Measurement of the <em>pp→tt</em> production cross section and the top quark mass at s√=1.96  TeV in the all-hadronic decay mode

Article

Reference

Measurement of the pp→tt production cross section and the top quark mass at s√=1.96  TeV in the all-hadronic decay mode

CDF Collaboration

CAMPANELLI, Mario (Collab.), et al.

Abstract

We report the measurements of the tt production cross section and of the top quark mass using 1.02  fb−1 of pp data collected with the CDF II detector at the Fermilab Tevatron. We select events with six or more jets on which a number of kinematical requirements are imposed by means of a neural network algorithm. At least one of these jets must be identified as initiated by a b-quark candidate by the reconstruction of a secondary vertex. The cross section is measured to be σtt=8.3±1.0(stat)+2.0−1.5(syst)±0.5(lumi)  pb, which is consistent with the standard model prediction. The top quark mass of 174.0±2.2(stat)±4.8(syst)  GeV/c2 is derived from a likelihood fit incorporating reconstructed mass distributions representative of signal and background.

CDF Collaboration, CAMPANELLI, Mario (Collab.), et al . Measurement of the pp→tt production cross section and the top quark mass at s√=1.96  TeV in the all-hadronic decay mode. Physical Review. D , 2007, vol. 76, no. 07, p. 072009

DOI : 10.1103/PhysRevD.76.072009

Available at:

http://archive-ouverte.unige.ch/unige:38439

Disclaimer: layout of this document may differ from the published version.

Measurement of the p p ! t t production cross section and the top quark mass at p s

1:96 TeV in the all-hadronic decay mode

T. Aaltonen,²³A. Abulencia,²⁴J. Adelman,¹³T. Affolder,¹⁰T. Akimoto,⁵⁵M. G. Albrow,¹⁷S. Amerio,⁴³D. Amidei,³⁵ A. Anastassov,⁵²K. Anikeev,¹⁷A. Annovi,¹⁹J. Antos,¹⁴M. Aoki,⁵⁵G. Apollinari,¹⁷T. Arisawa,⁵⁷A. Artikov,¹⁵ W. Ashmanskas,¹⁷A. Attal,³A. Aurisano,⁵³F. Azfar,⁴²P. Azzi-Bacchetta,⁴³P. Azzurri,⁴⁶N. Bacchetta,⁴³W. Badgett,¹⁷

A. Barbaro-Galtieri,²⁹V. E. Barnes,⁴⁸B. A. Barnett,²⁵S. Baroiant,⁷V. Bartsch,³¹G. Bauer,³³P.-H. Beauchemin,³⁴ F. Bedeschi,⁴⁶S. Behari,²⁵G. Bellettini,⁴⁶J. Bellinger,⁵⁹A. Belloni,³³D. Benjamin,¹⁶A. Beretvas,¹⁷J. Beringer,²⁹ T. Berry,³⁰A. Bhatti,⁵⁰M. Binkley,¹⁷D. Bisello,⁴³I. Bizjak,³¹R. E. Blair,²C. Blocker,⁶B. Blumenfeld,²⁵A. Bocci,¹⁶

A. Bodek,⁴⁹V. Boisvert,⁴⁹G. Bolla,⁴⁸A. Bolshov,³³D. Bortoletto,⁴⁸J. Boudreau,⁴⁷A. Boveia,¹⁰B. Brau,¹⁰ L. Brigliadori,⁵C. Bromberg,³⁶E. Brubaker,¹³J. Budagov,¹⁵H. S. Budd,⁴⁹S. Budd,²⁴K. Burkett,¹⁷G. Busetto,⁴³ P. Bussey,²¹A. Buzatu,³⁴K. L. Byrum,²S. Cabrera,^16,qM. Campanelli,²⁰M. Campbell,³⁵F. Canelli,¹⁷A. Canepa,⁴⁵ S. Carrillo,^18,iD. Carlsmith,⁵⁹R. Carosi,⁴⁶S. Carron,³⁴B. Casal,¹¹M. Casarsa,⁵⁴A. Castro,⁵P. Catastini,⁴⁶D. Cauz,⁵⁴ M. Cavalli-Sforza,³A. Cerri,²⁹L. Cerrito,^31,mS. H. Chang,²⁸Y. C. Chen,¹M. Chertok,⁷G. Chiarelli,⁴⁶G. Chlachidze,¹⁷

F. Chlebana,¹⁷I. Cho,²⁸K. Cho,²⁸D. Chokheli,¹⁵J. P. Chou,²²G. Choudalakis,³³S. H. Chuang,⁵²K. Chung,¹² W. H. Chung,⁵⁹Y. S. Chung,⁴⁹M. Cilijak,⁴⁶C. I. Ciobanu,²⁴M. A. Ciocci,⁴⁶A. Clark,²⁰D. Clark,⁶M. Coca,¹⁶ G. Compostella,⁴³M. E. Convery,⁵⁰J. Conway,⁷B. Cooper,³¹K. Copic,³⁵M. Cordelli,¹⁹G. Cortiana,⁴³F. Crescioli,⁴⁶ C. Cuenca Almenar,^7,qJ. Cuevas,^11,lR. Culbertson,¹⁷J. C. Cully,³⁵S. DaRonco,⁴³M. Datta,¹⁷S. D’Auria,²¹T. Davies,²¹

D. Dagenhart,¹⁷P. de Barbaro,⁴⁹S. De Cecco,⁵¹A. Deisher,²⁹G. De Lentdecker,^49,cG. De Lorenzo,³M. Dell’Orso,⁴⁶ F. Delli Paoli,⁴³L. Demortier,⁵⁰J. Deng,¹⁶M. Deninno,⁵D. De Pedis,⁵¹P. F. Derwent,¹⁷G. P. Di Giovanni,⁴⁴C. Dionisi,⁵¹

B. Di Ruzza,⁵⁴J. R. Dittmann,⁴M. D’Onofrio,³C. Do¨rr,²⁶S. Donati,⁴⁶P. Dong,⁸J. Donini,⁴³T. Dorigo,⁴³S. Dube,⁵² J. Efron,³⁹R. Erbacher,⁷D. Errede,²⁴S. Errede,²⁴R. Eusebi,¹⁷H. C. Fang,²⁹S. Farrington,³⁰I. Fedorko,⁴⁶W. T. Fedorko,¹³

R. G. Feild,⁶⁰M. Feindt,²⁶J. P. Fernandez,³²R. Field,¹⁸G. Flanagan,⁴⁸R. Forrest,⁷S. Forrester,⁷M. Franklin,²² J. C. Freeman,²⁹I. Furic,¹³M. Gallinaro,⁵⁰J. Galyardt,¹²J. E. Garcia,⁴⁶F. Garberson,¹⁰A. F. Garfinkel,⁴⁸C. Gay,⁶⁰ H. Gerberich,²⁴D. Gerdes,³⁵S. Giagu,⁵¹P. Giannetti,⁴⁶K. Gibson,⁴⁷J. L. Gimmell,⁴⁹C. Ginsburg,¹⁷N. Giokaris,^15,a M. Giordani,⁵⁴P. Giromini,¹⁹M. Giunta,⁴⁶G. Giurgiu,²⁵V. Glagolev,¹⁵D. Glenzinski,¹⁷M. Gold,³⁷N. Goldschmidt,¹⁸

J. Goldstein,^42,bA. Golossanov,¹⁷G. Gomez,¹¹G. Gomez-Ceballos,³³M. Goncharov,⁵³O. Gonza´lez,³²I. Gorelov,³⁷ A. T. Goshaw,¹⁶K. Goulianos,⁵⁰A. Gresele,⁴³S. Grinstein,²²C. Grosso-Pilcher,¹³R. C. Group,¹⁷U. Grundler,²⁴ J. Guimaraes da Costa,²²Z. Gunay-Unalan,³⁶C. Haber,²⁹K. Hahn,³³S. R. Hahn,¹⁷E. Halkiadakis,⁵²A. Hamilton,²⁰

B.-Y. Han,⁴⁹J. Y. Han,⁴⁹R. Handler,⁵⁹F. Happacher,¹⁹K. Hara,⁵⁵D. Hare,⁵²M. Hare,⁵⁶S. Harper,⁴²R. F. Harr,⁵⁸ R. M. Harris,¹⁷M. Hartz,⁴⁷K. Hatakeyama,⁵⁰J. Hauser,⁸C. Hays,⁴²M. Heck,²⁶A. Heijboer,⁴⁵B. Heinemann,²⁹ J. Heinrich,⁴⁵C. Henderson,³³M. Herndon,⁵⁹J. Heuser,²⁶D. Hidas,¹⁶C. S. Hill,^10,bD. Hirschbuehl,²⁶A. Hocker,¹⁷

A. Holloway,²²S. Hou,¹M. Houlden,³⁰S.-C. Hsu,⁹B. T. Huffman,⁴²R. E. Hughes,³⁹U. Husemann,⁶⁰J. Huston,³⁶ J. Incandela,¹⁰G. Introzzi,⁴⁶M. Iori,⁵¹A. Ivanov,⁷B. Iyutin,³³E. James,¹⁷D. Jang,⁵²B. Jayatilaka,¹⁶D. Jeans,⁵¹ E. J. Jeon,²⁸S. Jindariani,¹⁸W. Johnson,⁷M. Jones,⁴⁸K. K. Joo,²⁸S. Y. Jun,¹²J. E. Jung,²⁸T. R. Junk,²⁴T. Kamon,⁵³

P. E. Karchin,⁵⁸Y. Kato,⁴¹Y. Kemp,²⁶R. Kephart,¹⁷U. Kerzel,²⁶V. Khotilovich,⁵³B. Kilminster,³⁹D. H. Kim,²⁸ H. S. Kim,²⁸J. E. Kim,²⁸M. J. Kim,¹⁷S. B. Kim,²⁸S. H. Kim,⁵⁵Y. K. Kim,¹³N. Kimura,⁵⁵L. Kirsch,⁶S. Klimenko,¹⁸

M. Klute,³³B. Knuteson,³³B. R. Ko,¹⁶K. Kondo,⁵⁷D. J. Kong,²⁸J. Konigsberg,¹⁸A. Korytov,¹⁸A. V. Kotwal,¹⁶ A. C. Kraan,⁴⁵J. Kraus,²⁴M. Kreps,²⁶J. Kroll,⁴⁵N. Krumnack,⁴M. Kruse,¹⁶V. Krutelyov,¹⁰T. Kubo,⁵⁵S. E. Kuhlmann,²

T. Kuhr,²⁶N. P. Kulkarni,⁵⁸Y. Kusakabe,⁵⁷S. Kwang,¹³A. T. Laasanen,⁴⁸S. Lai,³⁴S. Lami,⁴⁶S. Lammel,¹⁷ M. Lancaster,³¹R. L. Lander,⁷K. Lannon,³⁹A. Lath,⁵²G. Latino,⁴⁶I. Lazzizzera,⁴³T. LeCompte,²J. Lee,⁴⁹J. Lee,²⁸

Y. J. Lee,²⁸S. W. Lee,^53,oR. Lefe`vre,²⁰N. Leonardo,³³S. Leone,⁴⁶S. Levy,¹³J. D. Lewis,¹⁷C. Lin,⁶⁰C. S. Lin,¹⁷ M. Lindgren,¹⁷E. Lipeles,⁹A. Lister,⁷D. O. Litvintsev,¹⁷T. Liu,¹⁷N. S. Lockyer,⁴⁵A. Loginov,⁶⁰M. Loreti,⁴³R.-S. Lu,¹

D. Lucchesi,⁴³P. Lujan,²⁹P. Lukens,¹⁷G. Lungu,¹⁸L. Lyons,⁴²J. Lys,²⁹R. Lysak,¹⁴E. Lytken,⁴⁸P. Mack,²⁶ D. MacQueen,³⁴R. Madrak,¹⁷K. Maeshima,¹⁷K. Makhoul,³³T. Maki,²³P. Maksimovic,²⁵S. Malde,⁴²S. Malik,³¹ G. Manca,³⁰A. Manousakis,^15,aF. Margaroli,⁵R. Marginean,¹⁷C. Marino,²⁶C. P. Marino,²⁴A. Martin,⁶⁰M. Martin,²⁵ V. Martin,^21,gM. Martı´nez,³R. Martı´nez-Balları´n,³²T. Maruyama,⁵⁵P. Mastrandrea,⁵¹T. Masubuchi,⁵⁵H. Matsunaga,⁵⁵ M. E. Mattson,⁵⁸R. Mazini,³⁴P. Mazzanti,⁵K. S. McFarland,⁴⁹P. McIntyre,⁵³R. McNulty,^30,fA. Mehta,³⁰P. Mehtala,²³ S. Menzemer,^11,hA. Menzione,⁴⁶P. Merkel,⁴⁸C. Mesropian,⁵⁰A. Messina,³⁶T. Miao,¹⁷N. Miladinovic,⁶J. Miles,³³ R. Miller,³⁶C. Mills,¹⁰M. Milnik,²⁶A. Mitra,¹G. Mitselmakher,¹⁸A. Miyamoto,²⁷S. Moed,²⁰N. Moggi,⁵B. Mohr,⁸

C. S. Moon,²⁸R. Moore,¹⁷M. Morello,⁴⁶P. Movilla Fernandez,²⁹J. Mu¨lmensta¨dt,²⁹A. Mukherjee,¹⁷Th. Muller,²⁶

R. Mumford,²⁵P. Murat,¹⁷M. Mussini,⁵J. Nachtman,¹⁷A. Nagano,⁵⁵J. Naganoma,⁵⁷K. Nakamura,⁵⁵I. Nakano,⁴⁰ A. Napier,⁵⁶V. Necula,¹⁶C. Neu,⁴⁵M. S. Neubauer,⁹J. Nielsen,^29,nL. Nodulman,²O. Norniella,³E. Nurse,³¹S. H. Oh,¹⁶

Y. D. Oh,²⁸I. Oksuzian,¹⁸T. Okusawa,⁴¹R. Oldeman,³⁰R. Orava,²³K. Osterberg,²³C. Pagliarone,⁴⁶E. Palencia,¹¹ V. Papadimitriou,¹⁷A. Papaikonomou,²⁶A. A. Paramonov,¹³B. Parks,³⁹S. Pashapour,³⁴J. Patrick,¹⁷G. Pauletta,⁵⁴ M. Paulini,¹²C. Paus,³³D. E. Pellett,⁷A. Penzo,⁵⁴T. J. Phillips,¹⁶G. Piacentino,⁴⁶J. Piedra,⁴⁴L. Pinera,¹⁸K. Pitts,²⁴

C. Plager,⁸L. Pondrom,⁵⁹X. Portell,³O. Poukhov,¹⁵N. Pounder,⁴²F. Prakoshyn,¹⁵A. Pronko,¹⁷J. Proudfoot,² F. Ptohos,^19,eG. Punzi,⁴⁶J. Pursley,²⁵J. Rademacker,^42,bA. Rahaman,⁴⁷V. Ramakrishnan,⁵⁹N. Ranjan,⁴⁸I. Redondo,³² B. Reisert,¹⁷V. Rekovic,³⁷P. Renton,⁴²M. Rescigno,⁵¹S. Richter,²⁶F. Rimondi,⁵L. Ristori,⁴⁶A. Robson,²¹T. Rodrigo,¹¹ E. Rogers,²⁴S. Rolli,⁵⁶R. Roser,¹⁷M. Rossi,⁵⁴R. Rossin,¹⁰P. Roy,³⁴A. Ruiz,¹¹J. Russ,¹²V. Rusu,¹³H. Saarikko,²³ A. Safonov,⁵³W. K. Sakumoto,⁴⁹G. Salamanna,⁵¹O. Salto´,³L. Santi,⁵⁴S. Sarkar,⁵¹L. Sartori,⁴⁶K. Sato,¹⁷P. Savard,³⁴

A. Savoy-Navarro,⁴⁴T. Scheidle,²⁶P. Schlabach,¹⁷E. E. Schmidt,¹⁷M. P. Schmidt,⁶⁰M. Schmitt,³⁸T. Schwarz,⁷ L. Scodellaro,¹¹A. L. Scott,¹⁰A. Scribano,⁴⁶F. Scuri,⁴⁶A. Sedov,⁴⁸S. Seidel,³⁷Y. Seiya,⁴¹A. Semenov,¹⁵ L. Sexton-Kennedy,¹⁷A. Sfyrla,²⁰S. Z. Shalhout,⁵⁸M. D. Shapiro,²⁹T. Shears,³⁰P. F. Shepard,⁴⁷D. Sherman,²² M. Shimojima,^55,kM. Shochet,¹³Y. Shon,⁵⁹I. Shreyber,²⁰A. Sidoti,⁴⁶P. Sinervo,³⁴A. Sisakyan,¹⁵A. J. Slaughter,¹⁷ J. Slaunwhite,³⁹K. Sliwa,⁵⁶J. R. Smith,⁷F. D. Snider,¹⁷R. Snihur,³⁴M. Soderberg,³⁵A. Soha,⁷S. Somalwar,⁵²V. Sorin,³⁶

J. Spalding,¹⁷F. Spinella,⁴⁶T. Spreitzer,³⁴P. Squillacioti,⁴⁶M. Stanitzki,⁶⁰A. Staveris-Polykalas,⁴⁶R. St. Denis,²¹ B. Stelzer,⁸O. Stelzer-Chilton,⁴²D. Stentz,³⁸J. Strologas,³⁷D. Stuart,¹⁰J. S. Suh,²⁸A. Sukhanov,¹⁸H. Sun,⁵⁶I. Suslov,¹⁵

T. Suzuki,⁵⁵A. Taffard,^24,pR. Takashima,⁴⁰Y. Takeuchi,⁵⁵R. Tanaka,⁴⁰M. Tecchio,³⁵P. K. Teng,¹K. Terashi,⁵⁰ J. Thom,^17,dA. S. Thompson,²¹E. Thomson,⁴⁵P. Tipton,⁶⁰V. Tiwari,¹²S. Tkaczyk,¹⁷D. Toback,⁵³S. Tokar,¹⁴ K. Tollefson,³⁶T. Tomura,⁵⁵D. Tonelli,⁴⁶S. Torre,¹⁹D. Torretta,¹⁷S. Tourneur,⁴⁴W. Trischuk,³⁴S. Tsuno,⁴⁰Y. Tu,⁴⁵ N. Turini,⁴⁶F. Ukegawa,⁵⁵S. Uozumi,⁵⁵S. Vallecorsa,²⁰N. van Remortel,²³A. Varganov,³⁵E. Vataga,³⁷F. Vazquez,^18,i

G. Velev,¹⁷C. Vellidis,^46,aG. Veramendi,²⁴V. Veszpremi,⁴⁸M. Vidal,³²R. Vidal,¹⁷I. Vila,¹¹R. Vilar,¹¹T. Vine,³¹ M. Vogel,³⁷I. Vollrath,³⁴I. Volobouev,^29,oG. Volpi,⁴⁶F. Wu¨rthwein,⁹P. Wagner,⁵³R. G. Wagner,²R. L. Wagner,¹⁷ J. Wagner,²⁶W. Wagner,²⁶R. Wallny,⁸S. M. Wang,¹A. Warburton,³⁴D. Waters,³¹M. Weinberger,⁵³W. C. Wester III,¹⁷

B. Whitehouse,⁵⁶D. Whiteson,^45,pA. B. Wicklund,²E. Wicklund,¹⁷G. Williams,³⁴H. H. Williams,⁴⁵P. Wilson,¹⁷ B. L. Winer,³⁹P. Wittich,^17,dS. Wolbers,¹⁷C. Wolfe,¹³T. Wright,³⁵X. Wu,²⁰S. M. Wynne,³⁰A. Yagil,⁹K. Yamamoto,⁴¹

J. Yamaoka,⁵²T. Yamashita,⁴⁰C. Yang,⁶⁰U. K. Yang,^13,jY. C. Yang,²⁸W. M. Yao,²⁹G. P. Yeh,¹⁷J. Yoh,¹⁷K. Yorita,¹³ T. Yoshida,⁴¹G. B. Yu,⁴⁹I. Yu,²⁸S. S. Yu,¹⁷J. C. Yun,¹⁷L. Zanello,⁵¹A. Zanetti,⁵⁴I. Zaw,²²X. Zhang,²⁴

J. Zhou,⁵²and S. Zucchelli⁵ (CDF Collaboration)

1Institute of Physics, Academia Sinica, Taipei, Taiwan 11529, Republic of China

2Argonne National Laboratory, Argonne, Illinois 60439, USA

3Institut de Fisica d’Altes Energies, Universitat Autonoma de Barcelona, E-08193, Bellaterra (Barcelona), Spain

4Baylor University, Waco, Texas 76798, USA

5Istituto Nazionale di Fisica Nucleare, University of Bologna, I-40127 Bologna, Italy

6Brandeis University, Waltham, Massachusetts 02254, USA

7University of California, Davis, Davis, California 95616, USA

8University of California, Los Angeles, Los Angeles, California 90024, USA

9University of California, San Diego, La Jolla, California 92093, USA

10University of California, Santa Barbara, Santa Barbara, California 93106, USA

11Instituto de Fisica de Cantabria, CSIC-University of Cantabria, 39005 Santander, Spain

12Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, USA

13Enrico Fermi Institute, University of Chicago, Chicago, Illinois 60637, USA

14Comenius University, 842 48 Bratislava, Slovakia and Institute of Experimental Physics, 040 01 Kosice, Slovakia

15Joint Institute for Nuclear Research, RU-141980 Dubna, Russia

16Duke University, Durham, North Carolina 27708

17Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA

18University of Florida, Gainesville, Florida 32611, USA

19Laboratori Nazionali di Frascati, Istituto Nazionale di Fisica Nucleare, I-00044 Frascati, Italy

20University of Geneva, CH-1211 Geneva 4, Switzerland

21Glasgow University, Glasgow G12 8QQ, United Kingdom

22Harvard University, Cambridge, Massachusetts 02138, USA

23Division of High Energy Physics, Department of Physics, University of Helsinki and Helsinki Institute of Physics, FIN-00014, Helsinki, Finland

24University of Illinois, Urbana, Illinois 61801, USA

25The Johns Hopkins University, Baltimore, Maryland 21218, USA

26Institut fu¨r Experimentelle Kernphysik, Universita¨t Karlsruhe, 76128 Karlsruhe, Germany

27High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305, Japan

28Center for High Energy Physics, Kyungpook National University, Taegu 702-701, Korea;

Seoul National University, Seoul 151-742, Korea;

and SungKyunKwan University, Suwon 440-746, Korea

29Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA

30University of Liverpool, Liverpool L69 7ZE, United Kingdom

31University College London, London WC1E 6BT, United Kingdom

32Centro de Investigaciones Energeticas Medioambientales y Tecnologicas, E-28040 Madrid, Spain

33Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA

34Institute of Particle Physics, McGill University, Montre´al, Canada H3A 2T8 and University of Toronto, Toronto, Canada M5S 1A7

35University of Michigan, Ann Arbor, Michigan 48109, USA

36Michigan State University, East Lansing, Michigan 48824, USA

37University of New Mexico, Albuquerque, New Mexico 87131, USA

38Northwestern University, Evanston, Illinois 60208, USA

39The Ohio State University, Columbus, Ohio 43210, USA

40Okayama University, Okayama 700-8530, Japan

41Osaka City University, Osaka 588, Japan

42University of Oxford, Oxford OX1 3RH, United Kingdom

43University of Padova, Istituto Nazionale di Fisica Nucleare, Sezione di Padova-Trento, I-35131 Padova, Italy

44LPNHE, Universite Pierre et Marie Curie/IN2P3-CNRS, UMR7585, Paris, F-75252 France

45University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA

46Istituto Nazionale di Fisica Nucleare Pisa, Universities of Pisa, Siena and Scuola Normale Superiore, I-56127 Pisa, Italy

47University of Pittsburgh, Pittsburgh, Pennsylvania 15260, USA

48Purdue University, West Lafayette, Indiana 47907, USA

49University of Rochester, Rochester, New York 14627, USA

50The Rockefeller University, New York, New York 10021, USA

51Istituto Nazionale di Fisica Nucleare, Sezione di Roma 1, University of Rome ‘‘La Sapienza,’’ I-00185 Roma, Italy

52Rutgers University, Piscataway, New Jersey 08855, USA

53Texas A&M University, College Station, Texas 77843, USA

54Istituto Nazionale di Fisica Nucleare, University of Trieste/Udine, Italy

55University of Tsukuba, Tsukuba, Ibaraki 305, Japan

56Tufts University, Medford, Massachusetts 02155, USA

57Waseda University, Tokyo 169, Japan

58Wayne State University, Detroit, Michigan 48201, USA

aVisiting from University of Athens, 15784 Athens, Greece.

bVisiting from University of Bristol, Bristol BS8 1TL, United Kingdom.

cVisiting from University Libre de Bruxelles, B-1050 Brussels, Belgium.

dVisiting from Cornell University, Ithaca, NY 14853, USA.

eVisiting from University of Cyprus, Nicosia CY-1678, Cyprus.

fVisiting from University College Dublin, Dublin 4, Ireland.

gVisiting from University of Edinburgh, Edinburgh EH9 3JZ, United Kingdom.

hVisiting from University of Heidelberg, D-69120 Heidelberg, Germany.

iVisiting from Universidad Iberoamericana, Mexico D.F., Mexico.

jVisiting from University of Manchester, Manchester M13 9PL, United Kingdom.

kVisiting from Nagasaki Institute of Applied Science, Nagasaki, Japan.

lVisiting from University de Oviedo, E-33007 Oviedo, Spain.

mVisiting from University of London, Queen Mary College, London, E1 4NS, United Kingdom.

oVisiting from Texas Tech University, Lubbock, TX 79409, USA.

pVisiting from University of California, Irvine, Irvine, CA 92697, USA.

qVisiting from IFIC (CSIC–Universitat de Valencia), 46071 Valencia, Spain.

nVisiting from University of California Santa Cruz, Santa Cruz, CA 95064, USA.

pp tt. . .

59University of Wisconsin, Madison, Wisconsin 53706, USA

60Yale University, New Haven, Connecticut 06520, USA (Received 25 June 2007; published 22 October 2007)

We report the measurements of the tt production cross section and of the top quark mass using 1:02 fb¹ofppdata collected with the CDF II detector at the Fermilab Tevatron. We select events with six or more jets on which a number of kinematical requirements are imposed by means of a neural network algorithm. At least one of these jets must be identified as initiated by a b-quark candidate by the reconstruction of a secondary vertex. The cross section is measured to bett8:31:0stat^2:0_1:5 syst 0:5lumipb, which is consistent with the standard model prediction. The top quark mass of 174:02:2stat 4:8systGeV=c² is derived from a likelihood fit incorporating reconstructed mass distributions representative of signal and background.

DOI:10.1103/PhysRevD.76.072009 PACS numbers: 14.65.Ha, 13.85.Ni, 13.85.Qk

I. INTRODUCTION

The measurement of the top quark properties allows one to verify the consistency of the standard model. At the Fermilab Tevatron Collider top quarks are produced mostly in pairs and the measurement of thettcross section tests the next-to-leading-order QCD calculations. Moreover, accurate measurements of the top quark mass and of the mass of theWboson provide constraints on the mass of the hypothetical standard model Higgs boson [1].

At the Tevatron center-of-mass energy, ps

1:96 TeV, the predictedttproduction cross section is 6.7 pb [2] for an assumed top quark mass of 175 GeV=c², with a total theoretical uncertainty of approximately 15%, due to the choice of renormalization and factorization scales. The top quark decays into aWboson and abquark almost 100% of the time; the W boson subsequently decays to either a quark-antiquark pair or a lepton-neutrino pair. The result- ing final states are then usually distinguished by the num- ber of charged energetic leptons (eor) and the number of jets.

In this analysis, we examine events characterized by a multijet topology and no energetic lepton (‘‘all-hadronic’’

mode). This tt final state has the advantage of a large branching ratio (4=9) and of having no undetected top decay physical observables in the final state. In addition, discrepancies in top quark cross section and mass mea- surements between this and other decay channels could indicate contributions from physics beyond the standard model [3]. The major challenge of this channel is the very large QCD multijet background which dominates the sig- nal by 3 orders of magnitude after the application of the online trigger selection. To improve the signal-to- background ratio (S/B), requirements based on the kine- matical and topological characteristics of standard modeltt events are expressed in terms of a neural network and applied to the data. This neural network selection reaches a S/B of about 1=12, improving the S/B by 60% with respect to the selection based on kinematical cuts used in previous analyses [4,5]. The neural network selection is followed by the requirement of jets identified as originating

from b quarks using a secondary vertex b-tagging algo- rithm, achieving a S/B of about1=2.

The purity of the selected sample enables the clear observation of ttcandidates and the measurement of the ttproduction cross section and of the top quark mass. Top quark events result in final states with a number of b-tagged jets larger than in inclusive QCD multijet pro- duction, so we use the excess of such tags to measure thett production cross section. A reconstructed top quark mass is then determined from a kinematical fit of the six leading jets in the event to a tt final state. The distribution of reconstructed top quark masses is then compared to the distributions expected from background and ttevents si- mulated at various values of the top quark mass, to obtain the value which best describes the data.

Given the theoretical uncertainties on the production cross section for events generated with N partons at tree level, a more accurate background estimate is obtained from the data themselves (‘‘data-driven’’) rather than from theoretical prediction of cross section and simulations.

The CDF and D0 Collaborations previously measured the ttproduction cross section and the top quark mass in the all-hadronic channel [6] using data sets with integrated luminosities of approximately110 pb¹ collected at

ps 1:8 TeV during the Tevatron run I (1992 –1996). More recent run II measurements of the cross section [4] and mass [5] have been performed by CDF using 311 pb¹ collected with the CDF II detector at

ps

1:96 TeV (March 2002 –August 2004). The results reported here are based on the data taken between March 2002 and February 2006, corresponding to an integrated luminosity of 1:02 fb¹. These measurements complement other re- centttcross section and top quark mass determinations by CDF [7–10] and D0 [11–13] in other final states.

The organization of the paper is as follows: Sec. II contains a brief description of the CDF II detector. The trigger and the neural-network-based sample selection are described in Sec. III. The b-tagging algorithm and its efficiency for identifying bjets are described in Sec. IV.

In Sec. V the data-driven method we use for estimating the

background from experimental multijet data is applied and the related systematic uncertainties are evaluated.

Section VI describes the optimization of the kinematical selection and the associated efficiency. Thettproduction cross section is presented in Sec. VII. The reconstruction of the top quark mass and the corresponding background and signal distributions are described in Sec. VIII; the method for fitting these distributions is discussed in Sec. IX.

Section X summarizes the expected contributions to the systematic uncertainty on the top quark mass measure- ment, while Sec. XI describes the top quark mass mea- surement. Finally, cross section and mass measurements are summarized in Sec. XII.

II. THE CDF II DETECTOR

The CDF II detector [14] is an azimuthally and forward- backward symmetric apparatus designed to studypp col- lisions at the Tevatron. A cylindrical coordinate system as described in [15] is used. The detector consists of a mag- netic spectrometer surrounded by calorimeters and muon chambers. The charged particle tracking system is im- mersed in a 1.4 T solenoidal magnetic field with axis parallel to the beam line. A set of silicon microstrip de- tectors provides charged particle tracking in the radial range from 1.5 to 28 cm, while a 3.1 m long open-cell drift chamber, the central outer tracker (COT), covers the radial range from 40 to 137 cm. In combination the silicon and COT detectors provide excellent tracking up to pseudor- apidities [15] jj 1:1, with decreasing coverage up to jj 2:0. Segmented electromagnetic and hadronic calo- rimeters surround the tracking system and measure the energy of interacting particles. The electromagnetic and hadronic calorimeters are lead-scintillator and iron- scintillator sampling devices, respectively, covering the rangejj 3:6. They are segmented in the central region (jj<1:1) in towers subtending 15in azimuth and 0.1 in , and in the forward region (1:1<jj<3:6) in towers subtending 7.5forjj<2:11and 15forjj>2:11. The electromagnetic calorimeters [16,17] are instrumented with proportional and scintillating strip detectors that mea- sure the transverse profile of electromagnetic showers at a depth corresponding to the expected shower maximum.

Drift chambers located outside the central hadronic calo- rimeters and behind a 60 cm iron shield detect muons with jj 0:6[18]. Additional drift chambers and scintillation counters detect muons in the region0:6<jj<1:5. Gas Cherenkov counters [19] with a coverage of3:7<jj<

4:7measure the average number of inelasticpp collisions and thereby determine the luminosity.

III. MULTIJET EVENT SELECTION

The all-hadronic final state ofttevents is characterized by the presence of at least six jets from the decay of the two top quarks. Collisions are selected in real time using a

multijet trigger, relying on calorimeter information, which was specially developed to collect the events used in this analysis. Subsequently, jets are identified off-line by grouping clusters of energy in the calorimeter using a fixed-cone algorithm with a radius 0.4 in space [20]. The typical jet transverse energy [15] resolution is approximately 0:1E_T1:0 GeV [21], where E_T is the jet transverse energy in GeV. After a preliminary selection of multijet events, a neural network selection based on relevant kinematical variables is used to provide the most precise cross section and mass measurements.

A. Multijet trigger

The CDF trigger system has three levels. The first two levels consist of special-purpose electronics and the third one of conventional digital processors. For triggering pur- poses the calorimeter granularity is simplified to a2424 grid in space with each ‘‘trigger tower’’ spanning approximately 15inand 0.2 incovering one or two physical towers. At level 1, the jet trigger requires a single tower with transverse energy E^tow_T 10 GeV. At level 2 we require the total transverse energy, summed over all the trigger towers,P

E^tow_T , to be175 GeV, and the presence of four or more clusters of calorimeter towers, each cluster with transverse energy E^cls_T 15 GeV. In order to main- tain an acceptable trigger rate, as the peak instantaneous luminosity increased, the P

E^tow_T threshold has been in- creased over the course of the data taking, 175 GeV being the latest value. This threshold has been applied, off-line, to the whole data set. Finally, the third trigger level con- firms the level 2 selection using a more accurate determi- nation of the jet energy, requiring four or more reconstructed jets with E_T 10 GeV. A total of 4 340 143 events satisfy the trigger requirements, with an effi- ciency of about 58% for inclusivettevents, and of 80% for all-hadronic tt events. The efficiency has been estimated using Monte Carlo generated tt events, described later.

This corresponds to the S/B of approximately 1=1100for the theoretical cross section of 6.7 pb, where the back- ground is represented by the multijet events themselves.

B. Preselection and topology requirements Events satisfying the trigger requirements are recon- structed in terms of their final state observables (tracks, vertices, charged leptons, jets). We retain only those events that are well contained in detector acceptance, requiring the primary event vertex [22] to lie inside the luminous region (jzj<60 cm). The jet energies are corrected for detector response and multiple interactions. First, we take into account the dependence of detector response and energy loss in the uninstrumented regions. After a small correction for the extra energy deposited by multiple col- lisions in the same beam-beam bunch crossing, a correc- tion for calorimeter nonlinearity is applied so that the jet energies correspond, on average, to the energy of all the pp tt. . .

particles within the jet cone. All systematic uncertainties for the individual corrections are added in quadrature to obtain the total uncertainty on the estimate of the initial parton energy. This uncertainty goes from 8% to 3% with jet transverse energy increasing from 15 GeV to 50 GeV, and remains approximately constant at 3% above 50 GeV [23].

For this analysis, each jet is required to have E_T 15 GeV and jj 2 after all corrections have been ap- plied. In order to remove the events from the tt leptonic channels, we veto events containing any well-identified high-p_Telectrons and muons as defined in [8], and require that P^{E6 T}

ET

p be<3 GeV

p [9], where the missing transverse energy,E6 _T [24], is computed with reference to the detector origin and is corrected for any identified muons and the position of thepp collision point, whileP

E_T is obtained by summing theE_T’s of all the selected jets. At this stage, called ‘‘preselection,’’ simulations show that the fraction of leptonic events amounts to about 14% of all accepted tt events. To avoid overlaps between jets we require jet pairs to be separated by at least 0.5 units in the space.

About 3:510⁶ events pass these preselection require- ments (S/B1=1000). Finally, we define the topology of the signal region by selecting events with a number of jets 6 N_jets 8to optimize the signal fraction. A total of 506 567 events pass this additional requirement with an ex- pected S/B of approximately 1=370, i.e. about 0.3%; the remaining events with lower jet multiplicity have much smaller values of S/B and are used as control regions. The residual fraction of leptonicttevents amounts to only 5%

of all accepted events.

C. Neural-network-based kinematical selection In order to further improve the S/B we use a neural network approach to recognize in more detail the features of signal and background events, including correlations between the kinematical variables which enter as input nodes in the network. We thus expect a better separation between signal and background relative to the former technique [4] where correlations were not fully considered.

The network uses theMLPFITpackage [25] as implemented byROOT[26] through theTMultiLayerPerceptronclass. We define a kinematical selection based on dynamical and topological properties of the candidate event. The number of variables used should allow the best possible description of the event properties, but, at the same time, too many input variables can worsen the performance, given the limited training statistics. As a guideline we studied differ- ent neural network configurations, in terms of inputs and hidden nodes, adding a few variables at a time, looking for the best performance in terms of largest S/B. The first quantities considered are those used in [4]: the total trans- verse energy of the jets, PE_T; the quantity P

3E_T PE_TE¹_TE²_T, obtained by removing the contribution

of the two jets with the highestE_T; the centrality, defined as C

P_{E ^}T

s

p , where

^ ps

is the invariant mass of the multijet system; and the aplanarity A, defined as A³₂Q1, Q1

being the smallest of the three normalized eigenvalues of the sphericity tensor, M^abP

jP^a_jP^b_j, calculated in the center-of-mass system of all jets, where the indicesaand brefer to the spatial components of the jet four-momentum P_j. In addition, we consider the dynamical properties of dijet and trijet systems through the use of the minimum and maximum values of the invariant mass among all possible jet permutations: M_2j^min,M^max_2j ,M^min_3j , andM_3j^max. We obtain another set of discriminating variables combining the transverse energy of the jets with their emission direction, represented by the angle ^? between the jet direction, as measured in the center-of-mass frame of all jets, and the proton beam axis. The variablecos^?has been shown [27]

to have discriminating power against the background, and we use here the quantity E^?_T E_Tsin^2? which tends to have larger values in the signal in comparison to the background events; this effect is enhanced for the jets with higher E_T. The variables we choose as additional inputs to the neural network are then E^?;1_T and E^?;2_T for the two highest-E_Tjets, andhE^?_Tidefined as the geometric mean over the remainingN_jets2jets. The 11 variables used as inputs to the neural network are summarized in Table I. Comparisons of the background-dominated data and Monte Carlo generated signal events for the 11 kine- matical variables are shown in Figs. 1–5. The network is trained on same-size samples of signal and background events with6 N_jets 8(about 507 000 events). In order to model the signal we use the PYTHIAv6.2 [28] leading- order Monte Carlo generator with parton showering fol- lowed by a simulation of the CDF II detector. The refer- ence top quark mass chosen for the training is M_top175 GeV=c². The background is obtained from the multijet data events themselves, since the signal frac- tion at the initial stage is expected to be very small. Among

TABLE I. Input variables to the neural network.

Variable Description

PE_T Scalar sum of the transverse energies of all jets P

3E_T As above, except the two highest-E_Tjets

C Centrality

A Aplanarity

M^min_2j Minimum dijet invariant mass M^max_2j Maximum dijet invariant mass M^min_3j Minimum trijet invariant mass M^max_3j Maximum trijet invariant mass E^?;1_T ETsin^2?for the highest-E_T jet E^?;2_T ETsin^2? for the next-to-highest-E_Tjet hE^?_Ti Geometric mean over the remaining jets

Aplanarity

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Fraction/(0.005)

0 0.01 0.02 0.03 0.04 0.05 0.06

QCD t t

Centrality

0.5 0.6 0.7 0.8 0.9 1

Fraction/(0.005)

0 0.005 0.01 0.015

0.02

QCD

t t

FIG. 2. Aplanarity (top panel) and centrality (bottom panel) distributions in QCD multijet (solid histogram) and tt Monte Carlo (dashed histogram) events with 6 N_jets 8.

All histograms are normalized to unity.

2) (GeV/c

min

M2j

0 20 40 60 80 100

)2 Fraction/(1 GeV/c ₀

0.01 0.02 0.03 0.04 0.05 0.06

QCD t t

2) (GeV/c

max

M2j

0 100 200 300 400 500

)2 Fraction/(6 GeV/c 0

0.01 0.02 0.03

0.04 QCD

t t

FIG. 3. M^min_2j (top panel) andM^max_2j (bottom panel) distributions in QCD multijet (solid histogram) and ttMonte Carlo (dashed histogram) events with 6 N_jets 8. All histograms are nor- malized to unity.

2) (GeV/c

min

M3j

0 100 150 200

)2 Fraction/(2 GeV/c 0

0.01 0.02 0.03

0.04

QCD

t t

2) (GeV/c

max

M3j 0

0 5

200 400 600 800

)2 Fraction/(8 GeV/c 0

0.01 0.02 0.03

0.04

QCD

t t

FIG. 4. M^min_3j (top panel) andM^max_3j (bottom panel) distributions in QCD multijet (solid histogram) and ttMonte Carlo (dashed histogram) events with 6 N_jets 8. All histograms are nor- malized to unity.

(GeV) ET

100 200 300 400 Σ500 600

Fraction/(5 GeV)

0 0.01 0.02 0.03 0.04 0.05

QCD t t

(GeV) ET

Σ3

100 150 200 250 300 350

Fraction/(5 GeV)

0 0.02 0.04 0.06 0.08

QCD t t

FIG. 1. P

ET (top panel) andP

3ET (bottom panel) distribu- tions in QCD multijet (solid histogram) and tt Monte Carlo (dashed histogram) events with 6 N_jets 8. All histograms are normalized to unity.

pp tt. . .

the configurations investigated, the one which provides the largest expected S/B has two hidden layers with 20 and 10 hidden nodes, respectively, and 1 output node. The value of the output node, N_out, is the quantity we use as the dis- criminator between signal and background, as shown in Fig.6for the6 N_jets 8sample.

IV. TAGGINGbQUARKS IN THE MULTIJET SAMPLE

In order to estimate thettcontent in the event sample, we exploit the heavy flavor content of ttevents using a b-tagging algorithm based on secondary vertex reconstruc- tion as described in detail in [22,29]. The algorithm aims at the identification of jets containing a b-hadron state by reconstructing its decay vertex with at least two high- quality tracks with hits in the silicon vertex detector. A b-tagged jet (‘‘tag,’’ in brief ) must have an associated secondary vertex with a displacement from the primary vertex in the transverse plane larger than 7.5 times the typical resolution of the vertex displacement of about 190m. Since, as shown in the next section, the back- ground will be estimated in terms of inclusive tags rather than events, for the signal we do not consider the efficiency for tagging an event, but rather the quantity to be used in the cross section calculation is the average number of tags per event,n^ave_tag, which is related to the cross section usingtt Monte Carlo calculations. The tagging efficiencies for jets coming from the fragmentation ofb-,c-, or light-flavored quarks are corrected according to the efficiency seen in the

min

N

n

0.75 0.8 0.85 0.9 0.95 1 1.05

FIG. 7. Average number of tags, n^ave_tag, as a function of the thresholdN^min_out forttMonte Carlo events passing the preselection and with 6 N_jets 8. The dashed lines represent the 1- uncertainty onn^ave_tag.

(GeV)

*,1

ET

0 20 40 60 80 100 120 140 160 180 200

Fraction/(3 GeV) ⁰

0.01 0.02 0.03

0.04 QCD

t t

(GeV)

*,2

ET

0 20 40 60 80 100 120 140 160 180

Fraction/(2 GeV) ⁰

0.01 0.02 0.03

0.04 QCD

t t

> (GeV)

*

<ET

0 10 20 30 40 50

Fraction/(1 GeV) ⁰

0.02 0.04

0.06 QCD

t t

FIG. 5. Kinematical distributions in QCD multijet (solid his- togram) andttMonte Carlo (dashed histogram) events with6 N_jets 8. From top to bottom:E^?;1_T ,E^?;2_T , and hE^?_Ti. All histo- grams are normalized to unity.

N

0 0.2 0.4 0.6 0.8 1

Fraction/(0.02)

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

QCD t t

FIG. 6. Neural network output,N_out, for QCD multijet (solid histogram) andttMonte Carlo (dashed histogram) events with 6 N_jets 8. Histograms are normalized to unity. The neural network implementation that we use in the TMultiLayer- Perceptron produces an output which is not strictly bound between 0 and 1.

data, by a factor0:890:07forbjets and0:890:14for cjets, respectively. These factors are described in detail in [22].

We find that the average number of tags present in att event after the preselection depends on the choice of the lower threshold,N^min_out (see Fig.7) for the neural network outputN_out. For any value of this threshold the systematic uncertainty onn^ave_tag is dominated by the uncertainty of the correction factors for taggingbandcjets.

V. BACKGROUND ESTIMATE

The background for the tt multijet final state comes mainly from QCD production of heavy-quark pairs (bb andcc) and false tags in light-quark jets. Other standard model processes such asW=Zjets have a smaller pro- duction cross section and small acceptance due to the selection cuts. Because of the current uncertainties on the Monte Carlo generation of QCD multijet events, we esti- mate the background from the data themselves. Given the theoretical uncertainties on the production cross section for events generated with N partons at tree level, a more accurate background estimate is obtained from the data themselves rather than from theoretical prediction of cross

section and simulations. The tag rate per jet is defined as the probability of tagging a jet whose tracks are recon- structed in the vertex detector. This rate is extracted from events depleted in thettsignal and is used as an estimate of the rate expected in events with different jet multiplicity.

These depleted events, with exactly four jets passing the preselection (‘‘control sample’’), have S/B1=3600. This method intrinsically provides an inclusive estimate in terms of the number of tags rather than the number of tagged events. The tag rate per jet is evaluated in this control sample and is parametrized in terms of variables sensitive to both the tagging efficiency for true heavy- flavored objects and the rate of false tags. These variables are the jet E_T, the number of tracks reconstructed in the vertex detector and associated to the jet, N_trk^jet, and the number of primary vertices in the event, N_vert. The tag rates per jet as a function of these variables are shown in Fig. 8 for jets with at least two tracks within the vertex detector acceptance (‘‘fiducial’’ jets).

The tag rate estimates the probability that a given fidu- cial jet in the signal sample is tagged. Summing this probability over all fiducial jets, we obtain the expected number of tags from nonsignal processes, that is, QCD heavy- and light-flavored production altogether. Before the

(GeV) Jet ET

Tag rate

0.01 0.02 0.03 0.04 0.05 0.06

jet

Ntrk

Tag rate

0 0.02 0.04 0.06 0.08 0.1

Nvert

20 40 60 80 100 120 140 160 2 8 10 12

1 2

4 6

3 4 5 6 7 8 9

Tag rate

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04

FIG. 8. Tag rate for fiducial jets as a function of jetET,N^jet_trk, andNvert, as measured in the control sample with exactly four jets.

pp tt. . .

neural-network-based kinematical selection, the multijet sample is composed essentially of background events.

The accuracy of our modeling of the background pro- cesses, assuming all jets to be uncorrelated, is shown in Figs. 9–11 where we compare the N_out distributions for tags in the data to the expected background in events with 4, 5, and 6 or more jets. The disagreement between the total number of observed and expected tags, possibly due to the presence of a small sample of tt events, amounts to no more than 0.8%. This small discrepancy observed at high jet multiplicity is accounted for as a systematic uncertainty on the background estimate due to the different jet multi- plicity in the signal and control samples. The kinematical selection cutting onN_outalso changes the event character- istics with respect to those found in the sample with exactly four jets, where the parametrization has been derived. This selection modifies the jet-E_T and spectra so that the average tag rate per event for jets from QCD background becomes higher. However, the parametrization of the tag rate in terms of properties of the jet (E_T andN^jet_trk) is shown to describe this increase well. Residual biases due to the neural network selection are treated as systematic uncer- tainties on the background prediction considering, in the control sample with four jets, subsets of events withN_out N_out^min with N_out^min in the range 0.8–1.0. In this case the disagreement between the total number of observed tags and the expected background is no more than about 2.4%, due to the neural network kinematical selection. The total systematic uncertainty expected on the background esti-

mate amounts then to 2.5%, the quadrature sum of the two uncertainties. The contributions from running conditions, such as instantaneous luminosity and detector configura- tion, have been studied and found to be negligible.

Nout

0 0.2 0.4 0.6 0.8 1

Tags/(0.01)

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200

Data

Background

FIG. 11. N_outdistribution for tags in the data events with 6 to 8 jets, compared with the estimate from the tag rate parametriza- tion. Events with multiple tags have multiple entries.

N

0 0.2 0.4 0.6 0.8 1

Tags/(0.01)

0 500 1000 1500 2000 2500 3000 3500 4000

Data Background

FIG. 10. Noutdistribution for tags in the data events with 5 jets, compared with the estimate from the tag rate parametrization.

Events with multiple tags have multiple entries.

Nout

0 0.2 0.4 0.6 0.8 1

Tags/(0.01)

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400

Data Background

FIG. 9. N_outdistribution for tags in the data events with 4 jets, compared with the estimate from the tag rate parametrization.

Events with multiple tags have multiple entries.