Contents lists available atScienceDirect
Physics
Letters
B
www.elsevier.com/locate/physletb
Precise
measurement
of
the
top
quark
mass
in
dilepton
decays
using
optimized
neutrino
weighting
D0
Collaboration
V.M. Abazov
af,
B. Abbott
bp,
B.S. Acharya
z,
M. Adams
au,
T. Adams
as,
J.P. Agnew
ap,
G.D. Alexeev
af,
G. Alkhazov
aj,
A. Alton
be,
1,
A. Askew
as,
S. Atkins
bc,
K. Augsten
g,
C. Avila
e,
F. Badaud
j,
L. Bagby
at,
B. Baldin
at,
D.V. Bandurin
bv,
S. Banerjee
z,
E. Barberis
bd,
P. Baringer
bb,
J.F. Bartlett
at,
U. Bassler
o,
V. Bazterra
au,
A. Bean
bb,
M. Begalli
b,
L. Bellantoni
at,
S.B. Beri
x,
G. Bernardi
n,
R. Bernhard
t,
I. Bertram
an,
M. Besançon
o,
R. Beuselinck
ao,
P.C. Bhat
at,
S. Bhatia
bg,
V. Bhatnagar
x,
G. Blazey
av,
S. Blessing
as,
K. Bloom
bh,
A. Boehnlein
at,
D. Boline
bm,
E.E. Boos
ah,
G. Borissov
an,
M. Borysova
am,
12,
A. Brandt
bs,
O. Brandt
u,
R. Brock
bf,
A. Bross
at,
D. Brown
n,
X.B. Bu
at,
M. Buehler
at,
V. Buescher
v,
V. Bunichev
ah,
S. Burdin
an,
2,
C.P. Buszello
al,
E. Camacho-Pérez
ac,
B.C.K. Casey
at,
H. Castilla-Valdez
ac,
S. Caughron
bf,
S. Chakrabarti
bm,
K.M. Chan
az,
A. Chandra
bu,
E. Chapon
o,
G. Chen
bb,
S.W. Cho
ab,
S. Choi
ab,
B. Choudhary
y,
S. Cihangir
at,
D. Claes
bh,
J. Clutter
bb,
M. Cooke
at,
11,
W.E. Cooper
at,
M. Corcoran
bu,
F. Couderc
o,
M.-C. Cousinou
l,
J. Cuth
v,
D. Cutts
br,
A. Das
bt,
G. Davies
ao,
S.J. de Jong
ad,
ae,
E. De La Cruz-Burelo
ac,
F. Déliot
o,
R. Demina
bl,
D. Denisov
at,
S.P. Denisov
ai,
S. Desai
at,
C. Deterre
ap,
3,
K. DeVaughan
bh,
H.T. Diehl
at,
M. Diesburg
at,
P.F. Ding
ap,
A. Dominguez
bh,
A. Dubey
y,
L.V. Dudko
ah,
A. Duperrin
l,
S. Dutt
x,
M. Eads
av,
D. Edmunds
bf,
J. Ellison
ar,
V.D. Elvira
at,
Y. Enari
n,
H. Evans
ax,
A. Evdokimov
au,
V.N. Evdokimov
ai,
A. Fauré
o,
L. Feng
av,
T. Ferbel
bl,
F. Fiedler
v,
F. Filthaut
ad,
ae,
W. Fisher
bf,
H.E. Fisk
at,
M. Fortner
av,
H. Fox
an,
S. Fuess
at,
P.H. Garbincius
at,
A. Garcia-Bellido
bl,
J.A. García-González
ac,
V. Gavrilov
ag,
W. Geng
l,
bf,
C.E. Gerber
au,
Y. Gershtein
bi,
G. Ginther
at,
bl,
O. Gogota
am,
G. Golovanov
af,
P.D. Grannis
bm,
S. Greder
p,
H. Greenlee
at,
G. Grenier
q,
r,
Ph. Gris
j,
J.-F. Grivaz
m,
A. Grohsjean
o,
3,
S. Grünendahl
at,
M.W. Grünewald
aa,
T. Guillemin
m,
G. Gutierrez
at,
P. Gutierrez
bp,
J. Haley
bq,
L. Han
d,
K. Harder
ap,
A. Harel
bl,
J.M. Hauptman
ba,
J. Hays
ao,
T. Head
ap,
T. Hebbeker
s,
D. Hedin
av,
H. Hegab
bq,
A.P. Heinson
ar,
U. Heintz
br,
C. Hensel
a,
I. Heredia-De La Cruz
ac,
4,
K. Herner
at,
G. Hesketh
ap,
6,
M.D. Hildreth
az,
R. Hirosky
bv,
T. Hoang
as,
J.D. Hobbs
bm,
B. Hoeneisen
i,
J. Hogan
bu,
M. Hohlfeld
v,
J.L. Holzbauer
bg,
I. Howley
bs,
Z. Hubacek
g,
o,
V. Hynek
g,
I. Iashvili
bk,
Y. Ilchenko
bt,
R. Illingworth
at,
A.S. Ito
at,
S. Jabeen
at,
13,
M. Jaffré
m,
A. Jayasinghe
bp,
M.S. Jeong
ab,
R. Jesik
ao,
P. Jiang
d,
K. Johns
aq,
E. Johnson
bf,
M. Johnson
at,
A. Jonckheere
at,
P. Jonsson
ao,
J. Joshi
ar,
A.W. Jung
at,
A. Juste
ak,
E. Kajfasz
l,
D. Karmanov
ah,
I. Katsanos
bh,
M. Kaur
x,
R. Kehoe
bt,
S. Kermiche
l,
N. Khalatyan
at,
A. Khanov
bq,
A. Kharchilava
bk,
Y.N. Kharzheev
af,
I. Kiselevich
ag,
J.M. Kohli
x,
A.V. Kozelov
ai,
J. Kraus
bg,
A. Kumar
bk,
A. Kupco
h,
T. Kurˇca
q,
r,
V.A. Kuzmin
ah,
S. Lammers
ax,
P. Lebrun
q,
r,
H.S. Lee
ab,
S.W. Lee
ba,
W.M. Lee
at,
X. Lei
aq,
J. Lellouch
n,
E-mailaddress:kehoe@physics.smu.edu(R. Kehoe).
http://dx.doi.org/10.1016/j.physletb.2015.10.086
0370-2693/©2015TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP3.
D. Li
n,
H. Li
bv,
L. Li
ar,
Q.Z. Li
at,
J.K. Lim
ab,
D. Lincoln
at,
J. Linnemann
bf,
V.V. Lipaev
ai,
R. Lipton
at,
H. Liu
bt,
Y. Liu
d,
A. Lobodenko
aj,
M. Lokajicek
h,
R. Lopes de Sa
at,
R. Luna-Garcia
ac,
7,
A.L. Lyon
at,
A.K.A. Maciel
a,
R. Madar
t,
R. Magaña-Villalba
ac,
S. Malik
bh,
V.L. Malyshev
af,
J. Mansour
u,
J. Martínez-Ortega
ac,
R. McCarthy
bm,
C.L. McGivern
ap,
M.M. Meijer
ad,
ae,
A. Melnitchouk
at,
D. Menezes
av,
P.G. Mercadante
c,
M. Merkin
ah,
A. Meyer
s,
J. Meyer
u,
9,
F. Miconi
p,
N.K. Mondal
z,
M. Mulhearn
bv,
E. Nagy
l,
M. Narain
br,
R. Nayyar
aq,
H.A. Neal
be,
J.P. Negret
e,
P. Neustroev
aj,
H.T. Nguyen
bv,
T. Nunnemann
w,
J. Orduna
bu,
N. Osman
l,
J. Osta
az,
A. Pal
bs,
N. Parashar
ay,
V. Parihar
br,
S.K. Park
ab,
R. Partridge
br,
5,
N. Parua
ax,
A. Patwa
bn,
10,
B. Penning
ao,
M. Perfilov
ah,
Y. Peters
ap,
K. Petridis
ap,
G. Petrillo
bl,
P. Pétroff
m,
M.-A. Pleier
bn,
V.M. Podstavkov
at,
A.V. Popov
ai,
M. Prewitt
bu,
D. Price
ap,
N. Prokopenko
ai,
J. Qian
be,
A. Quadt
u,
B. Quinn
bg,
P.N. Ratoff
an,
I. Razumov
ai,
I. Ripp-Baudot
p,
F. Rizatdinova
bq,
M. Rominsky
at,
A. Ross
an,
C. Royon
o,
P. Rubinov
at,
R. Ruchti
az,
G. Sajot
k,
A. Sánchez-Hernández
ac,
M.P. Sanders
w,
A.S. Santos
a,
8,
G. Savage
at,
M. Savitskyi
am,
L. Sawyer
bc,
T. Scanlon
ao,
R.D. Schamberger
bm,
Y. Scheglov
aj,
H. Schellman
aw,
M. Schott
v,
C. Schwanenberger
ap,
R. Schwienhorst
bf,
J. Sekaric
bb,
H. Severini
bp,
E. Shabalina
u,
V. Shary
o,
S. Shaw
ap,
A.A. Shchukin
ai,
V. Simak
g,
P. Skubic
bp,
P. Slattery
bl,
D. Smirnov
az,
G.R. Snow
bh,
J. Snow
bo,
S. Snyder
bn,
S. Söldner-Rembold
ap,
L. Sonnenschein
s,
K. Soustruznik
f,
J. Stark
k,
D.A. Stoyanova
ai,
M. Strauss
bp,
L. Suter
ap,
P. Svoisky
bp,
M. Titov
o,
V.V. Tokmenin
af,
Y.-T. Tsai
bl,
D. Tsybychev
bm,
B. Tuchming
o,
C. Tully
bj,
L. Uvarov
aj,
S. Uvarov
aj,
S. Uzunyan
av,
R. Van Kooten
ax,
W.M. van Leeuwen
ad,
N. Varelas
au,
E.W. Varnes
aq,
I.A. Vasilyev
ai,
A.Y. Verkheev
af,
L.S. Vertogradov
af,
M. Verzocchi
at,
M. Vesterinen
ap,
D. Vilanova
o,
P. Vokac
g,
H.D. Wahl
as,
M.H.L.S. Wang
at,
J. Warchol
az,
G. Watts
bw,
M. Wayne
az,
J. Weichert
v,
L. Welty-Rieger
aw,
M.R.J. Williams
ax,
14,
G.W. Wilson
bb,
M. Wobisch
bc,
D.R. Wood
bd,
T.R. Wyatt
ap,
Y. Xie
at,
R. Yamada
at,
S. Yang
d,
T. Yasuda
at,
Y.A. Yatsunenko
af,
W. Ye
bm,
Z. Ye
at,
H. Yin
at,
K. Yip
bn,
S.W. Youn
at,
J.M. Yu
be,
J. Zennamo
bk,
T.G. Zhao
ap,
B. Zhou
be,
J. Zhu
be,
M. Zielinski
bl,
D. Zieminska
ax,
L. Zivkovic
naLAFEX,CentroBrasileirodePesquisasFísicas,RiodeJaneiro,Brazil bUniversidadedoEstadodoRiodeJaneiro,RiodeJaneiro,Brazil cUniversidadeFederaldoABC,SantoAndré,Brazil
dUniversityofScienceandTechnologyofChina,Hefei,People’sRepublicofChina eUniversidaddelosAndes,Bogotá,Colombia
fCharlesUniversity,FacultyofMathematicsandPhysics,CenterforParticlePhysics,Prague,CzechRepublic gCzechTechnicalUniversityinPrague,Prague,CzechRepublic
hInstituteofPhysics,AcademyofSciencesoftheCzechRepublic,Prague,CzechRepublic iUniversidadSanFranciscodeQuito,Quito,Ecuador
jLPC,UniversitéBlaisePascal,CNRS/IN2P3,Clermont,France
kLPSC,UniversitéJosephFourierGrenoble1,CNRS/IN2P3,InstitutNationalPolytechniquedeGrenoble,Grenoble,France lCPPM,Aix-MarseilleUniversité,CNRS/IN2P3,Marseille,France
mLAL,UniversitéParis-Sud,CNRS/IN2P3,Orsay,France nLPNHE,UniversitésParisVIandVII,CNRS/IN2P3,Paris,France oCEA,Irfu,SPP,Saclay,France
pIPHC,UniversitédeStrasbourg,CNRS/IN2P3,Strasbourg,France qIPNL,UniversitéLyon1,CNRS/IN2P3,Villeurbanne,France rUniversitédeLyon,Lyon,France
sIII.PhysikalischesInstitutA,RWTHAachenUniversity,Aachen,Germany tPhysikalischesInstitut,UniversitätFreiburg,Freiburg,Germany
uII.PhysikalischesInstitut,Georg-August-UniversitätGöttingen,Göttingen,Germany vInstitutfürPhysik,UniversitätMainz,Mainz,Germany
wLudwig-Maximilians-UniversitätMünchen,München,Germany xPanjabUniversity,Chandigarh,India
yDelhiUniversity,Delhi,India
zTataInstituteofFundamentalResearch,Mumbai,India aaUniversityCollegeDublin,Dublin,Ireland
abKoreaDetectorLaboratory,KoreaUniversity,Seoul,RepublicofKorea acCINVESTAV,MexicoCity,Mexico
adNikhef,SciencePark,Amsterdam,TheNetherlands aeRadboudUniversityNijmegen,Nijmegen,TheNetherlands afJointInstituteforNuclearResearch,Dubna,Russia
agInstituteforTheoreticalandExperimentalPhysics,Moscow,Russia ahMoscowStateUniversity,Moscow,Russia
aiInstituteforHighEnergyPhysics,Protvino,Russia ajPetersburgNuclearPhysicsInstitute,St.Petersburg,Russia
alUppsalaUniversity,Uppsala,Sweden
amTarasShevchenkoNationalUniversityofKyiv,Kiev,Ukraine anLancasterUniversity,LancasterLA14YB,UnitedKingdom aoImperialCollegeLondon,LondonSW72AZ,UnitedKingdom apTheUniversityofManchester,ManchesterM139PL,UnitedKingdom aqUniversityofArizona,Tucson,AZ 85721,USA
arUniversityofCaliforniaRiverside,Riverside,CA 92521,USA asFloridaStateUniversity,Tallahassee,FL 32306,USA atFermiNationalAcceleratorLaboratory,Batavia,IL 60510,USA auUniversityofIllinoisatChicago,Chicago,IL 60607,USA avNorthernIllinoisUniversity,DeKalb,IL 60115,USA awNorthwesternUniversity,Evanston,IL 60208,USA axIndianaUniversity,Bloomington,IN 47405,USA ayPurdueUniversityCalumet,Hammond,IN 46323,USA azUniversityofNotreDame,NotreDame,IN 46556,USA baIowaStateUniversity,Ames,IA 50011,USA bbUniversityofKansas,Lawrence,KS 66045,USA bcLouisianaTechUniversity,Ruston,LA 71272,USA bdNortheasternUniversity,Boston,MA 02115,USA be
UniversityofMichigan,AnnArbor,MI 48109,USA bfMichiganStateUniversity,EastLansing,MI 48824,USA bgUniversityofMississippi,University,MS 38677,USA bhUniversityofNebraska,Lincoln,NE 68588,USA biRutgersUniversity,Piscataway,NJ 08855,USA bjPrincetonUniversity,Princeton,NJ 08544,USA bkStateUniversityofNewYork,Buffalo,NY 14260,USA blUniversityofRochester,Rochester,NY 14627,USA bmStateUniversityofNewYork,StonyBrook,NY 11794,USA bnBrookhavenNationalLaboratory,Upton,NY 11973,USA boLangstonUniversity,Langston,OK 73050,USA bpUniversityofOklahoma,Norman,OK 73019,USA bqOklahomaStateUniversity,Stillwater,OK 74078,USA brBrownUniversity,Providence,RI 02912,USA bsUniversityofTexas,Arlington,TX 76019,USA btSouthernMethodistUniversity,Dallas,TX 75275,USA buRiceUniversity,Houston,TX 77005,USA
bvUniversityofVirginia,Charlottesville,VA 22904,USA bwUniversityofWashington,Seattle,WA 98195,USA
a
r
t
i
c
l
e
i
n
f
o
a
b
s
t
r
a
c
t
Articlehistory:
Received13August2015
Receivedinrevisedform19October2015 Accepted31October2015
Availableonline11November2015 Editor:H.Weerts
Wemeasure thetopquarkmassindileptonfinalstatesoft¯t eventsinpp collisions¯ at√s=1.96 TeV, using data corresponding to an integrated luminosity of 9.7 fb−1 at the Fermilab Tevatron Collider. Theanalysisfeaturesacomprehensiveoptimizationoftheneutrinoweightingmethodtominimizethe statisticaluncertainties.Wealsoimprovethecalibrationofjetenergiesusingthecalibrationdetermined intt¯→lepton+jets events, whichreduces theotherwiselimitingsystematicuncertaintyfromthejet energyscale.Themeasuredtopquarkmassismt=173.32±1.36(stat)±0.85(syst)GeV.
©2015TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP3.
1 VisitorfromAugustanaCollege,SiouxFalls,SD,USA. 2 VisitorfromTheUniversityofLiverpool,Liverpool,UK. 3 VisitorfromDESY,Hamburg,Germany.
4 VisitorfromCONACyT,MexicoCity,Mexico. 5 VisitorfromSLAC,MenloPark,CA,USA.
6 VisitorfromUniversityCollegeLondon,London,UK.
7 VisitorfromCentrodeInvestigacionenComputacion–IPN,MexicoCity,Mexico. 8 VisitorfromUniversidadeEstadualPaulista,SãoPaulo,Brazil.
9 Visitorfrom KarlsruherInstitut fürTechnologie (KIT)–Steinbuch Centrefor Computing(SCC),D-76128Karlsruhe,Germany.
10 Visitor from Office ofScience, U.S. Departmentof Energy, Washington, D.C. 20585,USA.
11 VisitorfromAmericanAssociationfortheAdvancementofScience,Washington, D.C.20005,USA.
12 VisitorfromKievInstituteforNuclearResearch,Kiev,Ukraine. 13 VisitorfromUniversityofMaryland,CollegePark,Maryland20742,USA. 14 Visitor from European Organization for Nuclear Research (CERN), Geneva, Switzerland.
1. Introduction
The discovery of the top quark in 1995 [1,2] completed the three quark families of thestandard model (SM).Since then, the topquarkhasbeenoneofthefocalpointsoftheFermilabTevatron andoftheCERNLHCprograms.Thetopquark standsoutbecause ofitslargemass,mt,whichisafundamentalparameterintheSM. ItsYukawacouplingtotheHiggsboson,Yt
=
√
2mt
/
v,wherev is thevacuumexpectationvalueoftheHiggsfield,isclosetounity, implyingthatthetopquarkmayplayaspecialroleinelectroweak symmetry breaking.Inaddition,mt islinkedto theW andHiggs bosonmasses,MW andMH,throughradiativecorrections[3]. Fol-lowingtheHiggsbosondiscovery [4,5],aprecisemeasurementofmt provides atestoftheelectroweaksector oftheSMand infor-mation onwhetherouruniverseresidesinastableormetastable region of that theory [6–8]. The short lifetime of the top quark preventsitsconfinementinthestrongcolorfield,sincetopquarks decaybefore hadronizing.Thisallows a particularlyprecisestudy of pure quantumchromodynamic (QCD) effects.A comparisonof the measured mt and the mt extracted from cross section
mea-surements [9–12] may provide a probe of higher order andsoft QCDcorrectionstotheobservedmass[13].
Assuming the SM branching ratio of t
→
W b≈
100%, tt de-¯
cays yield distinct final state categoriesaccording tothe number ofchargedleptonswithhightransversemomentum(pT)fromW boson decays. Dilepton (2,
=
e orμ
) events, such asee, eμ
, andμμ
,withneutrinosfromtwo W→
ν
decays,are relatively rarebuthavelow background.We presenta measurementofmt using pp collider¯
datacollected withthe D0 detectoratthe Fer-milabTevatroncollider,correspondingtoanintegratedluminosity of9.7 fb−1,ineventswithtwohigh-pT electronsormuonsof op-posite electric charge. Two high-pT jets must also be observed, one of which must be identified as being consistent with orig-inating from a b quark. This analysis is based on our previous dilepton measurement [14], but with increased integrated lumi-nosityandmultipleoptimizationstoimprovetheprecision ofmt. Wereducethedominantstatisticalcontributiontotheuncertainty onmt throughan optimization ofthe methodsfor kinematic re-constructionandstatisticalanalysis.LackingadijetsignaturefromW
→
qq¯
,whichispresentint¯
t→
lepton+
jets (+
jets)events andwas used to improve the precision of jet energy calibration witha W massconstraint[15],previousdilepton analysesatthe Tevatronhavereacheda sensitivitylimitimposed bystandard jet calibrationmethods[16,17].Progressincalibratingjetenergiesin the dilepton channel [14] provides improved cross-checks across different channels and a more significant contribution from the dilepton channel to the world average mt [18]. For comparison, themostrecentmeasurementsofmt inthedileptonchannelfrom CDF, ATLAS, and CMS are, respectively, mt=
171.5±
1.9(stat)±
2.5(syst)GeV[19],mt=
173.79±
0.54(stat)±
1.30(syst)GeV[20], andmt=
172.50±
0.43(stat)±
1.46(syst)GeV[21].Inthis analy-sis,wesubstantiallyreducetheotherwisedominantuncertaintyin thejetenergyscalebyapplyingthemethodsofRef.[14].2. Detectoranddatasample
2.1.Detector
The D0 detector [22,23] has a central-tracking system, con-sisting ofa siliconmicrostrip tracker anda central fibertracker, both located within a 1.9 T superconducting solenoidal magnet, withdesigns optimizedfor identificationof the p
¯
p collision ver-tex and track reconstruction at pseudorapidities [24] of|
η
|
<
3 and|
η
|
<
2.5, respectively. The liquid-argon/uranium calorimeter has a central section covering|
η
|
≤
1.1, and two end sections that extend coverage to|
η
|
≈
4.2, with all three housed in sep-aratecryostats. An outer muon system, covering|
η
|
<
2, consists ofalayeroftrackingdetectorsandscintillationtriggercountersin frontof1.8 Tirontoroids,followedbytwosimilarlayersafterthe toroids.2.2.Objectreconstruction
Werequireelectronstosatisfyanidentificationcriterionbased onboosteddecisiontrees [25] usingcalorimeterandtracking in-formation. Muons must satisfy requirements that match hits in themuon system toa track inthe central trackingdetector that is required to have a small distance of closest approach to the beam axis [26]. We require hits in the muon layers inside and outsidethetoroid.Muonsandchargedhadronmomentaare mea-suredinthe centraltrackingdetector,whileelectron,photon(
γ
), jet, and charged hadron energies are measured in the calorime-ters. Muons must be isolated from jets and from nearby tracks. Electronsandmuonsmusthavetheir extrapolatedtrack trajecto-riesisolated fromcalorimeterenergydepositionsgreater than an energythreshold. Electrons andmuons must have pT>
15 GeV,and
|
η
|
<
2.5 and<
2.0,respectively.Wereconstructjetsusingan iterative, midpoint-seededcone algorithmwitha cone parameter ofR
cone=
0.5 [27]. Jets with embedded muons from the decayofb-hadrons requirean additionalcorrectionto jetenergyto ac-countfortheassociatedneutrino.Amultivariatediscriminant[28]
isusedtoidentifyjetsthatcontainab-hadron(i.e.,b jets)froma vertexdisplacedfromtheinteractionpoint.Wedefinethemissing transverse momentum (/ET) attributed to the escaping neutrinos asthenegative ofthevector sumofalltransversecomponentsof calorimeter cell energies, corrected for the measured muon mo-menta andthe response ofthe calorimeterto electrons. We also correct
/
ET fordetector response inthe jet energy calibration,as described below. Details of objectreconstruction are provided in Ref.[29].2.3. Standardjetenergycalibration
Wecalibratetheenergyofjetstobetheenergyoftheparticle jetsreconstructed using themidpoint algorithm [27].We correct fortheeffectsofthecalorimeterresponsetoparticleconstituents ofjets,energyleakingintotheconefromparticlesdirectedoutside it,aswellasenergydepositsoutsidetheconefromparticlesinside it[30].Chargedhadronshaveanenergy-dependentresponsethat is lower than that of electrons and photons. We therefore apply correctionsobtainedfrom
γ
+
jet eventstoaccountfortheenergy dependenceofthejet responseinthecentral|
η
|
region.We also applyarelativeη
-dependentcorrectionobtainedfromγ
+
jet and dijetevents.Weemploythesamemethodstocalibratejetenergies independentlyintheMonteCarlo(MC)simulationandindata.The MCisusedtohelpstudypotentialbiasesinthedata.We incorpo-rateacorrectionforjetsintheMCsimulationthataccountsforthe difference in single-particle response betweendataand MC. This procedure ensuresthat theflavor dependenceofthejet response in data is replicated in MC. In the MC we account for multiplepp interactions
¯
bycorrectingthejetenergytotheparticlelevelof onlythoseparticlesthataredirectedwithinthejetconeatparticle level.The typicalsystematicuncertaintyinthe energycalibration ofeach jetin thedileptonsample is2%. Thisprecision islimited bysystematicuncertaintiesoftheγ
+
jet methodinthe pT range ofjetsint¯
t events. Detailsaboutthis“standard jetenergyscale” calibration can be found in Ref. [30]. We require that jets havepT
>
20 GeV and|
η
|
<
2.5 after calibration,but before applying additionalcorrectionsfromtheW→
qq¯
constraintinthe+
jets channeldiscussedbelow.3. Absolutejetcalibrationfroma
W
→
qq¯
constraintAsinRef.[14],weapplyamultiplicativecorrectionfactortothe energyofjetsindatabasedonananalysisoft
¯
t→
+
jets events usingtheW→
qq¯
decaysasaconstraint.Application ofthis fac-tor, 1.0250±
0.0046(stat)
[15],improvesthe agreementbetween MCanddataandallowsustouseitsuncertaintytoreducethe un-certainty onthe absoluteenergyscaleby a factorof≈
4 relative tothestandard jetenergyscale,whileretainingitsη
and pT de-pendence.Toapply thisscale,whichcomesfromlight-quarkjets, tothedileptonsample,whichhasb jets,itisimportanttoensure that the variation in the ratio of data over MC jet response be-tweendifferentflavorsbeplacedonanequalfooting.Thestandard jet energy scale [30] achieves this on a jet-by-jetbasis by using single particlesinMCjetstocorrectthesimulationsothat ithas the samekinematic andflavor-dependentjet response as that in data.Thisensuresthattheenergiesofb jetsindileptonsimulated samplesagreewiththoseofb jetsinthedileptondatasample at the same level aslight-quark jets. Aside fromfragmentation dif-ferences between data and MC which are discussed below, thisapproach justifiesthe use ofthe
+
jets constraint inthe dilep-tonchannel.4. Eventselection
The tt candidate
¯
events in the ee andμμ
channels are re-quired to pass single-lepton triggers. The full suite of triggers is usedfor selectingeμ
events.The dileptoneventselection before optimizationis described inRef. [29]. We optimizethe selection based on MC events to provide the smallest expected statistical uncertaintyinmt.We requiretwo isolatedleptons withopposite electric charge. We require at least two jets, where at least one of the two jets with highest pT must be identified as a b jet. Forthe eμ
channel, ourselectionshave an efficiencyfor taggingb jets of 72%, and a light-quark mistag rate of 12% in the cen-tralregionin
η
.Thesame-flavor channelsemployslightlytighterb tagging requirements and thus have a few percent lower
effi-ciency, and30% lower mistag rate. Werequire eventsin the
μμ
channeltohave/
ET>
40 GeV.This/
ET selection isalsoappliedtoee eventswhen thedielectron invariantmass isbetween70 and
100 GeV,toreduce the Z
→
ee backgroundcontribution. We de-finea/
ET significancevariable,S
,which measures thelikelihood fortheobserved/
ET tobeafluctuation from/
ET=
0 GeV.We re-quireS >
3.5 (4) forthe ee (μμ
) channel.Werequireeμ
events tohave HT>
100 GeV,where HT isthescalarsumofthe pT of thetwohighest-pT jetsandoftheleptonwithhighestpT.TheHT,b tagging,and
/
ET-basedrequirementsareoptimized tominimize theexpectedstatisticaluncertaintyonmt ineachchannel.The ex-pectedsignal-to-background(S/B)ratiois≈
7forthesechannels. Theserequirementsyielda3%improvementinstatisticalprecision inmt relativetotheselectionsinRef.[14].Afterimplementingall theseselections,weobtain340,115,and110eventsintheeμ
,eeand
μμ
channels,respectively.5. Modelingsignalandbackground
The t
¯
t events are simulated at 15 mass points over therange 130
≤
mMCt≤
200 GeV using the tree level generator alp-gen 2.11 [31] with up to 2 additional light partons and pythia6.409 [32] with modified underlying event Tune A for parton
showeringandhadronization.Here,mMC
t refers totheinputmass in alpgen. An additional, larger sample is generated at mMCt
=
172.5 GeV to study systematic uncertainties. We normalize the
tt production
¯
crosssection toσ
tt¯=
7.24±
0.23 pb [33],which is calculatedatnext-to-next-to-leadingorderwitha next-to-next-to-leadinglogarithmsoftgluonresummation.Themainbackgrounds arise from three sources: Z/
γ
∗→
+−, diboson (W W , W Z ,
and Z Z )processes,andinstrumentaleffects.Wemodelthe Z
/
γ
∗ backgroundusing alpgen withupto2lightpartonsand pythia for showeringandhadronization. We employ pythia forthe diboson background. The instrumental background arises from W+
jets, multijet, or+
jets tt events¯
where one or two jets are either mis-identified aselectrons, orthey contain a hadron decayingto a non-isolated lepton that passesour selection. Thisbackground is estimated from data as in Ref. [29]. We apply a full detector simulationbasedon geant 3.14[34] forall simulatedevents.The objects reconstructed in simulation are smeared to ensure that their resolutionsreflect thoseindata.Scale factorsin object effi-cienciesareappliedtoimproveagreementbetweendataandMC.6. Kinematicreconstruction
6.1. Neutrinoweighting
Thepresenceoftwoneutrinosinthet
¯
t decaymakesit impossi-bletofullyconstrainthekinematicsandthusextracta uniquemtmeasurement from each event. Given the measured momenta of
leptons, jetsand
/
ET, theavailable constraintsfrom MW, andthe condition mt=
mt¯,we are missingone constraintto provide fullt
¯
t reconstructionin dileptonevents.We integrate overthe phase spaceofneutrinorapidities forchosen valuesofhypothesizedmt (mht)[35],andcompare
/
EcalcT ,thevectorsumofneutrino momen-tumsolutionsateachchosenpointofphasespace,totheobserved/
EobsT todetermine a“weight”ω
characterizing thelevelof agree-ment:ω
=
1 N N i=1 j=x,y exp−
(
/
E calc j,i−
/
Eobsj)
2 2σ
2 / ET,
(1)where i runsoverallneutrinosolutions foranytwo possible jet-leptonassignmentsinthet
¯
t finalstate(upto N=
8), j standsfor thetwo orthogonalcoordinatesinthetransverseplane(x and y), andσ
/ET isaparameterrepresentingtheRMSofthedifferencebe-tweenthetransversecomponentsofthemeasured
/
ET andthesum ofthesolvedneutrinotransversemomenta.Theparameterσ
/ET istakentobethesameinbothx and y directions. Weperformthis calculation over a range of mht, integrating
ω
over the neutrino phasespace,toyieldadistributionofω
(
mt)
versusmht.Prior stud-ies [36] haveshownthat thefirst twomoments (μ
ω,
σ
ω )ofthisdistributionextractmostoftheinformationaboutmt.Theanalysis ofRef.[14]usedtherangeofmh
t valuesbetween80and330 GeV in1 GeVstepsanda
σ
/ET of7 GeVintheweightcalculation.Thenewoptimizeddetermination oftheseparametersisbriefly sum-marizedbelow.
6.2. Optimizationofweightcalculationparameters
After applying the methods described above to improve the
jet energycalibration,thestatisticalcontribution isthedominant source ofmeasurement uncertainty on mt in the dilepton chan-nel.We thereforeexamine theparametersusedforthekinematic reconstruction oftt events
¯
andforthemaximumlikelihoodfitto reducetheexpectedstatisticaluncertainty.Ateachstep,weverify through MC simulations that the optimization does not increase thesystematicuncertainty.All neutrinosolutions and jetassignments yieldmass estima-torssuchas
μ
ω thatarecorrelatedwithmt.However,the correla-tion issubstantiallygreater, andμ
ω valuesarelessbiased, whenthe correct jet assignments and solutions of neutrino momenta are chosen. Since now mt has been measured with high preci-sion[18],wecanoptimizetherangeofmht basedonknownvalues of mt. Consideringa wide range inmth causes incorrect configu-rationstooverwhelmthecorrectconfiguration,therebyworsening the mass resolution. Likewise, scanning over too narrow a range biasesthe backgroundandworsensthemass sensitivityby caus-ing t
¯
t andbackgrounddistributions tobe similar. Examinationof a two-dimensional grid of upper and lower limits of the mass rangeyields theoptimalrangeofmht=
115 to220 GeVin1 GeV steps. The value ofσ
/ET also has a noticeable impact on theex-pectedprecisionoftheanalysis.ThiswasnotthecaseinRef.[14],
mainly because the final top quark mass measurement was less
precise. InRef. [14],the value of7 GeV for
σ
/ET was obtainedastheunclustered
/
ET resolutioninanearlierdataset[36],wherethe unclustered/
ET is themagnitudeof thevector sumofall energy depositions inthe calorimeterthat are not included inlepton or jet reconstruction. However, accounting only for the unclustered energyresolutionastheoriginofthedifference betweenthe cal-culatedandmeasured/
ET ignorestheeffectofassumptionsthatgo into thekinematicreconstruction. Forinstance,thefinitebinning oftheneutrinorapiditiesdiscretizesthesolvedneutrinomomenta andthereforethesolved/
ET.Also,thesolved/
ET doesnotincludeFig. 1. Thedistributioninthemassestimator,μw,forthecombinationoftheee, eμ, andμμchannelsfor(a)thepreselectedsampleand(b)thefinaleventsample.The MCeventsarenormalizedseparatelytothenumberofobservedeventsindatain eachchannel.Theratiosshowthetotalnumberofobservedeventsdividedbythe numberofexpectedeventsinagivenbinofμw formtMC=172.5 GeV.Theband ofsystematicuncertaintyisshownas theshaded areainthe ratioplots,which includescontributionsfromthedominantsources:jetenergyscale,lepton identifi-cation,leptonmomentumscale,luminosity,b quarkmodeling,initialandfinalstate radiation,colorreconnection,aswellashadronizationandhigher-orderQCDeffects fortt events.¯
additional jets, reconstructed or not, since only the two leading jetsareconsidered in thekinematic reconstruction.Due to these additionalcontributions,ascaninawiderangefrom7to100 GeV isperformedandwefindtheoptimalvaluefor
σ
/ET tobe25 GeV,whichislarger thanthe 7 GeVofRef. [14]. Combined,these op-timizationsimprovetheexpectedcombinedstatisticaluncertainty onmt by11%comparedtotheparametersinRef.[14].
6.3.Efficiencyofkinematicreconstructionandeventyields
Eventsusedintheanalysismusthaveatleastonepairof neu-trino solutions for at least one mht value. The efficiency for this kinematic reconstruction is over 99% for tt events,
¯
and 91% to 98% forthe background.In the final sample, a total of336, 113, and109eventsintheeμ
,ee,andμμ
channels,respectively,pass the kinematic reconstruction. The expected sum of tt and¯
back-groundyieldsandtheircorrespondingasymmetrictotal uncertain-ties(stat⊕
syst)are298.1−+2227..12,106.5+−1110..46,and103.5+−79..41 events fortheeμ
,ee,andμμ
channels,respectively.Thedistributionsof themassestimatorμ
ω ina preselectedsample,omittingrequire-ments on b tagging,
/
ET,/
ET significance, and HT, are shown inFig. 1(a).Thet
¯
t componentisevidentinthepreselecteddata.The massdependence oftheμ
ω distribution is giveninFig. 1(b) forthreemMC
t masspointswithallselectionsapplied.
7. Extractingthetopquarkmass
7.1. Maximumlikelihood
We perform a binned maximum likelihood fit to the
ex-tracted moment distributions
[
μ
ω,
σ
ω]
in data. Expectedproba-bility densities are calculated using the MC samples for each of the 16 mt points, yielding a two-dimensional probability den-sityhS
(
μ
ω,
σ
ω|
mMCt)
distributionparametrizedbymt.Background samples are used to construct a background template for each channel, hB(
μ
ω,
σ
ω),
with each background contributingaccord-ing toits expectedyield.Bins insignal templateswithno events are given a weighted value corresponding to a single signal MC eventtoensurethatthelogoflikelihoodisnotinfinite.The likeli-hoodisgivenby:
L
(
μ
ω{1..N},
σ
ω{1..N},
N|
nS,
nB,
mt)
=
N i=1 nS·
hS(
μ
ωi,
σ
ωi|
mt)+
nB·
hB(
μ
ωi,
σ
ωi) nS+
nB,
(2)where N isthenumber ofobservedeventsindata,nS is the
ex-pected number oft
¯
t events (for mt=
172.5 GeV), andnB is theexpectedtotalnumberofbackgroundevents.Wefit(
−
lnL
)versusmMCt to a parabola in a window of mMCt that is iteratively
var-ied until a stable minimum is found. We take the minimum of
thefinalparabolatobethefittedtopquarkmass,mfit
t .The uncer-taintyonthefittedmassisobtainedbyconsideringthemMCt range overwhichthefitfunctionincreasesby0.5unitsin(
−
lnL
)above this minimum.Using pseudo-experiments, we optimizethe tem-platebinningofeachchannelseparatelyinatwo-dimensionalgrid thatletsμ
ω andσ
ω binsizesvaryindependently.Finerbinninginμ
ω andσ
ω ,especiallyfortheeμ
channel,improvestheexpectedstatisticalprecision inmfitt by 5%. The fittedmass windowis op-timizedto
±
15 GeV forall channels.Taking alltheoptimizations together, including eventselection, weight calculation, and max-imum likelihood fitting, the statistical sensitivity of this analysis isimproved relativeto Ref.[14] by 20%beyondthe 35%gain ex-pectedfromincreasedintegratedluminosity.7.2. Ensembletestinganddataresults
Weobtain alinearrelationshipbetweenmfit
t andmtMC by
per-forming randomized pseudo-experiments using all signal mass
points. The numbers of signal and background events in the
pseudo-experimentsare allowed tofluctuatewithin their Poisson uncertaintiesaroundtheirexpectedvalues.Werequirethatthe to-talnumberofeventsmatchesthatobserved indata.Tominimize the effect of statistical fluctuations on our systematic uncertain-ties, we optimizethe numberof pseudo-experimentsby dividing theMC sampleintofivesubsamples,andmeasuresystematic un-certaintieswitheachsubsample.WecalculatetheRMSofthefive uncertainties,averageoverallsystematiceffects,anddivideby
√
5 to estimate the statisticalcomponentof systematicuncertainties. The averageRMS decreases untilwe oversample, orreuse, thet¯
tMC eventsbyroughly afactorofthree.Thiscorresponds to3000 pseudo-experiments.Weperformalinearfitofmfit
t versusmMCt to obtain a calibration slope andoffsetformfitt using3000 pseudo-experiments:
mfitt
=
Slope· (
mMCt−
170)+
Offset+
170. (3) We account for oversamplingby increasing the statistical uncer-taintiesateachmasspointbytheappropriateoversamplingfactor. Likewise, we compute the pull, or the ratio of mfitt−
mMCt over theaverage estimateduncertaintyateachmasspoint. TheslopesTable 1
Slopes,offsets,andpullwidthsofthemt calibrationandtheexpectedstatistical uncertaintiesinthemass(σmt)fortheee,eμ,andμμchannels,andtheir combi-nation.
Slope Offset [GeV] Pull width σmt [GeV]
ee 0.984±0.004 0.671±0.043 0.994 2.98 eμ 0.986±0.006 0.548±0.065 0.998 1.72
μμ 0.989±0.010 0.717±0.103 1.004 3.31
2 0.988±0.006 0.617±0.063 0.995 1.35
Table 2
Systematic uncertainties on mt for the combined dilepton measurement using 9.7 fb−1 ofintegratedluminosity. For symmetrizeduncertainties,the “±” sym-bolindicatesthatthe corresponding systematicparametersinMCarepositively correlatedwithmt indata,andthe “∓”symbolindicatesananticorrelation.The uncertaintiesshownas+or−onlyarecomputedbycomparingastandardchoice withanalternate,butaresymmetrizedincalculatingthetotaluncertainty.
Source σmt [GeV]
Jet energy calibration
Absolute scale ∓0.47 Flavor dependence ∓0.27 Residual scale +0.36 −0.35 b quark fragmentation +0.10 Object reconstruction Trigger −0.06 Electron pTresolution ±0.01 Muon pTresolution ∓0.03
Electron energy scale ±0.01
Muon pTscale ±0.01 Jet resolution ∓0.12 Jet identification +0.03 b tagging ∓0.19 Signal modeling Higher-order effects −0.33 ISR/FSR ±0.15 pT(t¯t) −0.07 Hadronization −0.11 Color reconnection −0.22 Multiple pp interactions¯ −0.06 PDF uncertainty ±0.08 Background modeling Signal fraction ±0.01
Heavy-flavor scale factor ±0.04
Method
Template statistics ±0.18
Calibration ±0.07
Total systematic uncertainty ±0.85
of mfitt versus mtMC are close to 1, and pull widths are consis-tent with unity, as shown in Table 1. We calculate the final mt by correcting mfitt from a given measurement by the slope and offset. We correctthe statistical uncertainty using the slope and thepullwidth.Theexpectedcorrectedstatisticaluncertainties for each channel are given in Table 1. In data, we obtain corrected, fittedmt values ofmt
=
171.86±
1.71(stat), 173.99±
3.04(stat), and 178.58±
3.56(stat) GeV for the eμ
, ee, andμμ
channels respectively, andmt=
173.32±
1.36(stat) GeV for the combined channels.8. Systematicuncertainties
Systematicuncertainties summarizedinTable 2arise fromjet energycalibration,objectreconstruction,modelingoft
¯
t and back-groundevents, and the mass-extractionmethod. The energies of jetsare shifted up anddown by theuncertainty onthe absolute energy scale, which is taken from+
jets events, thereby pro-viding shifts in mt. This scale is appropriate forlight-quark jets, which, after correcting for jet flavors to improve the agreementbetweendataandMC,havedifferentkinematicdistributions than
b jets fromt
¯
t decays. Wecalculate a residualuncertainty dueto the kinematicdifferencesbetweenthe+
jets calibrationsample anddileptonsampleofb jets.Weuseseparateupanddown esti-mates toextract theenergy- andη
-dependentshifts inmt based on uncertainties inthe standardjet energyscale relative totheir average value in the+
jets calibration sample. We cross-check thiswithanalternativemethodthatappliesshiftedlight-quarkjet energyscalestob jetsinthe+
jets channel[15].Thesemethods agree, andthereby validate the useof the+
jets scaleas a jet calibration.Wealsocross-checkusingajet-energy-dependent lin-earparameterizationoftheresidualjetenergyscaleasinRef.[15], obtaining resultsthatdonotexceedourestimate ofuncertainties fromthejetenergyscale.Toestimatetheuncertainty correspond-ing to possible differences in the flavor dependence of the MC scale relative todata,we changethe single-particle responsesup and down by their uncertainties and obtain the shift in mt. To estimate the possible dependence on the b quark fragmentation inthe MC,we replacethe pythia b quarkfragmentationfunction with the Bowler scheme [37], andcompare mt with the Bowler free parameters tuned toLEP (ALEPH, OPAL, andDELPHI) orSLD data[38].The systematicuncertainty dueto the triggerefficiency is es-timated by applying the ratio of single lepton trigger efficiency parameterization in data divided by the MC parameterization to
the ee and
μμ
channels. The uncertainties in the modeling ofthe energy and momentum resolutions ofelectrons, muons, and jetsareapplied independentlyofeachother,andtheshiftsinmt areextractedasuncertaintiesonmt.Leptonenergyormomentum scalesandtheiruncertaintiesareextractedfromZ
→
2eventsin data. An additional uncertainty is estimatedfor jet identification byshiftingthejetidentificationefficiencywithinitsuncertaintyin MC samples toestimate their effecton mt.The uncertaintyfrom modeling b tagging is evaluated by changing within their uncer-tainties the corrections that account for the agreement between dataandMCinb taggingefficiency.Higher-order virtual corrections to mt are absent in the alp-genusedtogenerateourstandardt
¯
t samples.Wetherefore com-pare an ensembleof pseudo-experimentsusing mc@nlo 3.4 [39] t¯
t eventswithone using alpgen events,wherebothemploy her-wig 6.510 [40] for modeling of hadronization. To evaluate the uncertaintyassociatedwiththemodelingofinitialandfinal-state radiation(ISR/FSR),wecompare alpgen+pythia withthe renormal-ization andfactorization scale changedup anddown by a factor of 1.5[15].The+
jets analysisexhibitsadiscrepancyintheshape ofthepT distributionofthet¯
t system,which,althoughthe dilep-ton statistics arelimited, maybe presentin thedileptonsample. We evaluate theuncertainty inthe modelingof thet¯
t pT distri-bution by reweighting MC events to makethem match thedata. The observed shift in mt is taken as the uncertainty. Since the hadronization inour standardtt sample¯
ismodeled with pythia, we estimate a hadronization uncertainty on mt by performing pseudo-experimentsusingan alpgen+herwig sample.Weevaluate the effect of color reconnection by comparing mt measurements in alpgen+pythia sampleswithtwo pythia tunes:thePerugia2011 tune thatincorporates anexplicitcolor-reconnectionscheme,and thePerugia2011NOCRtunethatdoesnot[41].Data andMCmay havedifferentdistributionsininstantaneousluminosityafterevent selection.Thisuncertaintyduetomultiple pp interactions¯
is esti-matedbyreweightingthedistributionofinstantaneousluminosity tomakeMCagreewiththedataforrespectivedata-takingepochs, and then take the shift in mt withrespect to the default value. The uncertaintyduetothe protonstructure isobtainedfromthe 20setsofCTEQ6L1partondistributionfunctions(PDF)reweightedtoCTEQ6M,wherethedeviationsinmtforthe20eigenvectorssets areaddedinquadrature[42].
Weestimatetheeffectoftheuncertaintyonthefractionof sig-nalor backgroundby changingthe expected tt event
¯
yields (nS)upanddownandtheexpectedbackgroundyields (nB)down and
up within their total uncertainties. The heavy-flavor scale factor, whichisappliedtothe Z
→
2crosssectiontocorrectthe heavy-flavorcontent,isalsochangedupanddownwithinitsuncertainty toestimateitssystematiceffectonmt.OurtemplatesareconstructedfromMCsamplesfort
¯
t,Z→
2, anddibosonbackgrounds,aswellasdatasamplesforinstrumental background,yieldingstatisticaluncertaintiesontheirbincontents. Weuse Poissondistributionsto modify bincontents withintheir statisticaluncertaintiestoobtain1000newtemplates.Wemeasuremt indatausingthesetemplates,andtheRMSofthemeasuredtop quarkmass istakenasits uncertainty.Ourmethodofmt extrac-tionreliesonthecorrectionofthefittedmt totheinputMCmass. Theuncertainties fromthiscalibrationare appliedto providethe uncertainty inmt. The uncertainty is reduced substantially from Ref.[14] dueprimarily tothe reduction inthe uncertaintyin jet energycalibrationandtheoptimizationsforimprovementsin sta-tisticaluncertainty.LargerMCsamplesalsocontributebylowering statisticalfluctuationsonsystematicuncertainties,orreducing sta-tisticallylimitedsystematicuncertainties.
9. Conclusions
Wehavemeasuredthetopquarkmassinthecombined dilep-tonchannels(e
μ
,ee,μμ
):mt
=
173.32±
1.36(stat)±
0.85(syst)GeV=
173.32±
1.60 GeV.This measurement is consistent with the current world average value of mt [18]. Our measurement is the mostprecise dilepton resultfromtheTevatron,andiscompetitive withthemostrecent LHCdilepton measurements. The systematicuncertainty of0.49% isthesmallestofalldileptonmeasurements.
Wethankthe staffsatFermilab andcollaborating institutions, andacknowledgesupportfromtheDepartmentofEnergyand Na-tionalScience Foundation (United Statesof America); French Al-ternative Energies and Atomic Energy Commission and National CenterforScientificResearch/NationalInstituteofNuclearand Par-ticle Physics (France); Ministry of Education and Science of the RussianFederation, NationalResearchCenter“KurchatovInstitute” of the Russian Federation, and Russian Foundation for Basic Re-search (Russia);National Council forthe Developmentof Science andTechnology and CarlosChagas Filho Foundation forResearch Support in the State of Rio de Janeiro (Brazil); Department of
Atomic Energy and Department of Science and Technology
(In-dia); Administrative Department of Science, Technology and In-novation (Colombia);National Council ofScience andTechnology (Mexico); National Research Foundation of Korea (Korea); Foun-dation for Fundamental Research on Matter (The Netherlands); Science and Technology Facilities Council and The Royal Society (UnitedKingdom);MinistryofEducation,YouthandSports(Czech Republic);Bundesministeriumfür BildungundForschung(Federal Ministry of Education and Research) and Deutsche Forschungs-gemeinschaft (German Research Foundation) (Germany); Science FoundationIreland(Ireland);SwedishResearchCouncil (Sweden); ChinaAcademy ofSciences andNationalNaturalScience Founda-tion of China (China); and Ministry of Education and Science of Ukraine(Ukraine).
References
[1]S.Abachi,etal.,D0Collaboration,Observationofthetopquark,Phys.Rev.Lett. 74(1995)2632.
[2]F.Abe,etal.,CDFCollaboration,Observationoftopquarkproductioninpp col-¯
lisionswiththeColliderDetectoratFermilab,Phys.Rev.Lett.74(1995)2626. [3] TheGfitterGroup,Resultsfortheglobalelectroweakstandardmodelfit,http://
project-gfitter.web.cern.ch/project-gfitter/Standard_Model/.
[4]G.Aad,etal.,ATLASCollaboration,Observationofanewparticleinthesearch forthestandardmodelHiggsboson,Phys.Lett.B716(2012)1.
[5]S.Chatrchyan,etal.,CMSCollaboration,Observationofanewbosonatamass of125GeVwiththeCMSexperimentattheLHC,Phys.Lett.B716(2012)30. [6]G.Degrassi,etal.,Higgsmassandvacuumstabilityinthestandardmodelat
NNLO,J.HighEnergyPhys.08(2012)098.
[7]A.DeSimone,etal.,Runninginflationinthestandardmodel,Phys.Lett.B678 (2009)1.
[8]F.Bezrukov,etal.,ThestandardmodelHiggsbosonastheinflaton,Phys.Lett. B 659(2008)703.
[9]V.M.Abazov,etal.,D0Collaboration,DeterminationofthepoleandM S masses
ofthetopquarkfromthet¯t crosssection,Phys.Lett.B703(2011)422. [10]T.Aaltonen,etal.,CDFCollaboration,Cross-section-constrainedtop-quarkmass
measurementfromdileptoneventsatthetevatron,Phys.Rev.Lett.100(2008) 062005.
[11]S.Chatrchyan,etal.,CMSCollaboration,Determinationofthetop-quarkpole massandstrongcouplingconstantfromthet¯t productioncrosssectioninpp
collisionsat√s=7 TeV,Phys.Lett.B728(2014)496.
[12]G.Aad,etal.,ATLASCollaboration,Measurementofthet¯t production cross-sectionusingeμeventswithb-taggedjetsinpp collisionsat√s=7 and8 TeV withtheATLASdetector,Eur.Phys.J.C74(2014)3109.
[13]F.Jegerlehner,M.Kalmykov,B.Kniehl,Onthedifferencebetweenthepoleand themassesofthetopquarkattheelectroweakscale,Phys.Lett.B722(2013) 123.
[14]V.M.Abazov,etal.,D0Collaboration,Measurementofthetopquarkmassin
p¯p collisionsusingeventswithtwoleptons,Phys.Rev.D86(2012)051103(R). [15]V.M.Abazov,etal.,D0Collaboration,Precisionmeasurementofthetopquark
massinlepton+jetsfinalstates,Phys.Rev.Lett.113(2014)032002; V.M.Abazov,etal.,D0Collaboration,Precisionmeasurementofthetop-quark massinlepton+jetsfinalstates,Phys.Rev.D91(2015)112003.
[16]V.M.Abazov,etal., D0Collaboration,Precisemeasurementofthetopquark massinthedileptonchannelatD0,Phys.Rev.Lett.107(2011)082004. [17]T.Aaltonen,etal.,CDFCollaboration,Topquarkmassmeasurementusingthe
templatemethodatCDF,Phys.Rev.D83(2011)111101.
[18]ATLAS,CDF,CMS,andD0Collaborations,FirstcombinationoftevatronandLHC measurementsofthetop-quarkmass,arXiv:1403.4427.
[19]T.Aaltonen,etal.,CDFCollaboration,Measurementofthetop-quarkmassin thet¯t dileptonchannelusingthefullCDFrunIIdataset,arXiv:1505.00500. [20]G.Aad,etal.,ATLASCollaboration,Measurementofthetopquarkmassinthe
t¯t→lepton+jets andtt¯→dilepton channelsusing√s=7 TeV ATLASdata, arXiv:1503.05427.
[21]S.Chatrchyan,etal.,CMSCollaboration,Measurementofthetop-quarkmassin
t¯t eventswithdileptonfinalstatesinppcollisionsat√s=7 TeV,Eur.Phys. J. C 72(2012)2202.
[22]V.M.Abazov,etal.,D0Collaboration,TheupgradedD0detector,Nucl.Instrum. MethodsA565(2006)463.
[23]R. Angstadt,etal.,The layer0innersilicondetector oftheD0 experiment, Nucl.Instrum.MethodsA622(2010)298.
[24] Thepseudorapidityisdefinedasη= −ln[tan(θ/2)]whereθisthepolarangle relativetotheprotonbeamdirection
[25]V.M.Abazov,etal.,D0Collaboration,Electronandphotonidentificationinthe D0experiment,Nucl.Instrum.MethodsA750(2014)78.
[26]V.M.Abazov,etal.,D0Collaboration,Muonreconstructionandidentification withtherunIID0detector,Nucl.Instrum.MethodsA737(2014)281. [27]G.Blazey,etal.,RunIIjetphysics:proceedingsoftherunIIQCDandweak
bosonphysicsworkshop,arXiv:hep-ex/0005012,2000.
[28]V.M.Abazov,etal.,D0Collaboration,Improvedb quarkjetidentificationatthe D0experiment,Nucl.Instrum.MethodsA763(2014)290.
[29]V.M.Abazov,etal.,D0Collaboration,Measurementoftheasymmetryin angu-lardistributionsofleptonsproducedindileptont¯t finalstatesinpp collisions¯
at√s=1.96 TeV,Phys.Rev.D88(2013)112002.
[30]V.M.Abazov,etal.,D0Collaboration,JetenergyscaledeterminationintheD0 experiment,Nucl.Instrum.MethodsA763(2014)442.
[31]M. Mangano, et al., alpgen, a generator for hard multipartonprocesses in hadroniccollisions,J.HighEnergyPhys.07(2003)001;
M.Mangano,M.Moretti,R.Pittau,Multijetmatrixelementsandshower evo-lutioninhadroniccollisions:W bb¯+n jetsasacasestudy,Nucl.Phys.B632 (2002)343;
F.Caravaglios,etal.,Anewapproachtomultijetcalculationsinhadron colli-sions,Nucl.Phys.B539(1999)215.
[32]T.Sjöstrand,etal.,High-energy-physicseventgenerationwith pythia 6.1, Com-put.Phys.Commun.135(2001)238.
[33]M.Czakon,A.Mitov,Top++:aprogramforthecalculationofthetop-pair cross-sectionathadroncolliders,Comput.Phys.Commun.185(2014)2930; M.Cacciari,etal.,Top-pairproductionathadroncolliderswith next-to-next-to-leadinglogarithmicsoft-gluonresummation,Phys.Lett.B710(2012)612; P.Baernreuther, M. Czakon,A.Mitov, Percent-level-precisionphysics at the tevatron:next-to-next-to-leadingorderQCDcorrectionstoqq¯→t¯t+X ,Phys. Rev.Lett.109(2012)132001.
[34] R.Brun,F.Carminati,CERNprogramlibrarylongwriteupW5013,1993 (un-published).
[35]S.Abachi,etal.,D0Collaboration,Measurementofthetopquarkmassusing dileptonevents,Phys.Rev.Lett.80(1998)2063.
[36]V.M.Abazov,etal.,D0Collaboration,Measurementofthetopquarkmassin finalstateswithtwoleptons,Phys.Rev.D80(2009)092006.
[37]M.G.Bowler,e+e−productionofheavyquarksinthestringmodel,Z.Phys.C 11(1981)169.
[38] Y.Peters,K.Hamacher,D.Wicke,PreciseTuningoftheb Fragmentationforthe D0MonteCarlo,FERMILAB-TM-2425-E,2006.
[39]S.Frixione,etal.,Single-tophadroproductioninassociationwithaW boson,
J. HighEnergyPhys.07(2008)029;
S.Frixione,etal.,Single-topproductionin mc@nlo,J.HighEnergyPhys.03 (2006)092;
S.Frixione,B.Webber,MatchingNLOQCDcomputationsandpartonshower simulations,J.HighEnergyPhys.06(2002)029.
[40]G.Corcella,etal.,HERWIG6:aneventgeneratorforhadronemissionreactions withinterferinggluons(includingsupersymmetricprocesses),J.HighEnergy Phys.01(2001)010.
[41]P.Skands,TuningMonteCarlogenerators:thePerugiatunes,Phys.Rev.D82 (2010)074018;
P.Skands,TuningMonteCarlogenerators:thePerugiatunes,arXiv:1005.3457, 2011.
[42]J.Pumplin,et al.,Newgenerationofpartondistributionswithuncertainties fromglobalQCDanalysis,J.HighEnergyPhys.07(2002)012.
