University of Michigan, Ann Arbor, MI 48109, USA bj Michigan State University, East Lansing, MI 48824, USA bk University of Mississippi, University, MS 38677, USA bl University of Nebraska, Lincoln, NE 68588, USA bm Rutgers University, Piscataway, NJ 08855, USA bn Princeton University, Princeton, NJ 08544, USA bo State University of New York, Buffalo, NY 14260, USA bp Columbia University, New York, NY 10027, USA bq University of Rochester, Rochester, NY 14627, USA br State University of New York, Stony Brook, NY 11794, USA bs Brookhaven National Laboratory, Upton, NY 11973, USA bt Langston University, Langston, OK 73050, USA bu University of Oklahoma, Norman, OK 73019, USA bv Oklahoma State University, Stillwater, OK 74078, USA bw Brown University, Providence, RI 02912, USA bx University of Texas, Arlington, TX 76019, USA by Southern Methodist University, Dallas, TX 75275, USA bz Rice University, Houston, TX 77005, USA ca University of Virginia, Charlottesville, VA 22901, USA cb University of Washington, Seattle, WA 98195, USA ai
a r t i c l e
i n f o
Article history: Received 27 September 2010 Received in revised form 21 November 2010 Accepted 26 December 2010 Available online 31 December 2010 Editor: H. Weerts Keywords: Supersymmetry Squarks Supersymmetric top
a b s t r a c t We report the result of a search for the pair production of the lightest supersymmetric partner of the top quark (t˜1 ) in p p¯ collisions at a center-of-mass energy of 1.96 TeV at the Fermilab Tevatron collider corresponding to an integrated luminosity of 5.4 fb−1 . The scalar top quarks are assumed to decay into a b quark, a charged lepton, and a scalar neutrino (ν˜ ), and the search is performed in the electron plus muon final state. No significant excess of events above the standard model prediction is detected, and substantially improved exclusion limits at the 95% C.L. are set in the the (M t˜1 , M ν˜ ) mass plane. © 2010 Elsevier B.V. All rights reserved.
Supersymmetric theories [1] predict the existence of scalar partners for each of the standard model (SM) fermions. In the minimal supersymmetric standard model (MSSM) [2], the mixing between the chiral states of the scalar partners of the SM fermions
4 5 6 7 8
SLAC, Menlo Park, CA, USA. ICREA/IFAE, Barcelona, Spain. Centro de Investigacion en Computacion – IPN, Mexico City, Mexico. ECFM, Universidad Autonoma de Sinaloa, Culiacán, Mexico. Universität Bern, Bern, Switzerland.
is greatest for the partners of the top quark due to its large Yukawa coupling [3]. Thus, it is possible that the scalar top quark (t˜1 ) is the lightest squark and has the largest production cross section. If R-parity [1] is conserved, then scalar top quarks would be produced by p p¯ collisions in pairs with the dominant processes being quark–antiquark annihilation and gluon fusion [3]. In this Letter we report on a search for the production of t˜1 t¯˜1 ¯ ± μ∓ ν˜ ν˜ final state, where ν˜ is the sneutrino, the pairs in the bbe supersymmetric partner of the neutrino. In this search, we assume that the two-body decays of t˜1 to t χ˜ 10 , t g˜ , and bχ˜ 1+ are kine-
324
D0 Collaboration / Physics Letters B 696 (2011) 321–327
matically forbidden, where χ˜ 10 is the neutralino, g˜ is the gluino, and χ˜ 1+ is the chargino. For this scenario, the three-body decay t˜1 → blν˜ is predicted to dominate over the flavor changing twobody decay t˜1 → c χ˜ 10 [4]. We assume that the t˜1 has a 100% branching fraction to blν˜ , with equal fraction to each lepton type, that R-parity is conserved, and that the sneutrino is the lightest supersymmetric particle or decays invisibly into a neutrino and a neutralino. This analysis uses data corresponding to an integrated luminosity of 5.4 fb−1 collected using √ the D0 detector operating at the Fermilab Tevatron collider at s = 1.96 TeV. The data were collected from April 2002 through June 2009. The D0 Collaboration has previously searched [5–7] for top squark pair production in the final states bb¯ ± ∓ ν˜ ν˜ where the lepton pair is ee, μμ, or e μ. Two of these earlier searches used subsets of this data set corresponding to integrated luminosities of 0.43 fb−1 and 1.1 fb−1 , while the earliest search used data from the Tevatron Run I, corresponding to an integrated luminosity of 108 pb−1 . Searches for top squark pair production in the bb¯ ± ∓ ν˜ ν˜ final states have also been reported by the CDF Collaboration [8] and by the ALEPH, L3, and OPAL Collaborations [9]. The main components of the D0 detector [10] include a central tracking system located inside a 2 T superconducting solenoid. The inner-most tracking element is the silicon microstrip tracker (SMT), followed by a scintillating fiber tracker. These two detectors together measure the momenta of charged particles. The tracking system provides full coverage in the azimuthal (φ ) direction for |η| < 2, where the pseudorapidity η is defined as η = − ln(tan θ/2) and θ is the polar angle with respect to the proton beam direction. Outside the solenoid is the uranium/liquid argon calorimeter which is divided into a central calorimeter and two end-cap calorimeters. Each of these three calorimeters have electromagnetic layers followed by hadronic layers. The outermost component of the detector is the muon system, which consists of proportional drift tubes and scintillator trigger counters, followed by 1.8 T iron toroids and two additional layers of drift tubes and scintillators. Events are selected for offline analysis by a three-level trigger system. All events are required to pass one of a suite of single-electron triggers or single-muon triggers using information from the tracking system, the calorimeter, and the muon system. For each event, a primary p p¯ interaction vertex is defined. If more than one vertex is reconstructed, the primary vertex is taken to be the vertex least consistent with originating from a soft collision. The location of the primary vertex along the beam direction is required to be within ±60 cm of the detector center. Jets are reconstructedusing the D0 Run II cone algorithm [11] with cone of radius R ≡ (φ)2 + ( y )2 < 0.5, where y is the rapidity. Jet energies are calibrated using the standard D0 procedure [12]. A jet is retained in an event if it has transverse energy E T > 20 GeV, |η| < 2.5, and if R(jet, electron) = (φ)2 + (η)2 > 0.5. No requirement on the number of jets is applied. Electrons are required to have transverse momentum p eT > 15 GeV and |η| < 1.1. They are required to be isolated, defined as having [ E tot (0.4) − E EM (0.2)]/ E EM (0.2) < 0.15 where E tot (0.4) is the total calorimeter energy in a cone of radius R = 0.4 and E EM (0.2) is the electromagnetic energy in a cone of radius R = 0.2. In addition, the shower development in the calorimeter is required to be consistent with that of an electromagnetic shower both transversely and longitudinally. An eight-variable likelihood function is constructed to further distinguish between electromagnetic and hadronic showers. The output of this function ranges from 0 to 1, and electrons are required to have a likelihood value greater than 0.85. Electromagnetic showers associated with electrons are also required to match a central track within η < 0.05 and φ < 0.05 of the electromagnetic cluster.
μ
Muons are required to have transverse momentum p T > 10 GeV and |η| < 2. They are required to have both drift tube and scintillator hits in the muon system and to match a track in the central tracker. If the central track includes hits in the SMT, the distance of closest approach (DCA) between the muon track and the primary vertex is required to be less than 0.02 cm. If there are no SMT hits, then the DCA is required to be less than 0.2 cm. Muons are also required to satisfy isolation requirements in both the calorimeter and the central tracker. For the calorimeter isolation, the transverse energy in the cone R < 0.5 around the muon μ track divided by p T must be less than 0.15. For the central tracker isolation, the sum of the transverse energy of the tracks in the μ hollow cone 0.1 < R < 0.5 divided by p T must be less than 0.15. Events are required to have exactly one electron and one muon with opposite charge and to have a minimum separation between the electron and the muon R(e , μ) > 0.5. The missing transverse energy (/ E T ) is calculated from the calorimeter energy corrected for the jet and electron calibrations. It is then adjusted to account for the transverse momentum of the muon. All retained events are required to have / E T > 7 GeV. We refer to this preliminary set of selection requirements as the preselection. Signal Monte Carlo (MC) events are generated in a 2-D grid, i.e., for t˜1 masses ranging from 100 GeV to 240 GeV, and for ν˜ masses ranging from 40 GeV to 140 GeV, each in 10 GeV steps. For each point, the MSSM decay parameters are calculated with suspect [13] and sdecay [14]. madgraph/madevent [15] is used to generate four vectors for the signal events with pythia [16] providing the showering and hadronization. The next-to-leading order (NLO) cross section for t˜1 pair production is calculated by prospino 2.0 [17] with the CTEQ6.1M [18] parton distribution functions (PDFs). The calculations are performed with the factorization and renormalization scales set to one, one half, and two times the t˜1 mass to determine the nominal value and the negative and positive uncertainties. The scale factor uncertainties are combined quadratically with the PDF uncertainties [18,19] to give the total theoretical uncertainties for the signal cross sections. The dominant SM backgrounds for this decay are Z /γ ∗ → τ τ with τ → lν ; diboson production including W W , W Z , and Z Z ; top quark pairs; W + jets; and instrumental background coming from multijet (MJ) processes where jets are misidentified as electrons or contain muons that pass the isolation criterion and with / E T arising from energy mismeasurement. All the background processes in this analysis except for MJ are modeled using MC simulation. Vector boson pair production is simulated with pythia, while all other backgrounds are simulated at the parton level with alpgen [20], with pythia used for hadronization and showering. The detector geometry and response are fully simulated for each MC event using geant [21] based software, and in order to simulate detector noise and multiple p p¯ interaction effects, each MC event is overlayed with a data event from a randomly chosen p p¯ crossing. MC correction factors determined from data are applied to make distributions consistent between data and MC. These corrections include factors for the luminosity profile, beam spot position, muon and electron identification efficiencies, boson transverse momentum, and jet, electron, and muon energy resolutions. The MJ background is estimated from a selection of data events not overlapping with the search sample which is selected by inverting the electron likelihood and muon isolation requirements. Jets fake isolated leptons due to detector or jet fragmentation effects. It is therefore expected that inverting the electron likelihood or the muon isolation requirements does not significantly change the event kinematics and that this background sample is a good representation of the MJ events in the search sample. Because most same-sign di-lepton events come from MJ processes, we obtain the
D0 Collaboration / Physics Letters B 696 (2011) 321–327
325
Fig. 1. (Color online.) E T versus φ(e , μ) for Z /γ ∗ → τ τ MC events. Events inside the black box in the lower right corner are removed. The sizes of the boxes are proportional to the number of events in the underlying cell.
/
Fig. 3. (Color online.) Distribution of δ Z versus δ W W for (a) the small- M signal benchmark, ( M t˜1 , M ν˜ ) = (110 GeV, 90 GeV), MC events, (b) Z /γ ∗ → τ τ MC events, and (c) W W MC events. The top right quadrant is used in the limit-setting procedure.
/
Fig. 2. (Color online.) Distribution of (a) E T , (b) electron p T , and (c) muon p T comparing data and all background processes after Selection 1. The thick dashed and thick solid lines represent the small- M and large- M signal benchmarks, respectively.
normalization factor for the MJ background by taking the ratio of the number of same-sign events that pass the likelihood and isolation requirements to the number of same-sign events that fail these requirements. To remove W + jet events from the MJ sameE T < 20 GeV, sign sample, we make the additional requirement / E T . We also correct this since W + jets events tend to have large /
ratio for non-MJ SM processes that produce like-sign leptons, using the MC samples. Data events are required to satisfy at least one of a suite of single-electron or single-muon triggers. The efficiency of the combination of the single-electron triggers is measured using a subset of the search sample in which at least one of the single-muon triggers fired, and vice-versa for the single muon triggers. The combination of these two efficiencies, taken to be the overall trigger efficiency, is then applied as a correction to the MC samples. The trigger efficiency averaged over all background MC events that pass the preselection is 95%. The mass difference, M = M t˜1 − M ν˜ , determines the kinematics of the final state. A larger M will lead, on average, to larger /E T , larger jet energy, and higher p T charged leptons. We divide the range of M into a “large- M” region ( M > 60 GeV) and “small M” region ( M < 60 GeV). To illustrate these regions, we have chosen two benchmark points, ( M t˜1 , M ν˜ ) = (200 GeV, 100 GeV) and (110 GeV, 90 GeV), which will be referred to as the large M and small- M benchmarks, respectively. Since there are many
326
D0 Collaboration / Physics Letters B 696 (2011) 321–327
Table 1 Expected numbers of background and signal events, and the number of events observed in the data at each stage of the analysis. The errors include statistical and systematic uncertainties. For M < 60 GeV, Selection 2 is δt t¯( M = 20 GeV) > 0, and for M 60 GeV, Selection 2 is δ Z ( M = 100 GeV) > 0. Sample
Preselection events
Selection 1 events
Selection 2
δt t¯ > 0
δZ > 0
events
events
Z → ττ Z → μμ Z → ee WZ WW ZZ t t¯ W MJ
1516 ± 150 33.1 ± 4.7 23.2 ± 3.9 12.7 ± 1.6 295 ± 32 2.2 ± 0.3 25 206+ −28 70 ± 9.2 33 ± 9.2
582 ± 61 22.9 ± 3.7 16.6 ± 3.0 12.0 ± 1.5 268 ± 30 2.0 ± 0.3 25 204+ −28 67.5 ± 9.0 19.7 ± 5.5
515 ± 54 16.3 ± 2.9 10.2 ± 2.2 6.3 ± 0.8 157 ± 18 1.0 ± 0.15 0 .8 6.6+ −0.9 55 ± 7.7 18.4 ± 5.1
17.3 ± 2.0 5.5 ± 1.2 0.1 ± 2.3 9.3 ± 1.2 237 ± 26 1.1 ± 0.16 22 179+ −24 53 ± 7.4 1.3 ± 0.35
Background total Data
160 2191+ −159 2168
1195 ± 73 1147
785 ± 57 776
513 ± 37 472
35 ± 5.6
25.5 ± 4.2
23.8 ± 3.9
–
55 ± 9.3
53.4 ± 9.0
–
51.8 ± 8.7
Small- M benchmark (110 GeV, 90 GeV) Large- M benchmark (200 GeV, 100 GeV)
signal points and their characteristics differ significantly, the analysis strategy is to optimize the signal selection as a function of M. For all values of M, the largest background after preselection is Z /γ ∗ → τ τ . A two-dimensional plot of the azimuthal angle between the electron and muon, φ(e , μ), vs. / E T for Z /γ ∗ → τ τ MC events is shown in Fig. 1. The two leptons from Z /γ ∗ → τ τ tend to be back-to-back in φ , and tend to have low / E T . We therefore reject events in which φ(e , μ) > 2.8 and / E T < 20 GeV and label this as “Selection 1”. Fig. 2 compares / E T , electron p T , and muon p T of the data and the sum of all backgrounds at this stage of the analysis. The agreement confirms our understanding of the SM backgrounds, of the trigger efficiency, and of other MC corrections. After Selection 1, the three largest backgrounds are Z /γ ∗ → τ τ , W W , and t t¯ production. To discriminate these backgrounds from signal we create for each of them a composite discriminant variable from a linear combination of kinematic quantities. We use the R software package [22] to calculate the maximum likelihood coefficients β for a generalized linear model (GLM) [23] of the form
δ A = ln
μ 1−μ
= β0 + β · X
(1)
to discriminate between signal and a specific background source A. Here, μ is the probability that an event is signal, β0 is a con is the vector of event stant, β is the vector of coefficients, and X kinematic variables. By construction, δ A = 0 when μ = 0.5, and signal-like events have positive δ A. The discriminant δ Z is constructed to separate signal from Z /γ ∗ → τ τ background, using an equal number of signal and Z /γ ∗ → τ τ MC events to de we use the following termine the coefficients β0 and β. For X μ variables: ln(/ E T ), ln( p T ), ln( p eT ), φ(e , μ), φ(e , / E T ), φ(μ, / E T ), and φ(e , / E T ) × φ(μ, / E T ). For each value of M, ranging from 20 to 200 GeV, we use the same variables with re-optimized coefficients. We use a similar method for creating the discriminants δ W W and δt t¯ to separate signal from W W and t t¯ backgrounds. μ For δ W W we use the variables ln(/ E T ), ln( p T ), ln( p eT ), number of jets, φ(e , μ), and ln( W W tag ). Here W W tag is the magnitude μ of the vector sum of p eT , p T , and / E T , which should be close to μ E T ), ln( p T ), zero for W W events. For δt t¯, we use the variables ln(/ ln( p eT ), ln(1 + H T ), the energy of the second most energetic jet,
and W W tag . The variable H T is the scalar sum of the transverse energies of all jets in an event. We first apply a requirement using the most effective discriminator of the three. For M < 60 GeV, we require δt t¯ > 0. The efficiency of this requirement is 0.95 for the small- M signal benchmark and 0.03 for t t¯. For M 60 GeV, we require δ Z > 0. The efficiency of this requirement is 0.96 for the large- M signal benchmark and 0.01 for Z /γ ∗ → τ τ . After making these requirements on one variable, we build 2-D distributions of the two remaining discriminants. Fig. 3 shows these distributions for the small- M benchmark signal and the two most significant remaining backgrounds. In calculating the signal exclusion confidence limits, we use only the bins in the upper right quadrant where the signal is concentrated. Table 1 summarizes the observed data events and the expected number of signal and background events. The overall selection efficiencies for the small and large M benchmarks are 0.16% and 12% respectively. The theoretical uncertainties on the signal cross section are approximately 20% as discussed above and are the dominant uncertainties in this analysis. The uncertainty on the integrated luminosity is 6.1%. Other systematic uncertainties included in the limit setting calculations are the lepton identification and track matching efficiencies (5%), the MJ background (27%) scale factor, the jet energy calibration (1–2)%, and the production cross section uncertainties on all the SM background processes (3–10)%. All uncertainties except for those on the MJ background and the SM production cross sections are treated as fully correlated. All systematic uncertainties are included in the limit calculations with Gaussian errors [24]. For M < 60 GeV, we use the two-dimensional histograms of the positive values of δ Z and δ W W in the limit setting procedure. For M 60 GeV, we use the positive values of δt t¯ and δ W W . A modified frequentist approach [25] is used to determine the 95% C.L. exclusion limits on scalar top quark production as a function of the ν˜ and t˜1 masses, as shown in Fig. 4. Also shown are the exclusion regions from the CERN LEP experiments [9] and a previous D0 search [6]. In conclusion, we set 95% C.L. exclusion limits on the cross section for scalar top quark pair production assuming a 100% branch¯ ±l∓ ν˜ ν¯˜ using 5.4 fb−1 of integrated luminosity ing fraction to bbl from the D0 experiment at the Fermilab Tevatron collider. We
D0 Collaboration / Physics Letters B 696 (2011) 321–327
327
FAPERJ, FAPESP and FUNDUNESP (Brazil); DAE and DST (India); Colciencias (Colombia); CONACyT (Mexico); KRF and KOSEF (Korea); CONICET and UBACyT (Argentina); FOM (The Netherlands); STFC and the Royal Society (United Kingdom); MSMT and GACR (Czech Republic); CRC Program and NSERC (Canada); BMBF and DFG (Germany); SFI (Ireland); The Swedish Research Council (Sweden); and CAS and CNSF (China). References [1] [2] [3] [4] [5] [6] [7] [8] [9] Fig. 4. (Color online.) The observed (expected) 95% C.L. exclusion region includes all mass points below the solid (dashed) line. The shaded band around the expected limit shows the effects of the scalar top quark pair production cross section uncertainty. The kinematically forbidden region is shown in the upper left, and the regions excluded by LEP I and LEP II [9], by previous D0 searches [6,7], and by a previous CDF search [8] are also indicated.
have excluded stop pair production for M t˜1 < 210 GeV when M ν˜ < 110 GeV and the difference M t˜1 − M ν˜ > 30 GeV. This extends the previous limits on the top squark mass by more than 40 GeV for sneutrino masses less than 90 GeV and the limits on the sneutrino mass by more than 30 GeV for top squark mass equal to 150 GeV.
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]
Acknowledgements
[22]
We thank the staffs at Fermilab and collaborating institutions, and acknowledge support from the DOE and NSF (USA); CEA and CNRS/IN2P3 (France); FASI, Rosatom and RFBR (Russia); CNPq,
[23] [24] [25]
P. Fayet, S. Ferrara, Phys. Rep. 32 (1977) 249. S. Dimopoulos, Nucl. Phys. B 193 (1981) 150. W. Beenakker, et al., Nucl. Phys. B 515 (1998) 3. W. Porod, Phys. Rev. D 59 (1999) 095009. V.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 659 (2008) 500. V.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 675 (2009) 289. W.M. Abazov, et al., D0 Collaboration, Phys. Lett. B 581 (2004) 147. T. Aaltonen, et al., CDF Collaboration, Phys. Rev. D 82 (2010) 092001. LEP SUSY Working Group, ALEPH, DELPHI, L3, OPAL Collaborations, LEPSUSYWG/01-02.1, http://lepsusy.web.cern.ch/lepsusy/, 2001. W.M. Abazov, et al., D0 Collaboration, Nucl. Instr. Methods Phys. Res. A 565 (2006) 463. G. Blazey, et al., arXiv:hep-ex/0005012, 2000. J. Hegeman, J. Phys.: Conf. Ser. 160 (2009) 012024. A. Djouadi, J.-L. Kneur, G. Moultaka, Comput. Phys. Comm. 176 (2007) 426. M. Mühlleitner, Acta Phys. Polon. B 35 (2004) 2753 (we use version 1.1a). J. Alwall, et al., J. High Energy Phys. 0709 (2007) 028 (we use version 4.4.13). T. Sjöstrand, et al., Comput. Phys. Comm. 135 (2001) (we use version 6.409). W. Beenakker, R. Höpker, M. Spira, arXiv:hep-ph/9611232, 1996 (we use version 2.0). J. Pumplin, et al., J. High Energy Phys. 0207 (2002) 012. D. Stump, et al., J. High Energy Phys. 0310 (2003) 046. M. Mangano, et al., J. High Energy Phys. 0307 (2003) 001 (we use version 2.11). R. Brun, F. Carminati, CERN Program Library Long Writeup W5013, 1993, unpublished. R Development Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2009, http://www.R-project.org (we use version 2.9.2). J.A. Nelder, R.W.M. Wedderburn, J. Roy. Statist. Soc. Ser. A 135 (1972) 370. W. Fisher, Fermilab-TM-2386-E, 2007. T. Junk, Nucl. Instr. Methods Phys. Res. A 434 (1999) 435.