Article
Reference
Measurement of B ( t→Wb )/ B ( t→Wq ) at the Collider Detector at Fermilab
CDF Collaboration
CAMPANELLI, Mario (Collab.), et al.
Abstract
We present a measurement of the ratio of top-quark branching fractions R=B(t→Wb)/B(t→Wq), where q can be a b, s, or a d quark, using lepton-plus-jets and dilepton data sets with an integrated luminosity of ∼162 pb−1 collected with the Collider Detector at Fermilab during Run II of the Tevatron. The measurement is derived from the relative numbers of tt events with different multiplicity of identified secondary vertices. We set a lower limit of R>0.61 at 95%
confidence level.
CDF Collaboration, CAMPANELLI, Mario (Collab.), et al . Measurement of B ( t→Wb )/ B ( t→Wq ) at the Collider Detector at Fermilab. Physical Review Letters , 2005, vol. 95, no. 10, p. 102002
DOI : 10.1103/PhysRevLett.95.102002
Available at:
http://archive-ouverte.unige.ch/unige:38292
Disclaimer: layout of this document may differ from the published version.
1 / 1
Measurement of B t ! Wb= B t ! Wq at the Collider Detector at Fermilab
D. Acosta,16J. Adelman,12T. Affolder,9T. Akimoto,54M. G. Albrow,15D. Ambrose,15S. Amerio,42D. Amidei,33 A. Anastassov,50K. Anikeev,15A. Annovi,44J. Antos,1M. Aoki,54G. Apollinari,15T. Arisawa,56J.-F. Arguin,32 A. Artikov,13W. Ashmanskas,15A. Attal,7F. Azfar,41P. Azzi-Bacchetta,42N. Bacchetta,42H. Bachacou,28W. Badgett,15 A. Barbaro-Galtieri,28G. J. Barker,25V. E. Barnes,46B. A. Barnett,24S. Baroiant,6G. Bauer,31F. Bedeschi,44S. Behari,24
S. Belforte,53G. Bellettini,44J. Bellinger,58A. Belloni,31E. Ben-Haim,15D. Benjamin,14A. Beretvas,15T. Berry,29 A. Bhatti,48M. Binkley,15D. Bisello,42M. Bishai,15R. E. Blair,2C. Blocker,5K. Bloom,33B. Blumenfeld,24A. Bocci,48 A. Bodek,47G. Bolla,46A. Bolshov,31D. Bortoletto,46J. Boudreau,45S. Bourov,15B. Brau,9C. Bromberg,34E. Brubaker,12
J. Budagov,13H. S. Budd,47K. Burkett,15G. Busetto,42P. Bussey,19K. L. Byrum,2S. Cabrera,14M. Campanelli,18 M. Campbell,33F. Canelli,7A. Canepa,46M. Casarsa,53D. Carlsmith,58R. Carosi,44S. Carron,14M. Cavalli-Sforza,3 A. Castro,4P. Catastini,44D. Cauz,53A. Cerri,28L. Cerrito,41J. Chapman,33Y. C. Chen,1M. Chertok,6G. Chiarelli,44 G. Chlachidze,13F. Chlebana,15I. Cho,27K. Cho,27D. Chokheli,13J. P. Chou,20S. Chuang,58K. Chung,11W.-H. Chung,58
Y. S. Chung,47M. Cijliak,44C. I. Ciobanu,23M. A. Ciocci,44A. G. Clark,18D. Clark,5M. Coca,14A. Connolly,28 M. Convery,48J. Conway,6B. Cooper,30K. Copic,33M. Cordelli,17G. Cortiana,42J. Cranshaw,52J. Cuevas,10A. Cruz,16
R. Culbertson,15C. Currat,28D. Cyr,58D. Dagenhart,5S. Da Ronco,42S. D’Auria,19P. de Barbaro,47S. De Cecco,49 A. Deisher,28G. De Lentdecker,47M. Dell’Orso,44S. Demers,47L. Demortier,48M. Deninno,4D. De Pedis,49 P. F. Derwent,15C. Dionisi,49J. R. Dittmann,15P. DiTuro,50C. Do¨rr,25A. Dominguez,28S. Donati,44M. Donega,18 J. Donini,42M. D’Onofrio,18T. Dorigo,42K. Ebina,56J. Efron,38J. Ehlers,18R. Erbacher,6M. Erdmann,25D. Errede,23
S. Errede,23R. Eusebi,47H.-C. Fang,28S. Farrington,29I. Fedorko,44W. T. Fedorko,12R. G. Feild,59M. Feindt,25 J. P. Fernandez,46R. D. Field,16G. Flanagan,34L. R. Flores-Castillo,45A. Foland,20S. Forrester,6G. W. Foster,15 M. Franklin,20J. C. Freeman,28Y. Fujii,26I. Furic,12A. Gajjar,29M. Gallinaro,48J. Galyardt,11M. Garcia-Sciveres,28
A. F. Garfinkel,46C. Gay,59H. Gerberich,14D. W. Gerdes,33E. Gerchtein,11S. Giagu,49P. Giannetti,44A. Gibson,28 K. Gibson,11C. Ginsburg,15K. Giolo,46M. Giordani,53M. Giunta,44G. Giurgiu,11V. Glagolev,13D. Glenzinski,15
M. Gold,36N. Goldschmidt,33D. Goldstein,7J. Goldstein,41G. Gomez,10G. Gomez-Ceballos,10M. Goncharov,51 O. Gonza´lez,46I. Gorelov,36A. T. Goshaw,14Y. Gotra,45K. Goulianos,48A. Gresele,42M. Griffiths,29C. Grosso-Pilcher,12
U. Grundler,23J. Guimaraes da Costa,20C. Haber,28K. Hahn,43S. R. Hahn,15E. Halkiadakis,47A. Hamilton,32 B.-Y. Han,47R. Handler,58F. Happacher,17K. Hara,54M. Hare,55R. F. Harr,57R. M. Harris,15F. Hartmann,25 K. Hatakeyama,48J. Hauser,7C. Hays,14H. Hayward,29B. Heinemann,29J. Heinrich,43M. Hennecke,25M. Herndon,24
C. Hill,9D. Hirschbuehl,25A. Hocker,15K. D. Hoffman,12A. Holloway,20S. Hou,1M. A. Houlden,29B. T. Huffman,41 Y. Huang,14R. E. Hughes,38J. Huston,34K. Ikado,56J. Incandela,9G. Introzzi,4M. Iori,49Y. Ishizawa,54C. Issever,9 A. Ivanov,6Y. Iwata,22B. Iyutin,31E. James,15D. Jang,50B. Jayatilaka,33D. Jeans,49H. Jensen,15E. J. Jeon,27M. Jones,46
K. K. Joo,27S. Y. Jun,11T. Junk,23T. Kamon,51J. Kang,33M. Karagoz Unel,37P. E. Karchin,57Y. Kato,40Y. Kemp,25 R. Kephart,15U. Kerzel,25V. Khotilovich,51B. Kilminster,38D. H. Kim,28H. S. Kim,23J. E. Kim,27M. J. Kim,11 M. S. Kim,27S. B. Kim,27S. H. Kim,54Y. K. Kim,12M. Kirby,14L. Kirsch,5S. Klimenko,16M. Klute,31B. Knuteson,31
B. R. Ko,14H. Kobayashi,54D. J. Kong,27K. Kondo,56J. Konigsberg,16K. Kordas,32A. Korn,31A. Korytov,16 A. V. Kotwal,14A. Kovalev,43J. Kraus,23I. Kravchenko,31A. Kreymer,15J. Kroll,43M. Kruse,14V. Krutelyov,51
S. E. Kuhlmann,2S. Kwang,12A. T. Laasanen,46S. Lai,32S. Lami,44S. Lammel,15M. Lancaster,30R. Lander,6 K. Lannon,38A. Lath,50G. Latino,44I. Lazzizzera,42C. Lecci,25T. LeCompte,2J. Lee,27J. Lee,47S. W. Lee,51R. Lefe`vre,3 N. Leonardo,31S. Leone,44S. Levy,12J. D. Lewis,15K. Li,59C. Lin,59C. S. Lin,15M. Lindgren,15E. Lipeles,8T. M. Liss,23 A. Lister,18D. O. Litvintsev,15T. Liu,15Y. Liu,18N. S. Lockyer,43A. Loginov,35M. Loreti,42P. Loverre,49R.-S. Lu,1
D. Lucchesi,42P. Lujan,28P. Lukens,15G. Lungu,16L. Lyons,41J. Lys,28R. Lysak,1E. Lytken,46D. MacQueen,32 R. Madrak,15K. Maeshima,15P. Maksimovic,24G. Manca,29F. Margaroli,4R. Marginean,15C. Marino,23A. Martin,59 M. Martin,24V. Martin,37M. Martı´nez,3T. Maruyama,54H. Matsunaga,54M. Mattson,57P. Mazzanti,4K. S. McFarland,47 D. McGivern,30P. M. McIntyre,51P. McNamara,50R. McNulty,29A. Mehta,29S. Menzemer,31A. Menzione,44P. Merkel,46 C. Mesropian,48A. Messina,49T. Miao,15N. Miladinovic,5J. Miles,31L. Miller,20R. Miller,34J. S. Miller,33C. Mills,9
R. Miquel,28S. Miscetti,17G. Mitselmakher,16A. Miyamoto,26N. Moggi,4B. Mohr,7R. Moore,15M. Morello,44 P. A. Movilla Fernandez,28J. Muelmenstaedt,28A. Mukherjee,15M. Mulhearn,31T. Muller,25R. Mumford,24A. Munar,43
P. Murat,15J. Nachtman,15S. Nahn,59I. Nakano,39A. Napier,55R. Napora,24D. Naumov,36V. Necula,16T. Nelson,15 C. Neu,43M. S. Neubauer,8J. Nielsen,28T. Nigmanov,45L. Nodulman,2O. Norniella,3T. Ogawa,56S. H. Oh,14Y. D. Oh,27
T. Ohsugi, T. Okusawa, R. Oldeman, R. Orava, W. Orejudos, K. Osterberg, C. Pagliarone, E. Palencia, R. Paoletti,44V. Papadimitriou,15A. A. Paramonov,12S. Pashapour,32J. Patrick,15G. Pauletta,53M. Paulini,11C. Paus,31
D. Pellett,6A. Penzo,53T. J. Phillips,14G. Piacentino,44J. Piedra,10K. T. Pitts,23C. Plager,7L. Pondrom,58G. Pope,45 X. Portell,3O. Poukhov,13N. Pounder,41F. Prakoshyn,13A. Pronko,16J. Proudfoot,2F. Ptohos,17G. Punzi,44 J. Rademacker,41M. A. Rahaman,45A. Rakitine,31S. Rappoccio,20F. Ratnikov,50H. Ray,33B. Reisert,15V. Rekovic,36
P. Renton,41M. Rescigno,49F. Rimondi,4K. Rinnert,25L. Ristori,44W. J. Robertson,14A. Robson,19T. Rodrigo,10 S. Rolli,55R. Roser,15R. Rossin,16C. Rott,46J. Russ,11V. Rusu,12A. Ruiz,10D. Ryan,55H. Saarikko,21S. Sabik,32 A. Safonov,6R. St. Denis,19W. K. Sakumoto,47G. Salamanna,49D. Saltzberg,7C. Sanchez,3L. Santi,53S. Sarkar,49 K. Sato,54P. Savard,32A. Savoy-Navarro,15P. Schlabach,15E. E. Schmidt,15M. P. Schmidt,59M. Schmitt,37T. Schwarz,33 L. Scodellaro,10A. L. Scott,9A. Scribano,44F. Scuri,44A. Sedov,46S. Seidel,36Y. Seiya,40A. Semenov,13F. Semeria,4
L. Sexton-Kennedy,15I. Sfiligoi,17M. D. Shapiro,28T. Shears,29P. F. Shepard,45D. Sherman,20M. Shimojima,54 M. Shochet,12Y. Shon,58I. Shreyber,35A. Sidoti,44A. Sill,52P. Sinervo,32A. Sisakyan,13J. Sjolin,41A. Skiba,25 A. J. Slaughter,15K. Sliwa,55D. Smirnov,36J. R. Smith,6F. D. Snider,15R. Snihur,32M. Soderberg,33A. Soha,6 S. V. Somalwar,50J. Spalding,15M. Spezziga,52F. Spinella,44P. Squillacioti,44H. Stadie,25M. Stanitzki,59B. Stelzer,32
O. Stelzer-Chilton,32D. Stentz,37J. Strologas,36D. Stuart,9J. S. Suh,27A. Sukhanov,16K. Sumorok,31H. Sun,55 T. Suzuki,54A. Taffard,23R. Tafirout,32H. Takano,54R. Takashima,39Y. Takeuchi,54K. Takikawa,54M. Tanaka,2
R. Tanaka,39N. Tanimoto,39M. Tecchio,33P. K. Teng,1K. Terashi,48R. J. Tesarek,15S. Tether,31J. Thom,15 A. S. Thompson,19E. Thomson,43P. Tipton,47V. Tiwari,11S. Tkaczyk,15D. Toback,51K. Tollefson,34T. Tomura,54
D. Tonelli,44M. To¨nnesmann,34S. Torre,44D. Torretta,15S. Tourneur,15W. Trischuk,32R. Tsuchiya,56S. Tsuno,39 D. Tsybychev,16N. Turini,44F. Ukegawa,54T. Unverhau,19S. Uozumi,54D. Usynin,43L. Vacavant,28A. Vaiciulis,47 A. Varganov,33S. Vejcik III,15G. Velev,15V. Veszpremi,46G. Veramendi,23T. Vickey,23R. Vidal,15I. Vila,10R. Vilar,10
I. Vollrath,32I. Volobouev,28M. von der Mey,7P. Wagner,51R. G. Wagner,2R. L. Wagner,15W. Wagner,25R. Wallny,7 T. Walter,25Z. Wan,50M. J. Wang,1S. M. Wang,16A. Warburton,32B. Ward,19S. Waschke,19D. Waters,30T. Watts,50 M. Weber,28W. C. Wester III,15B. Whitehouse,55D. Whiteson,43A. B. Wicklund,2E. Wicklund,15H. H. Williams,43 P. Wilson,15B. L. Winer,38P. Wittich,43S. Wolbers,15C. Wolfe,12M. Wolter,55M. Worcester,7S. Worm,50T. Wright,33 X. Wu,18F. Wu¨rthwein,8A. Wyatt,30A. Yagil,15T. Yamashita,39K. Yamamoto,40J. Yamaoka,50C. Yang,59U. K. Yang,12 W. Yao,28G. P. Yeh,15J. Yoh,15K. Yorita,56T. Yoshida,40I. Yu,27S. Yu,43J. C. Yun,15L. Zanello,49A. Zanetti,53I. Zaw,20
F. Zetti,44J. Zhou,50and S. Zucchelli4 (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
4Istituto Nazionale di Fisica Nucleare, University of Bologna, I-40127 Bologna, Italy
5Brandeis University, Waltham, Massachusetts 02254, USA
6University of California, Davis, Davis, California 95616, USA
7University of California, Los Angeles, Los Angeles, California 90024, USA
8University of California, San Diego, La Jolla, California 92093, USA
9University of California, Santa Barbara, Santa Barbara, California 93106, USA
10Instituto de Fisica de Cantabria, CSIC-University of Cantabria, 39005 Santander, Spain
11Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, USA
12Enrico Fermi Institute, University of Chicago, Chicago, Illinois 60637, USA
13Joint Institute for Nuclear Research, RU-141980 Dubna, Russia
14Duke University, Durham, North Carolina 27708, USA
15Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA
16University of Florida, Gainesville, Florida 32611, USA
17Laboratori Nazionali di Frascati, Istituto Nazionale di Fisica Nucleare, I-00044 Frascati, Italy
18University of Geneva, CH-1211 Geneva 4, Switzerland
19Glasgow University, Glasgow G12 8QQ, United Kingdom
20Harvard University, Cambridge, Massachusetts 02138, USA
21Division of High Energy Physics, Department of Physics, University of Helsinki and Helsinki Institute of Physics, FIN-00014, Helsinki, Finland
22Hiroshima University, Higashi-Hiroshima 724, Japan
23University of Illinois, Urbana, Illinois 61801, USA
102002-2
24The Johns Hopkins University, Baltimore, Maryland 21218, USA
25Institut fu¨r Experimentelle Kernphysik, Universita¨t Karlsruhe, 76128 Karlsruhe, Germany
26High Energy Accelerator Research Organization (KEK), Tsukuba, Ibaraki 305, Japan
27Center for High Energy Physics: Kyungpook National University, Taegu 702-701; Seoul National University, Seoul 151-742;
and SungKyunKwan University, Suwon 440-746; Korea
28Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA
29University of Liverpool, Liverpool L69 7ZE, United Kingdom
30University College London, London WC1E 6BT, United Kingdom
31Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA
32Institute of Particle Physics: McGill University, Montre´al, Canada H3A 2T8; and University of Toronto, Toronto, Canada M5S 1A7
33University of Michigan, Ann Arbor, Michigan 48109, USA
34Michigan State University, East Lansing, Michigan 48824, USA
35Institution for Theoretical and Experimental Physics, ITEP, Moscow 117259, Russia
36University of New Mexico, Albuquerque, New Mexico 87131, USA
37Northwestern University, Evanston, Illinois 60208, USA
38The Ohio State University, Columbus, Ohio 43210, USA
39Okayama University, Okayama 700-8530, Japan
40Osaka City University, Osaka 588, Japan
41University of Oxford, Oxford OX1 3RH, United Kingdom
42University of Padova, Istituto Nazionale di Fisica Nucleare, Sezione di Padova-Trento, I-35131 Padova, Italy
43University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA
44Istituto Nazionale di Fisica Nucleare Pisa, Universities of Pisa, Siena and Scuola Normale Superiore, I-56127 Pisa, Italy
45University of Pittsburgh, Pittsburgh, Pennsylvania 15260, USA
46Purdue University, West Lafayette, Indiana 47907, USA
47University of Rochester, Rochester, New York 14627, USA
48The Rockefeller University, New York, New York 10021, USA
49Istituto Nazionale di Fisica Nucleare, Sezione di Roma 1, University di Roma ‘‘La Sapienza’’, I-00185 Roma, Italy
50Rutgers University, Piscataway, New Jersey 08855, USA
51Texas A&M University, College Station, Texas 77843, USA
52Texas Tech. University, Lubbock, Texas 79409, USA
53Istituto Nazionale di Fisica Nucleare, University of Trieste/Udine, Italy
54University of Tsukuba, Tsukuba, Ibaraki 305, Japan
55Tufts University, Medford, Massachusetts 02155, USA
56Waseda University, Tokyo 169, Japan
57Wayne State University, Detroit, Michigan 48201, USA
58University of Wisconsin, Madison, Wisconsin 53706, USA
59Yale University, New Haven, Connecticut 06520, USA (Received 27 May 2005; published 29 August 2005)
We present a measurement of the ratio of top-quark branching fractionsRBt!Wb=Bt!Wq, whereq can be ab,s, or ad quark, using lepton-plus-jets and dilepton data sets with an integrated luminosity of162 pb1collected with the Collider Detector at Fermilab during Run II of the Tevatron.
The measurement is derived from the relative numbers ofttevents with different multiplicity of identified secondary vertices. We set a lower limit ofR >0:61at 95% confidence level.
DOI:10.1103/PhysRevLett.95.102002 PACS numbers: 14.65.Ha, 12.15.Hh
The top quark as described by the standard model (SM) is expected to decay to aW boson and a bottom quark at least 99.8% of the time at 90% confidence level (C.L.) [1].
The Cabibbo-Kobayashi-Maskawa (CKM) quark-mixing matrix [2,3] element jVtbjis expected to be very close to unity from the assumption of a unitary, three-generation matrix and the measured small values ofjVubjandjVcbj [1]. A measurement of the ratio of top-quark branching fractionsRBt!Wb=Bt!Wq, whereqcan be a b, s, or a d quark, significantly less than unity would contradict our current theoretical assumptions, implying either non-SM top decay, a non-SM background to top-pair production, or a fourth generation of quarks. A previous
measurement has set a lower limit ofR >0:56at 95% C.L.
[4]. In this Letter we present a measurement ofRusingtt events collected at the Collider Detector at Fermilab (CDF) during Run II of the Tevatron, a proton-antiproton collider with center of mass energy of ps
1:96 TeV. The inte- grated luminosity of the data sample used in this analysis is 162 pb1.
Our measurement usestt-pair events. The lifetime of top is too short for hadronization to occur, and the SM strongly favors an essentially immediate decay of each quark to a realWboson and weak-isospin1=2quark; ifR1, this is always abquark. To maintain high detection and trigger efficiencies and low background levels, we only considertt
final states in which at least oneW has decayed leptoni- cally. Events in which oneWdecays leptonically are called
‘‘lepton-plus-jets’’ L J events, and events with two leptonic decays are called ‘‘dilepton’’ (DIL) events.
Values of R are determined separately for each of these sets of events, and are combined in the end to set a lower limit on R. The greater statistical power comes from the L Jsample.
The measurement requires both the counting ofb-quark jets and the determination of thettcontent as a function of theb-quark multiplicity. We identify (‘‘tag’’)b-quark jets by identifying displaced secondary vertices using the
SECVTXalgorithm [5].Ris extracted from the relative rates of events with zero, one, and two tags; any two rates determine R uniquely, while all three rates jointly over- determine R. A novel feature of this measurement is the inclusion of the 0-tagL Jevent rate, which is determined using event kinematics and an artificial neural net (ANN) technique. AsR depends only on relative rates, this mea- surement is independent of any assumptions of the overall tt cross section. However, our measurement of R does depend critically on the knowledge of the efficiency to identify bjets. To extract R we use the efficiency to tag jets inttevents estimated with a Monte Carlo (MC) sample in which tagging efficiencies have been tuned to match jet data [5].
The CDF detector for Run II [6] consists of a charged- particle tracking system in a magnetic field of 1.4 T, seg- mented electromagnetic and hadronic calorimeters and muon detectors. A silicon microstrip detector provides tracking over the radial range 1.5 to 28 cm, and is essential for the detection of displaced secondary vertices. The fiducial region of the silicon detector covers the pseudor- apidity range jj<2, while the central tracking system and muon chambers provide coverage forjj<1[7]. A three-level trigger system is used to select events with electron (muon) candidates with ET pT>18 GeV (18 GeV=c), which form the data set for this analysis.
L Jevents consist of one isolated high-pTlepton (eor ), large missing transverse energy (6ET) due to the un- detected neutrino, and four hadronic jets. Two of these jets arise from the hadronic decay of the otherW, and the other two arise from the top-daughter quarks q. The L J selection requirements are described in detail elsewhere [5]. Briefly, we require the presence of an isolated lepton which has transverse momentum greater than 20 GeV=c, that6ET is at least 20 GeV, and that there is a minimum of four jets, clustered with a cone-based algorithm having
cone dimension R
2 2
p 0:4, within
jj<2 and with corrected transverse energy [5] greater than 15 GeV. These requirements select 107 events.
DIL events consist of two charged leptons (ee,, or e), large6ETdue to the undetected neutrinos, and two jets from the top-daughter quarksq. The DIL selection require- ments are described in detail elsewhere [8]. Compared to
the L Jselection, we demand an additional lepton, but only a minimum of two energy-corrected [8] jets, with the same requirements as before. These requirements select 11 events.
Both event samples are subdivided on the basis of the number of identified b jets in the event. The number of events in each subsample with i tagged jets are given in Table I. The 2-tag subsample is defined to include events with 2 tags; in this data sample we observe no events with more than two tagged jets.
In the L J sample, the dominant background is W production in association with jets from QCD processes (‘‘W jets’’ events). In the 1-tag and 2-tag subsamples we make an a priori estimate of the backgrounds with a collection of data-driven and simulation techniques that are described in detail elsewhere [5]. The backgrounds in these subsamples includeWproduction in association with heavy-flavor jets (Wbb, Wcc, Wc),W production in asso- ciation with light-flavor jets that are incorrectly identified as b jets (‘‘mistags’’), QCD multijet (‘‘QCD’’) events containing fake or real leptons and/or incorrectly- measured 6ET, dibosons (WW, WZ), and single-top-quark production. The background estimate requires a small correction forR1. The background estimate forR1 in these subsamples is given in Table I. The uncertainties on the estimate are dominated by uncertainties in the fraction of W jets events that include heavy flavor and on the normalization of the QCD background rate.
By construction, thea priorimethod cannot predict the background level in the 0-tag L J sample, where the W jets production rate dominates that for ttpairs; in- stead we make use of event kinematics [9]. The artificial neural net [10] is trained with thettsignal (HERWIG[11] ) andW jets background (HERWIG ALPGEN[12] ) events simulated with a detailed detector description based on
GEANT[13]. There is an additional QCD background which is modeled using data with nonisolated leptons. We find optimal signal to background discrimination with an ANN structure of nine input variables, one intermediate layer with ten nodes, and one output unit. The variables used are the transverse energies of the four leading jets, the mini- mum di-jet mass, the di-jet transverse mass with value closest to the mass of the W, the scalar sum of the trans- verse energies of all leptons and jets, the total longitudinal momentum divided by the total transverse momentum, and the event aplanarity.
The ANN output ranges from zero for backgroundlike events to one for signal-like events. We perform a binned maximum likelihood fit of the ANN output distribution for the tt fraction in the 0-tag subsample. The fraction of events from QCD backgrounds is fixed to 11.4% in this fit. These events are characterized by the nonisolation of the lepton and small 6ET, and the fixed rate is based on comparing to control regions with either low 6ET or poor isolation [5]. The resulting measurement of background
102002-4
rates in the 0-tagL Jsubsample is given in Table I. The fit of the distribution of ANN outputs for this subsample is shown in Fig. 1.
Systematic uncertainties in the ANN-determined back- grounds are dominated by our understanding of the jet energy scale, the renormalization and factorization scale, and the shape of the QCD template and are strongly anti- correlated between thettandW jets measurements. Our ANN-measuredttcontent in theL Jsample without any tagging requirement is consistent with that found in our earlier measurement of thettproduction cross section [14].
The procedure is repeated in the 1-tag and 2-tag samples, yielding background rates of5:85:2and0:1 1:00:1, respec- tively, consistent with the a priori estimates shown in Table I. As thea prioriestimates have smaller uncertainties in the 1-tag and 2-tag subsamples, the ANN-determined background level is used only for the L J 0-tag subsample.
The main backgrounds in the DIL sample are Drell-Yan production including lepton pairs from the Z resonance, dibosons, andW jets events with fake leptons. The total background level of2:20:6events in the DIL sample has been estimated elsewhere [8]. The Drell-Yan rate ineeand events is estimated using simulated data normalized to the observed rate ofZevents in the data. Other electroweak backgrounds are estimated from MC simulations. The fake-lepton background is estimated by multiplying each jet inWplus three or more jet events by a lepton fake rate, measured in complementary jet samples.
Most of the jets in the DIL background events arise from generic QCD radiation. To determine the background dis- tribution across thei-tag subsamples, we apply a parame- trization of the probability to tag a generic QCD jet [5], derived from jet-triggered data samples, to the jets in the DIL sample, correcting for the enrichedttcontent of the sample. The resulting estimates are given in Table I; the background in the 2-tag subsample is negligible.
Thettevent-tagging efficiencyi, defined as the proba- bility to observeitags in attevent, depends on the fiducial acceptances for jets that can potentially be tagged, and the efficiencies to tag those jets [9]. Those efficiencies in turn depend on the species of the underlying quark in the jet.
The efficiencyidepends strongly onR, asR1implies fewerbjets available for tagging, and more light-quark jets available instead. We use the jet acceptances and tagging efficiencies to parametrizeiR. These quantities are esti- mated with a sample of simulated tt events from the
PYTHIA [15] generator and CDF detector simulation, and their uncertainties are dominated by our understanding of the control samples of jet data used to calibrate tagging TABLE I. Summary of observed number of events withitags in theL Jand DIL samples, with estimates of nominalttevent- tagging efficiencies, background levels, and expected event yields. The L J0-tag background is measured with an ANN. The efficiency estimates and the 1-tag and 2-tagL Jbackground estimates are given forR1. Equations (1) and (2) are used for the calculation of the expected total number of eventsNexpi . The statistical and systematic uncertainties have been combined.
0-tag 1-tag 2-tag
Lepton Jets (L J)
EfficiencyiR1 0:450:03 0:430:02 0:120:02
Background (Nbkgi ) 62:49:0 4:20:7 0:20:1
Total expected (Nexpi ) 80:45:2 21:54:1 5:01:4
Observed (Nobsi ) 79 23 5
Dileptons (DIL)
EfficiencyiR1 0:470:03 0:430:02 0:100:02
BackgroundNbkgi 2:00:6 0:20:1 <0:01
Total expectedNexpi 6:10:4 4:00:2 0:90:2
ObservedNobsi 5 4 2
FIG. 1 (color online). Fit of the ANN output in the 0-tagL J data set (triangles) with a sum of 3 components: W jets (upper), QCD multijet (middle), and tt (lower). The QCD normalization is independently estimated and not varied in the fit; its shape is determined from the nonisolated lepton data.
efficiencies in the simulation. The leading determiner ofi is the efficiency to tag abjet from the decayt!Wb;b 0:440:04forbjets falling within the fiducial acceptance and having at least two tracks with silicon information. The i values also have small contributions from the efficien- cies to tag jets from W !cs hadronic decays and from additional QCD radiation inttevents. The nominal values ofiforR1are given in Table I. The value of0 2 changes by0:23(0.09) asRchanges from 0.5 to 1.
The expected event yield in each of the three tagged subsets of each of theL Jand DIL samples is
Nexpi Ninctt iR Nibkg; (1) whereNibkgis the number of background events in thei-tag subsample andNtinct is an estimate of the inclusive number ofttevents in the sample, determined by
Ninctt X
i
NiobsNibkg; (2)
where Nobsi is the observed number of events in each subsample. In this construction, the measured value ofR is independent of any assumption of the overall rate oftt production, and is thus sensitive only to the relative num- bers ofttevents withitags.
The full likelihood is a product of independent likeli- hoods for theL Jand DIL samples. Each likelihood is a product of Poisson functions comparing Niobs to Niexp for each value of i, multiplied by Gaussian functions which incorporate systematic uncertainties in the event-tagging efficiencies and backgrounds, taking into account the cor- relations across the different subsamples. These include correlations in the event-tagging efficiencies through the single-jet tagging efficiencies, in the common methodol- ogy of thea prioriestimates in the taggedL Jsamples, and in the overall normalization of the DIL backgrounds.
There are a total of five free parameters in the likelihood to account for these systematic uncertainties.
The resulting likelihood as a function ofRis shown in Fig. 2, along with the negative logarithm of the likelihood.
We find a central value of R1:12 0:210:19stat 0:170:13syst.
The dominant systematic uncertainties arise from the un- certainty on the background measurement in the 0-tag L J sample 0:140:11 and from the overall normalization of the tagging efficiencies 0:090:06. Taken separately, the two final states of ttgive consistent results for R; the L J sample alone yields R1:02 0:23 0:210:200:13, and the DIL sample alone yields R1:41 0:46 0:170:400:13. These R results are consistent with the SM expectations.
The ratio R can only take on physical values between zero and unity. We use the Feldman-Cousins prescription [16] to set a lower limit onR. We generate ensembles of pseudoexperiments for different input values of R(Rtrue), and vary the input quantities of the analysis, e.g., the background estimates, taking correlations into account.
Using the likelihood-ratio ordering principle, we find the acceptance intervals as shown in Fig. 2. With our measured value ofR, we find thatR >0:61at the 95% C.L..
Our lower limit onRis the strongest limit on this top- quark branching ratio to date. Within the SM, R
jVtbj2
jVtbj2 jVtsj2 jVtdj2, up to phase-space factors. Assuming three generations and the unitarity of the CKM matrix, the denominator is unity, and we estimate jVtbj>0:78 at 95% C.L.. All of our measurements of R are consistent with the SM expectations.
We thank the Fermilab staff and the technical staffs of the participating institutions for their vital contributions.
This work was supported by the U.S. Department of Energy and National Science Foundation; the Italian Istituto Nazionale di Fisica Nucleare; the Ministry of Edu- cation, Culture, Sports, Science, and Technology of Japan;
the Natural Sciences and Engineering Research Council of Canada; the National Science Council of the Republic of China; the Swiss National Science Foundation; the A. P.
Sloan Foundation; the Bundesministerium fu¨r Bildung und Forschung, Germany; the Korean Science and Engineering Foundation and the Korean Research Foundation; the Particle Physics and Astronomy Research Council and the Royal Society, UK; the Russian Foundation for Basic Research; the Comision Interministerial de Ciencia y Tecnologı´a, Spain; in part by the European Community’s Human Potential Programme under Contract No. HPRN- CT-2002-00292; and the Academy of Finland.
-ln(L)
0 1 2 3
CDF II R = 1.12
- 0.23 + 0.270.5 1 1.5 2 2.5
L
0 0.5 1
R
0.5 1 1.5R
trueR
0.4 0.6 0.8
1 0.5 1 1.5 R
95% C.L.
90% C.L.
68% C.L.
FIG. 2 (color online). The upper plot shows the likelihood as a function ofR(inset) and its negative logarithm. The intersections of the horizontal linelnL 0:5with the likelihood define the statistical 1 errors onR. The lower plot shows 95% (outer), 90% (central), and 68% (inner) C.L. bands forRtrueas a function ofR. Our measurement ofR1:12(vertical line) impliesR >
0:61at the 95% C.L. (horizontal line).
102002-6
[1] S. Eidelman et al. (Particle Data Group), Phys. Lett. B 592, 1 (2004).
[2] N. Cabibbo, Phys. Rev. Lett.10, 531 (1963).
[3] M. Kobayashi and T. Maskawa, Prog. Theor. Phys.49, 652 (1973).
[4] T. Affolderet al.(CDF Collaboration), Phys. Rev. Lett.86, 3233 (2001).
[5] D. Acosta et al. (CDF Collaboration), Phys. Rev. D 71, 052003 (2005).
[6] D. Acosta et al. (CDF Collaboration), Phys. Rev. D 71, 032001 (2005).
[7] In the CDF geometry,is the polar angle with respect to the proton beam axis, andis the azimuthal angle. The pseudorapidity is lntan=2. The transverse mo- mentum,pT, is the component of the momentum projected onto the plane perpendicular to the beam axis. The trans- verse energy ET of a shower or calorimeter tower is Esin, whereEis the energy deposited.
[8] D. Acostaet al.(CDF Collaboration), Phys. Rev. Lett.93, 142001 (2004); we restrict the193 pb1data set used here to runs in which the silicon detector was included.
[9] D. Smirnov, Ph.D. thesis, University of New Mexico, 2005.
[10] A. Zellet al., SNNS: Stuttgart Neural Network Simula- tor (2001), user Manual, V4.2. See also http://www-ra.
informatik.uni-tuebingen.de/SNNS/.
[11] G. Corcellaet al., J. High Energy Phys. 01 (2001) 010.
[12] M. L. Manganoet al., J. High Energy Phys. 07 (2003) 001.
[13] R. Brun and F. Carminati, CERN Programming Library Long Writeup No. W5013, 1993 (unpublished).
[14] D. Acostaet al.(CDF Collaboration) hepex/0504053.
[15] T. Sjo¨strand et al., Comput. Phys. Commun. 135, 238 (2001).
[16] G. Feldman and R. Cousins, Phys. Rev. D 57, 3873 (1998).
