Measurements of the Production, Decay and Properties of the Top Quark: A Review

0.47 MB·English

by  Kevin Lannon

EPJ manuscript No.
(will be inserted by the editor)
Measurements of the Production, Decay and Properties of the
Top Quark: A Review
Kevin Lannon1, Fabrizio Margaroli2, and Chris Neu3
1 University of Notre Dame, NotreDame, IN 46556, USA
2 Sapienza Universit`a diRoma and INFN Roma1, 00185 Rome, Italy
3 University of Virginia, Charlottesville, VA 22904, USA
2
1
0 Received: date/ Revised version: date
2
Abstract. With thefull Tevatron RunII and early LHCdatasamples, theopportunityfor furtheringour
n
a understanding of the properties of the top quark has never been more promising. Although the current
J knowledge of the top quark comes largely from Tevatron measurements, the experiments at the LHC
7 are poised to probe top-quark production and decay in unprecedented regimes. Although no current top
2 quark measurements conclusively contradict predictions from the standard model, the precision of most
measurements remains statistically limited. Additionally, some measurements, most notably AFB in top

] quark pair production, show tantalizing hints of beyond-the-Standard-Model dynamics. The top quark
x
sample is growing rapidly at the LHC, with initial results now public. This review examines the current
e
status of top quark measurements in the particular light of searching for evidence of new physics, either
-
p throughdirectsearchesforbeyondthestandardmodelphenomenaorindirectlyviaprecisemeasurements
e of standard model top quark properties.
h
[ PACS. 14.65.Ha Top quarks

1
v
3 1 Introduction because |Vtb| ∼ 1 in the Standard Model, the top quark
7 decays to a W boson and a b almost 100% of the time,
8 The observationof the top quark at the Tevatronin 1995 producing a signature quite distinct from the collimated
5 marked the end of the search for the isospin partner to hadronicjets thatsignalthe productionof lighterquarks.
.
1 the bottom quark,completing the three-generationstruc- Finally, as a result of its large mass, a significant fraction
0 ture of the quark sector in the Standard Model. Yet, it (roughly70%intheStandardModel)ofthetheW bosons
2 also marked the beginning of the quest to understand producedintopquarkdecaysarelongitudinallypolarized.
1
why the top quark is so different from the other quarks.
:
v The very feature that allowed the top quark to evade ex-
i perimental detection for so long—its extraordinarilyhigh Beyond the phenomenological implications, the large
X
mass—is also the cause of most of its unusual proper- top quark mass raises deeper questions. In the Standard
r ties. The 173.2 GeV/c2 mass of the top quark[1] makes Model, quark masses arise from the quarks’ couplings to
a
it roughly forty times more massive than the next most theHiggsboson.Thelargetopquarkmassimpliesthetop
massive quark, the bottom quark, and over 10000 times quarkhasaparticularlylargecouplingtotheHiggsboson.
moremassivethantheupquark.Perhapsevenmorestrik- In fact, the specific value of the top quark mass implies a
ingisthat,asafundamentalobjectbelievedtopossessno Yukawa coupling to the Higgs very near unity. From this
internalstructure,thetopquarkismoremassivethanthe perspective it might be more appropriate to ask why the
first seventy-four elements in the periodic table. rest of the quarks have such an unnaturally small cou-
The top quark’s large mass has phenomenological im- pling to the Higgs rather than asking why the top quark
plicationsaswell.Foremost,becausethetopquarkismore mass is so large. Regardless of whether the Higgs mecha-
massive than the W boson, top quark decays proceed nism provides the correctexplanation for the electroweak
rapidly throughte electroweakinteraction via an on-shell symmetry breaking, the large top quark mass raises the
W before the top quark has a chance to form a hadronic possibilitythatthe topquarkmayhavesomespecialcon-
bound state. This makes the top quark the only quark nection to or play some special role in the mechanism of
whose properties can be studied without the complica- electroweaksymmetry breaking.Severalalternative mod-
tions of disentangling hadronization effects. Furthermore, els to the Higgs picture of electroweak symmetry break-
ing,suchas the topquark see-sawtheory [2]or top-color-
Send offprint requests to: K. Lannon assistedtechnicolor[3]positsucharoleforthetopquark.
2 Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review
Therefore, the primary focus of top quark physics is significantly extended the energy frontier,reachinga cen-
the search for some evidence of physics beyond the Stan- ter of mass energy of 7 TeV. The LHC is designed to col-
dardModel–inparticularnewphysicsthatwouldhelpto lideprotonswithprotonsina26.7kmcircumferencering.
explainthe topquark’ssingulardifferences incomparison Using the roughly 35 pb 1 of integrated luminosity de-
−
with the rest of the quark sector. Strategies in the search liveredat 7 TeV in 2010,both ATLAS [13] and CMS [14]
fornewphysicsassociatedwiththetopquarkfallbroadly experimentsmeasuredthetopquarkpairproductioncross
into two categories:First there are the direct searches for section.By2011,electroweaksingletopquarkproduction
new physics associated with top quark production or de- had been observed, and the full complement of top quark
cay.Examplesincludesearchingfornewheavyresonances measurements and searches for new physics were well un-
decaying into top quarks or searching for new particles derway. As of the end of pp running in 2011, the LHC
producedintopquarkdecays.Analternativestrategyisto hasdeliveredover5.6fb 1 ofintegratedluminositytothe
−
measure properties of the top quark predicted within the ATLAS and CMS experiments, reaching a maximum in-
StandardModel,suchasitsproductioncrosssection,both stantaneous luminosity of 3.6×1033 cm 2s 1. Over the
− −
inclusivelyanddifferentially,aswellasitsdecaybranching coming years the LHC is expected to gradually increase
fractions,lookingfordeviationscomparedtotheStandard the instantaneous luminosity to 1034 cm 2s 1 or beyond
− −
Modelpredictions.This reviewconsidersanalysesofboth andthecenterofmassenergyuptoamaximumof14TeV.
typesandsummarizesthecurrentexperimentalpictureof Asidefromtheobviousdifferencesrelatedtothehigher
top quark physics. centerofmassenergyandluminosities,thereareafewim-
portant distinctions between the collider environments at
theTevatronandtheLHC.Perhapsmostsignificantly,be-
cause of the higher beam energy, and the lack of valence
1.1 Accelerators and the Top Quark
antiquarks in the LHC beams, gluon initiated top pro-
duction plays a dominant role at the LHC, and processes
Itisimpossibletorelatethetaleofthe topquarkwithout involving antiquarks, such as pair production through qq¯
mentioning the two acceleratorsthat to date are the only annhilation or s-channel single top production, have a
locationswheretopquarkshavebeen producedina labo- smaller cross section. In contrast, at the Tevatron, qq¯an-
ratorysetting.Thefirstacceleratortoproducetopquarks nihilationdominates toppair production.This is particu-
was the Tevatron located at Fermi National Accelerator larlyrelevantfor any new physics models where contribu-
Laboratory in Batavia, Illinois in the United States. The tions to the top quark sample proceed through qq¯ anni-
Tevatron collides protons with antiprotons in a 6.28 km hilation.Furthermore,the symmetric nature of the initial
circumferencering.Whenitbegancollidingbeamsopera- state at the LHC means that there is no natural way of
tions in 1985,the center ofmass energywas 1.8TeV, and defining a “forward” or “backwards” direction, increas-
before the end of the first collider run, referredto as Run ing the challenge of pursuing such measurements as the
I, it had achieved instantaneous luminosities in excess of forward-backward asymmetry A in top pair produc-
FB
1.1× 1031 cm−2s−1. Nearly ten years after the start of tion. Finally, as the LHC luminosities continue to push
Tevatron operations, the CDF[4,5,6,7] and D0 [8] exper- to ever increasing levels, the LHC is experiencing never
iments jointly announced first observation of top quark before encountered levels of so-called pile-up from addi-
pair production using 67 pb−1 and 50 pb−1 of integrated tional pp collisions overlappingwith the collision of inter-
luminosityrespectively[9,10].Shortlyafterthe topquark est.These differencesinenvironmentcanleadto different
observation,the Tevatronshutdownforupgradesbothto measurement strategies being produced at the LHC com-
the maximum beam energy and luminosity, and in 2001, pared to similar analyses performed at the Tevatron.
the second Tevatron run, known as Run II, began, with
a center of mass energy of 1.96 TeV. Roughly eight years
later,withadatasetcorrespondingto3.2fb−1atCDFand 1.2 Dominant production mechanism for top quarks
2.1 fb 1 at D0, both experiments announced observation
−
ofproductionofsingletopquarksthroughtheelectroweak The top quark has color charge, so it can be produced
interaction [11,12]. The long time delay and large differ- throughstronginteractions.Whatmakes its crosssection
enceindatasetsizeneededfortheobservationofsingletop solowwithrespecttogenericcolliderinelasticinteractions
quark production compared to top quark pair production is its very large mass and the fact that it is produced in
is a testament to the experimental difficulty of extracting pairsviathestronginteraction,thusprobinghighxregion
thesmallsingletopsignalfromthelargebackgrounds.On of PDFs. Top quark pair production at hadron colliders
Sep. 30,2011,after 26 yearsof colliding beams, the Teva- happens through quark-antiquark annihilation and gluon
tronacceleratorceasedoperations.Overthattimeperiod, gluonfusion.Theformerisdominantatthe Tevatroncol-
the Tevatrondeliveredanintegratedluminosity ofalmost liderwherethevalenceu(u¯)quarkcontributiondominates
12 fb−1 and achieved instantaneous luminosities as high in the proton(antiproton). There are no u¯ valence quarks
as4.1×1032 cm−2s−1.Awiderangeoftopquarkanalysis in the LHC colliding protons, so the corresponding par-
continues to be pursued using this dataset both at CDF ton density functions (PDF) are very small. On the other
and D0. handonlyarelativelysmallfractionoftheproton’senergy
In March of 2010, the Large Hadron Collider (LHC) is needed to produce top quarks at the 7TeV LHC colli-
located at the CERN laboratory in Geneva, Switzerland, sions;thisenergyrangeiswherethegluonPDFdominate.
Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review 3
q t g t g t
Top Pair Decay Channels
q t g t g t s s
c et s
Fpriogd.u1c.tiTorne.e-levelLOFeynmandiagramsthatcontributetott¯ on+j n+jet +jets all-hadronic
ctr uo au
d e m t
u el
–τ eτ µτnsττ tau+jets
o
–µ eµ eµptµ µτ muon+jets
dil
– ee eµ eτ electron+jets
e
W cay e+ µ+ τ+ ud cs
e
Fig. 2. Feynman diagrams for single top quark production. d
Represented are (a) a LO s channel diagram, (b) a NLO t-
−
channeldiagram, and (c) a NLO Wt production diagram.
Fig. 3. Final states of thett¯system.
The fraction of top quarks produced through quark-anti- theexistingcolliderconstraintsonSUSYsuggestthatthe
quark annihilation and gluon-gluon fusion is respectively supersymmetric partners of the third generation quarks
85%(15%) at the Tevatron and 15%(85%) at the LHC. could be the iightest SUSY squarks[37]. The production
The main Feynman diagrams for top quark productions anddecayofstopquarkswouldappearkinematicallysim-
are shown in Fig. 1. The total cross section is approxi- ilar to SM top quark production.
mately7.5pbatthe1.96TeVpp¯collisionsattheTevatron,
and 160pb at the 7TeV pp collisions of the LHC [15,16,
17,18,19,20,21,22,23,24]. For a review summarizing the 1.3 Top quark decay modes
currentstatus of top productioncalculations at NLO and
with approximate higher-order corrections, see [25]. Due to its lifetime being shorter than the hadronization
Topquarkscanalsobe producedsinglyathadroncol- time, the top quark is different from other quarks in that
liders. This production happens through electroweak dia- the SM predicts it does not produce resonances. Using
gramsinthes-ort-channel,orthroughassociatedproduc- precisionmeasurementsofCKMparametersandthe con-
tion with a W boson. The cross sections at the Tevatron straintofitsunitarity,the topquarkispredictedtodecay
are respectively approximately 1 pb, 2 pb and 0.3 pb [26, 99.8% of the time into a W boson and a b quark. The
27,28,29,30,31,32]. The rise in the cross section at the W boson decays 67.6% of the time in ud¯or cs¯[38] (the
LHC is again a function of the number of gluons in the conjugate decays implied for the oppositely charged W
initial state: the s-channel production is approximately boson) and the remaining times into a charged lepton ℓ
5 pb, the t-channel is approximately 64 pb, and the Wt and the corresponding neutrino ν in the isodoublet. The
ℓ
productioncrosssectionsreachesapproximately16pb[31, single top quark decay modes are thus completely speci-
33,34]. fied.Pairproductionoftopquarksleavesamorecomplex
Thetopquarkistheleastunderstoodduetothemuch picture: depending on both W boson decays, one is left
smaller datasets available when compared to the other with a many-quarks final state (“all-hadronic’), a final
quarks. In principle, any sizable deviation in the produc- state composed of four quarks, a charged and a neutral
tion rate or in the final state kinematics of top quark lepton (“lepton+jets”) or a final state with two jets, two
events, would be a sign of new physics. There are sev- chargedleptonsandtwoneutralleptons(“dilepton”).The
eral theoretically well-motivated models that would pre- relativefractionsare46.2%,43.5%and10.3%;aschematic
dict new physics affecting top quark samples. A fourth representationis shown in Fig.3.
generation of heavy quarks is allowed by the SM fit to
theexistingprecisionmeasurementofelectroweakobserv-
ables, and would allow for the right size of CP violation 2 Analysis Techniques
in the universe[35]. These exotic quarks would appear in
detectorsverysimilarlyto eventswith SMtopquarkpro- 2.1 Identifying Top Events
duction.Heavyresonancesofreplicasofthe knownvector
bosons Z and W, that appear for example in dynami- Top quarks are distinguished from the backgrounds in
cal electroweak symmetry-breaking schemes[3,36] would large part because of the distinctiveness of their event
affect either or both pair and single top quark produc- signature,involvingsomecombinationofhighenergylep-
tion.Supersymmetry theory(SUSY) incombinationwith tons,jets (includingjetsoriginatingfrombquarks),anda
4 Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review
significantamountofmissingtransverseenergyE .There- and p > 20 GeV/c to 33 GeV/c for muons. For CMS,
T T
fore, the first step of any top quark analysis is i(cid:14)dentifying the muon acceptance ranges from |η| < 2.1 to 2.5, while
the appropriate combination of these signatures. electrons are identified in the range |η|<2.5.
2.1.1 Leptons 2.1.2 Jets
In most cases, the term lepton for a top quark analysis
Depending on the decay topology, top quark events will
refers specifically to electrons and muons. Unless explic-
have between two and six jets, with up to two of these
itly noted, tau leptons are not included, although elec-
originatingfrombquarks.Identifying energeticjets iskey
trons and muons produced in leptonic tau decays can
to separating the top signal from the backgrounds.
contribute. Because they are relatively easy to identify,
CDFandD0use similarjetreconstructionandidenti-
compared to jets andE , leptons are used in the trigger
T fication algorithms [39,40]. Both experiments reconstruct
formanytopquarkan(cid:14)alyses.Thekinematicrequirements
jets with a cone-based algorithm that clusters calorime-
necessary to produce reasonable triggerrates often deter-
ter energy deposits into jets. Jet energies are corrected
mine the lepton acceptance for these analyses.
to account for gain variations across the detector, non-
Dilepton and lepton plus jets analyses at CDF uti-
linear energy response, and activity from pile-up. The jet
lize data collected with a trigger involving a single elec-
energies are then calibrated to reproduce the energy of
tron or muon with transversemomentum p >18 GeV/c
T the final state hadrons contained in the jet cone. Addi-
within the psuedorapidity1 range |η| < 2.0 for electrons
tional corrections to parton-level (in some cases specific
and|η|<1.1 formuons.These leptons are then identified
to the top quark events) are employed where appropriate
offline, requiring p >20 GeV, as well as such additional
T (for example, in the top mass measurement). Both CDF
requirements as having a large ratio of electromagnetic
and D0 require jets to have E > 20 GeV. D0 jets are
to hadronic energy and consistent shower profile for elec- T
reconstructedwitha cone sizeof0.5 andrequiredto have
trons, or a high quality of the track-muon matching for
|η|<2.5.At CDF, jets are reconstructedwith a cone size
muons.Additional,non-triggeredleptons,suchasthesec-
of 0.4 and |η|<2.0.
ond lepton in a dilepton event, or leptons collected on
BothATLAS and CMSreconstructjets using ananti-
jet- andE -based triggers are allowed to pass looser re-
T
k algorithm [41]. ATLAS reconstructs jets by clustering
quiremen(cid:14)ts.Forwardmuons(1.1<|η|<1.6)andisolated T
togethertopologicalenergyclusterswithinthecalorimeter
tracks are sometimes used to increase the muon accep-
above certain energy thresholds using R = 0.4. Energies
tance. To suppress leptons from heavy quark decays or
are corrected from the EM scale to the hadronic scale
hadrons faking leptons, leptons are often required to be
usingp -andη-dependentcorrectionfactors,andthenthe
isolated,meaningthattheratiobetweentheE contained T
T
absolutejetenergyscaleiscalibratedusingtestbeamdata
in a cone ∆R< 0.4 around the lepton and the lepton p
T
and Monte Carlo simulated collisions. Jets with p > 25
is less than 0.1. T
GeV and |η|<2.5 are considered for top quark analyses.
The D0 experiment uses a similar approach: The D0
CMS incorporates tracking information in its jet re-
single lepton trigger thresholds range from 15 GeV to 80
construction, using one of two algorithms: The Jet Plus
GeVforelectronsandfrom10GeVto 15GeVformuons.
Track(JPT)algorithmimprovesjetenergymeasurements
D0 supplements these single object triggers with a collec-
by combining tracking information into jets first recon-
tion of triggers requiring a single lepton with varying p
T
structed in the calorimeter alone. Corrections are made
thresholdsplusajet.Offlineleptonsareselectedbyrequir-
based on the measurements of track momentum both for
ing p > 20 GeV. Muons are reconstructed in the range
T
tracksthat project into the area of the jet cone as well as
|η| < 2.0 while electrons must have |η| < 1.1. Isolated
those that projectoutside. The second approachtaken at
tracks are also used to extend the muon acceptance.
CMS is particle-flow jet reconstruction. The particle flow
ATLAS and CMS are able to trigger on electrons and
algorithm attempts to use the full CMS detector to asso-
muons over a significantly wider η range than the Teva-
ciate all measured energies correctly to one of the follow-
tron experiments. The ATLAS trigger accepts electrons
ing categories: electrons, muons, taus, photons, charged
withE >22GeVand|η|<2.47,whilethe muontrigger
T
hadrons, and neutral hadrons. Once the various compo-
allows muons with p > 18 GeV and |η| < 2.5. Offline
T
nents are identified, charged and neutral hadrons, plus
thresholds for electrons andmuons are E >25 GeV and
T
non-isolated leptons and photons are clustered into jets.
p > 20 GeV respectively. CMS triggers on single elec-
T
Both jet reconstruction techniques use a size R = 0.5.
trons ranging from E > 27 GeV to 42 GeV, depending
T
To be considered for top quark analyses, jets should have
ontheinstantaneousluminosityandwhetherextrajetsare
p >30 GeV and |η|<2.4- 2.5.
required in the trigger. For muons, the threshold ranges T
For t-channel single top analyses, it is important to
from p > 17 GeV/c to 30 GeV/c. Offline these thresh-
T
be able to identify and reconstruct jets as far forward as
olds range from E > 30 GeV to 45 GeV for electrons
T
possible, because typically one of the two jets in this sig-
1 Psuedorapidity η isrelated tothepolaranglewith respect natureisintheforwardregion.Therefore,singletopanal-
to the beam direction θ as follows: η = ln[tan(θ/2)]. For a ysestypically extend the |η| of jets usedin their analyses.
−
massless particle, η is equivalent to rapidity y. CDF extends its jet η range to |η| < 2.8, while D0 uses
Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review 5
|η| < 3.4. ATLAS and CMS extend their jet η ranges to particle tracks. The jet probability (jp) tagger computes
4.5 and 5.0 respectively. the probability that the tracks in the jet come from the
primaryvertex,usingtheimpactparametersofthetracks.
A variant of the jp algorithm, known as the jbp algo-
2.1.3 b-Tagging rithm, gives increased weight to the four tracks with the
highestimpactparameter.Finallythetrackcounting(tc)
Because the topquark nearly alwaysdecays into a W bo- algorithmis basedon counting the number of trackswith
son and a b quark, identifying jets that originate from b significant impact parameters. Each of these algorithms
quarks is a powerful way to distinguish top production hasmultipleoperatingpoints,allowinganoptimizationof
from its backgrounds, as well as to help resolve jet-to- tagging efficiency (which ranges from 36% to 82%) and
partonmatchingwhenreconstructingtopquarkkinemat- mistag rate (ranging from 0.2% to 13%).
ics. Algorithms for identifying b-quark jets (referred to as ATLAS has implemented two main algorithms for b-
b-taggingalgorithms)typicallyrelyonthelonglifetime of tagging[47,47]:TheJetFitteralgorithmisbasedonre-
the B hadron, either explicitly reconstructing a displaced constructed vertices significantly displaced from the pri-
vertexfromtheB decay,orbyidentifyingtrackswithhigh mary vertex. In contrast, the ip3d tagger uses the im-
impact parameters originating from the B decay. pact parameters–both longitudinal and transverse–of the
The CDF experiment primarily uses the secvtx b- tracks in the jet to compute the probability that the jet
tagging algorithm [6] based on reconstructing displaced originatesfromthe primaryvertex.These two algorithms
secondary vertices using the intersection of at least two are combined using an ANN to yield an efficiency of 60%
displaced tracks. As an alternative, the JetProb algo- for b jets and approximately a 0.3% mistag rate.
rithm[42]issometimesused.Inthisalgorithm,thetracks
in a jet are examined and a probability is calculated for
all the tracks, based on their impact parameters, to have 2.1.4 Missing Transverse Energy
originated from the primary vertex. Some CDF analyses
also make use of an artificial neural network (ANN) ap-
Leptonic W bosons decays, associated with a top quark
plied after the secvtx to increase the purity of the b-
decays, produce energetic neutrinos that cannot be di-
tagged sample. The ANN uses input variables related to
rectly detected. Instead, the presence of these neutrinos
the tagged secondary vertex, like the decay length, num-
have to be inferred by looking at the transverse momen-
ber of tracks, and invariant mass of the tracks associated
tumbalanceofthevisibleparticlesinthedetector.Toac-
to the vertex,to discriminate betweenreal b-jetsandtags
countforbothchargedandneutralparticles,thismomen-
ofjetsthatdonotoriginatefromabquark(mistags).This
tumbalanceisusuallycalculatedusingenergiesmeasured
ANN variable is typically used as an input to other mul-
inthe calorimeter,weighting the energyin eachcalorime-
tivariate analysis methods. The efficiency of the secvtx
ter tower by the sine of the polar angle sinθ. The vector
algorithmisapproximately45%withamistagratearound
sum of these weighted calorimeter energies is called the
1%.
missing transverse energy E . At CDF and D0, the E
D0 uses b-tagging algorithm that combines variables T T
calculated from the raw cal(cid:14)orimeter energy deposits, a(cid:14)nd
fromseveralb-taggingapproachesusinganANN[43].This
thencorrectionsareappliedbasedonthe calibratedener-
tagging algorithm combines the features of three tagging
giesforreconstructedobjects,likejets,photons,andelec-
algorithms: The secondary vertex tagger (SVT) is based
trons. Muons as minimum ionizing particles, are partic-
onreconstructedvertices,whilethejetlifetimeprobability
ularly important to correct for because unlike other par-
(jlip) and counting signed impact parameter (csip) tag-
ticles, they do not deposit much of their energy in the
gers are based tracks with high impact parameters. The
calorimeter. At ATLAS and CMS the E is calculated
jliptaggercomputesaprobabilityforthe jetto originate T
startingfromcalibratedquantities:topol(cid:14)ogicalclustersat
from the primary vertex, while the csip tagger is based
ATLAS and particle flow candidates at CMS.
on requiring a certain number of high-impact-parameter
tracks. The ANN tagger uses variables related to each
of these algorithms, such as the significance of the de-
2.2 Signal and Background Modeling
cay length of the secondary vertex, the invariant mass of
thetracksincludedinthevertex,thejlipprobability,and
thenumberofcsiptracks.Bycombininginformationfrom 2.2.1 Top Pair and Single Top Production
multiple taggers,the ANN tagger is able to achieve a sig-
nificantly higher efficiency for a comparable fake rate to Top quark production is simulated through a variety of
the other tagging algorithms. With this tagger, different MonteCarloprograms.Fortopquarkpairproduction,the
operatingpointsmaybeselectedtobalanceefficiencyver- tree level process is usually described by Leading Order
sus mistag rate. For example, the D0 single top analysis (LO) Monte Carlo simulations such as pythia[48], Alp-
usesanoperatingpointthatyields47%b-jetefficiencyfor gen [49] and MadGraph[50] respectively by the CDF,
a mistag rate of 0.5% [44]. D0andCMScollaborations.Thelattertwocollaborations
TheCMSexperimentusesthreedifferentb-taggingal- model tt¯production through several LO diagrams repre-
gorithms [45,46]. The simple secondary vertex (ssv) tag- senting eachtt¯plus zero, one, two or three extra partons;
gerreconstructsdisplacedverticesusingtwoormorechargedthe events are then summed to describe extra radiation
6 Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review
at tree level. The ATLAS collaborationuses the Next-to- masslessandtheirgenerationisincludedinthegeneration
Leading-Order(NLO)MonteCarloprogrammc@nlo[51] of the rest of the V+jets events.
to describe tt¯ production. Single top quark production The QCD multijet process is difficult to model with
happensthroughthes-andt-channeldiagrams,orinasso- Monte Carlo.Typically,this processis modeledusing one
ciatedtW production,wherethelatterisnegligibleatthe or more side-band regions in the data. The specific side-
Tevatronpp¯collisions.ItsproductionismodeledbyMad- band region depends on the top signature being studied.
GraphbytheCDFandCMScollaborationsandsingle- For example, in lepton + jets analyses, the QCD mul-
top[52] by the D0 collaboration. ATLAS uses mc@nlo tijet background is typically modeled using a side-band
to model the single top quark production. Several other wherethe lepton failsone ormoreselectionrequirements,
Monte Carlo programs have been used either to evaluate like lepton isolation. For the all hadronic signature, the
possible systematic biases induced by the choice of the QCDmultijetbackgroundismodeledusingasamplewith
default program, or to suit analysis-specific needs. relaxed b-tagging or kinematic selection requirements. In
All tree-level computations are passed to pythia for the eventthat a Monte Carlo modelfor this processis re-
parton shower, hadronization and underlying event, with quired, typically alpgen or MadGraph plus pythia or
the exception of mc@nlo that is passed to herwig[53] herwig is used to simulate multijet production. In some
for parton shower/hadronization, and to the subroutine cases, pythia or herwig dijet production is sufficient.
jimmy[54] for the description of the underlying event. Additional,electroweakbackgroundsasdibosonWW/WZ/ZZ
Themostcommonchoiceforpartondistributionfunctions production, are usually modeled with pythia (at CDF,
set (PDF) is the CTEQ one[55,56,57]. Tau decays are D0, and CMS) or herwig (at ATLAS).
simulated through the tauola[58] package. The decayed
particles are then passed to a full detector response sim-
ulation produced using the geant[59] program. Pile-up 2.3 Pair Production
events are added to the primary collisions through either
Monte Carlo simulation or adding real collisions recorded 2.3.1 Lepton + Jets Final State
by means of minimum bias triggers.
Althoughcommonlyreferredtoasthe“lepton+jets”final
state, this signature generally encompasses only µ+jets
and e+jets final states. Final states involving τ leptons
2.2.2 Backgrounds
aretypically handled separately,as describedbelow. This
signatureoffersanumberofadvantagesfortopquarkanal-
In top quark analyses there are two major categories of yses.Thesingleenergeticchargedleptonprovidesaconve-
backgrounds: vector boson (W or Z referred to collec- nient signal for triggering on these events. This signature
tively as V) plus jets and multijet QCD. Both of those occursinapproximately30%oftopquarkpairproduction
background categories are modeled using a different ap- events (neglecting events with taus), offering good com-
proach. promise between the purity offered by leptonic W decays
The predominant technique used to model V+ jets and the statistics offered by the hadronic W branching
production is the matched matrix element plus parton fraction. Because only the z-component of the neutrino
shower(ME+PS) approach.Exact matrix elements for V goes unmeasured, the kinematics of the top quark pair
plus different numbers of partons are calculated at lead- final state can be fully reconstructed by constraining the
ing order precision. Parton-level events generated with chargedleptonandtheneutrinotogivetheW bosonmass,
theseleading-ordermatrixelementsarethenfedtoapar- andrequiringthesamemassforthetworeconstructedtop
ton shower MC, like pythia or herwig to account for quarks.
the effects of parton shower, hadronization, and the de- The primary backgrounds for this signature are irre-
cays of unstable particles. However, there is a double- ducible W+jets production (with and without additional
counting of the phase space that can be populated both heavyflavorquarks)and QCDmultijet productionwhere
by V +N parton events and V +(N −1) parton events oneoftheseveraljetsfakesaleptonsignature.Fortypical
with hard radiation from the parton shower. To remove event selections before b-taggingis applied (referred to as
thisdouble-counting,amatchingschemeisneededtoveto the “pretag” selection), the signal (S) to background (B)
some events from each sample. The primary matching ratio S : B ranges from approximately 1 : 1 to 1 : 5 de-
scheme used is the MLM matching scheme [60], but the pendingonthenumberofjetsrequiredandthe kinematic
CKKW scheme [61,62] is also sometimes studied. selection requirements.Although the purity in the pretag
D0 and CDF use alpgen + pythia to model V+jets sample is insufficient to allow a determination of the top
production, including matrix elements up to V +5 and 4 quark cross section simply by counting events, it is possi-
partons respectively. Events including b and c quarks are ble toextractarobusttopquarksignalby fitting either a
producedasdedicatedsamples,includingtheeffectsofthe single kinematic distribution or multiple kinematic vari-
heavy quark mass. ATLAS models V+ jets production ables combined using a multivariate analysis technique
using alpgen + herwig/jimmy, again with the events (MVA). Alternatively, the signal purity can be substan-
including b and c quarks being generated separately. At tially enhanced by requiring the presence of one or more
CMS, V+ jets processes are modeled with MadGraph b-taggedjets. Using b-tagging,the S :B can be improved
+ pythia. At CMS, heavy flavor quarks are treated as to approximately 2 : 1 or more. The primary challenge
Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review 7
for extracting the top signature in the b-tagged sample 2.3.4 Events with taus
is understanding the rates of W+bottom and W+charm
production.Theoreticalpredictionsarenotsufficientlyre- Approximately 20% of top quark pair events appear with
liable at this time, so techniques to extract these rates thirdgenerationleptonsinthefinalstate.Tauleptonsare
from data have to be used. themostdifficulttoidentifyathadroncolliders,duetothe
multiple ways in which they can appear in the detector.
Thebranchingratiooftauleptonstooneormorecharged
2.3.2 Dilepton Final State and/or neutral hadron and a tau neutrino is the largest,
BR(τ →hadrons+ν )∼ 65%. Hadronically decaying taus
τ
As with the lepton + jets final state, only electrons and appearasnarrowjets;thissignatureiseasilymimickedby
muons are considered as the “leptons” of this signature. hadronic jets or electrons. The decays of tau into lighter
Although this signature has the smallest branching frac- leptons has a lower BR(τ → ℓν ν ,ℓ = e,µ) ∼ 35% and
ℓ τ
tion(around5%afterneglectingeventswithtaus),itpro- canhardly be discriminatedfromelectrons ormuons pro-
vides by far the cleanest signature. The two energetic, duced from W decays. Tau identification algorithms thus
isolated charged leptons from the two W bosons decays address only the hadronic tau decays.
makethis signatureeasyto triggeron,andthe significant Due to the large fake rate in tau identification algo-
amount ofET from two neturinos and the two energetic rithms, the requirementof a tau, largemissing transverse
b-jets make(cid:14)this channel easy to separate from the main energy and jets is not sufficient to produce a sample with
backgrounds. Drell-Yan production provides the leading reasonablesignalpurity.The mostcommonchoiceis thus
background, and because it is difficult to predict accu- to identify the additional electron or muon in the dilep-
ratelytheETtails,thisbackgroundisoftenextractedfrom ton final state, in order to increase purity. In lepton+jets
data by lo(cid:14)oking at theE tails around the Z boson mass events where the ℓ=τ, it is helpful to suppress the dom-
T
peak. Another challeng(cid:14)ing background that frequently is inant QCD background by taking advantage of the dif-
extracted from data, is the background for fake leptons. ferent kinematic and topological characteristics of these
Events from the W+ jets process can be reconstructedin events, similarly to the what is done in the all-hadronic
thedileptonfinalstateifoneofthejetsintheeventfakesa final state[63].
chargedleptonofoppositesigntothechargedleptonfrom Ultimately,themosteffectivewaytocollecttopevents
theW decay.Inaddition,QCDmultijetswillcontributeif with taus in the final state has been proven to be by re-
there are two jets providing opposite-sign charged lepton quiringlargemissingtransverseenergy,severaljets outof
fakes. Despite these backgrounds, it is common for dilep- which at least one is identified as a b-jet, and vetoing the
toneventselectionstoachievesignaltobackgroundratios presence of electrons or muons. By exploiting again the
in excess of 2 : 1 with channels like the double-b-tagged peculiar kinematics of the signal events—either by means
eµsamplehavingvirtuallynobackground.Asidefromthe of a cut based event selection or through a multivariate
lack of statistics,the mainchallenge of using the dilepton eventselection—itispossibletoisolateakinematicregion
channel comes from the presence of two neutrinos whose with large signal purity.
four-momenta cannot be measured. All of the above choices will leave the remaining top
leptonic events as a background to the tauonic signal.
2.3.3 All-hadronic final state
2.4 Single Top
Thissignaturehasseveraladvantages:thebranchingratio
BR ≈46% is the largest of all decay modes. Also, the jet 2.4.1 Lepton + Jets Final State
energy can be measured using only the calorimeter, and
the calorimetercoverageisusuallybroaderthanthespec- Unliketopquarkpairproduction,insingletopquarkpro-
trometercoverageforcolliderdetectors,allowingmaximal duction, there is only one W boson from the top quark
acceptanceto the signal.Also,one canin principle recon- decay to provide a charged lepton. Therefore (except for
struct the full final state kinematics. theassociatedtW channelwhichwillbediscussedmomen-
On the other hand, given that it is extremely difficult tarily),thereis nodileptonic finalstate,andthe lepton+
toidentifythechargeorflavorofthequarkoriginatingthe jets final state provides the cleanest event signature. As
jets,thenumberofpermutationisverylarge.Forthesame in top quark pair production, the lepton + jets signature
reason,itisverydifficulttodiscernthetopquarkfromthe typicallyonlyinvolvesthe electronandmuon,leavingthe
antitopquark,makingseveralmeasurementsunfeasiblein tau channel for the E + jets signature. For single top
T
thisdecaymode.TheQCDmultijetfinalstateisthemost quarkproduction,this(cid:14)signatureconsistsofachargedlep-
common at a hadron collider, thus isolating the tt¯signal ton, E from the neutrino decay, and two to three jets,
T
in this signature requires a detailed understanding of the of wh(cid:14)ich, one or two originate from a b quark. Because
QCD kinematics and topology.The production of six jets thissignatureinvolvesfewerjetsthantoppairproduction,
inthe finalstate ispoorlyunderstoodattheoreticallevel. the backgrounds are substantially more challenging, and
It is thus important to utilize the collider data itself to extracting the signal without the use of b-tagging is not
derive a model for the QCD background in this complex feasible. Even with b-tagging, the typical signal to back-
final state. ground ratio in this channel begins at roughly 1 : 20 at
8 Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review
the Tevatronor1:7attheLHC.Therefore,extractionof theory could indicate the presence of new physics. Fur-
the single top quark production signature typically relies ther,preciseunderstandingofthe top-quarkpairandsin-
onacombinationofb-tagging,andmultivariatekinematic gle top quark production cross sections enable searches
discriminants. for new physics in which these processes pose significant
Because the assocated tW-channel involves the decay backgrounds.Hence, measurements of the top-quark pro-
ofanadditionalW bosonproducedinassociationwiththe duction cross sections are important components of the
topquark,thissignaturediffersslightlyfromthet-channel LHC and Tevatron physics programs.
ands-channelproduction.TheW decayprovideseitheran AtCDFandD0,thettcrosssectionhasbeenmeasured
additionaljetoranextrachargedleptonandETcompared in many different channels. See Table 1 for a summary of
totheothersingletopproductionchannels.T(cid:14)hissignature a selection of recently published Tevatron results [66,67,
is only relevant at the LHC because the production cross 68,69,70,71].Duetochannel-dependentbackgroundcom-
section at the Tevatron is negligibly small. position and the tt branching ratios to the various acces-
sible final states, each channel pursued in the measure-
ment of σ has its own purity and expected yield. There
2.4.2 Events with taus tt
is great value in measuring the tt production cross sec-
tion in multiple channels then, since each attempt must
The extractionofsingle topsignalfromthe kinematically
necessarily approach the measurement in a unique way.
similarW+jetsbackgroundhasbeenamajorchallengein
Anultimate combinationofallindependentresultswould
events withwellidentified electrons andmuons;the isola-
have enhanced sensitivity.
tion of single top events with taus in the final state poses
Table 1 is not intended to be an exhaustive history of
an even greater challenge, as in addition to the W+jets
Tevatron Run II tt measurements. The most sensitive tt
background with real taus, QCD multijet process con-
cross section measurements are in the ℓ+jets and dilep-
tribute to the sample whenever a jet originating from a
ton channels; hence, the most modern published results
quark or gluon is misidentified as a tau jet.
exploiting the largest data samples are listed for those
Similarly to the analyses of events where top quarks
channels. Other preliminary high statistics Tevatron re-
are produced in pairs, leptonically decaying taus are im-
sultsareavailablewithcompetitiveorsuperiorsensitivity
plicitlyincludedinanalysesthatcollectelectronsandmuons.
but are not included here. Further complicating matters
Asofthewritingofthisdocument,onlyCDFandD0col-
is that some recent Run II published results make a dif-
laboration measured the single top production cross sec-
ferentassumptionforthemassofthetopquark,m ;since
tionineventswithtaus.Twodifferentstrategieshavebeen t
the acceptance for tt events is dependent onm , the mea-
set forward in both the triggering strategy and the isola- t
suredσ mustbequotedatsomeassumedm value.Here
tion of tau events at the two collaborations. The CDF tt t
we chose to only include results here that make the same
experiment analyzes events collected from a trigger path
assumption (m =172.5 GeV/c2) as the main ℓ+jets and
requiring large missing transverse energy from the neu- t
dileptonanalysestofacilitatecomparison.Lastlysomere-
trino, and two energetic jets, while the D0 experiment
sultshavinglargeuncertaintiesrelativetotheresultsfrom
analyzeseventscollectedfromamultitude oftriggerstyp-
the ℓ+jets and dileptonchannels,these were notincluded
icallyrequiringlargeenergydepositinthecalorimeterdue
either.
to jet activity. Events with identified electrons or muons
are rejected. Severalqualitativefeaturesoftheseresultscanbeiden-
Thetwocollaborationimplementedalsodifferentstrate- tified:First,theNLOpredictionforthettproductioncross
gies to address the otherwise dominant QCD background section at the Tevatron has been recently calculated by
toeventswithhadronicallydecayingtaus.TheD0collaboratiMono[6ch4],and Uwer [15]: σNLO = 7.5±0.7 pb. The results
tt
uses a multivariate tau identification algorithm to sup- listed in Table 1 are all completely consistent with this
press the QCD multijet production where a jet mimics a prediction from theory. Each measurement listed is con-
tau. The CDF collaboration[65] does not attempt to ex- sistent with the NLO prediction within 1σ uncertainties.
plicitly identify taus, but rather focuses on suppressing Themostprecisemeasurementofσ [66]isextracted
tt
the QCD contribution through multivariate techniques, from a measurement of the ratio R = σ /σ . In the
tt Z
exploiting the different QCD kinematics and topology. In measurement of R, several sources of systematic uncer-
both scenarios, single top decays including electrons and tainty cancel; one can then exploit the superior precision
muons will contaminate the tauonic signal. of the theoretical Z production cross section to achieve
a measured σ with significantly reduced systematic un-
tt
certainty over conventional methods. Two such measure-
3 Experimental Results ments were executed at CDF; the most precise measure-
ment discardedb-tagginginformation, due to the system-
3.1 Top Quark Pair and Single Production Cross aticuncertaintiesoneincurswhenusingtaggingvariables,
Section and exploited a neural network for final event classifica-
tion,examiningthekinematicvariablesoftheevents.This
Measurements of the top quark production cross section superior Tevatron σ result is now statistics limited; this
tt
are good tests of perturbative QCD. Deviations in the measurementwillthereforebenefitfromanupdatedresult
observed cross sections from the predictions provided by exploitingthefullTevatronRunIIstatistics.Theremain-
Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review 9
Table 1. SummaryofpublishedTevatronresultsonthettproductioncrosssection.Allmeasuredσ assumeatop-quarkmass
tt
of 172.5 GeV/c2.
Channel Experiment (fb−1) σ (pb)
Lint tt
ℓ+jets CDF 4.6 7.82 0.38(stat) 0.37(syst) 0.15(Ztheory)
(ℓ=e,µ) D0 5.3 ± 7.78+0.77±(stat+syst+±lumi)
−0.64
dilepton CDF 5.1 7.3 0.7(stat) 0.5(syst) 0.4(lumi)
ee,µµ,eµ D0 5.4 ±7.36+0.90(s±tat+syst+±lumi)
−0.79
all-hadronic CDF 2.9 7.2 0.5(stat) 1.1(syst) 0.4(lumi)
± ± ±
MET+jets CDF 2.2 7.99 0.55(stat) 0.76(syst) 0.46(lumi)
± ± ±
ing systematic uncertainty is comparable to the statisti- selectedsamplethroughaprofilelikelihoodfittothenum-
cal uncertainty, so it is important to keep the system- ber of total reconstructed jets, the number of b-tagged
atic uncertainty in mind for future analyses; dominant jets, and the secondary vertex mass distribution in the
remaining sources of systematic uncertainty include the data.Themainsystematicuncertaintiesaretakenintoac-
uncertainty on the scale for jet energy measurements and count when maximizing the profile likelihood, hence the
uncertainty in the modeling of signal tt and background systematic uncertainty on the measured tt cross section
W+jets events. is reduced with respect to more conventional techniques.
The best Tevatron measurement of σ has a relative Under these conditions, this result achieves a sensitivity
tt
totaluncertaintyof∼7%.NotethattheNLOtheoretical to σtt comparable to that of NLO theory.
predictionhasarelativeuncertaintyof∼9%.Hence,the All these 2010 LHC results will soon be eclipsed in
precision of the measurement of the tt is now exceeding terms of sensitivity by the results from the full 2011LHC
that of the theoretical prediction; to identify new physics datasample.AtthetimeofpreparationofthisReview,all
throughanobserveddiscrepancyinthemeasuredσ with of the 2011 remain preliminary. With the high statistics
tt
respect to theory, then new, more precise theoretical pre- 2011 data samples, reduction of systematic uncertainty
dictionswillbenecessary.Whileapproximatehigherorder will be the top priority in all measurements of σ .
tt
calculations exist, no complete NNLO is currently avail- Afterdiscoveryofthetopquarkthroughthestrongin-
able. Hence, many in the top-quark physics community teractionand subsequentmeasurementofthe top quark’s
are eager for the measurement of the differential produc- mass,electroweakproductionofsingletopquarksbecame
tioncrosssectionsoftop-quarkpairsinthehigh-statistics thenextmajorgoalintop-quarkphysics.Singletopquark
samples at the LHC experiments to look for possible in- productionisnotjusta curiosity;the single topcrosssec-
dications of new physics. tion is sensitive to the CKM matrix element |Vtb|, pro-
As discussed above, the main production mechanism vidingtheopportunityforfirstdirectmeasurementofthis
for top-quark pairs at the Tevatron is quark-antiquark parameter within the SM.
annihilation, whereas at the LHC top-quark pairs come Atthe Tevatron,asdiscussedabove,singletopquarks
mostly through gluon-gluon fusion. Hence, in addition to comethroughtwoproductionmechanisms,s-andt-channel
simplyprobingtop-quarkproductionatahighercenter-of- production. In early searches, the s- and t-channel mech-
mass collision energy, measurements of the tt production anismswere searchedfor together,lookingfor evidence of
cross section at the LHC also test our understanding of theircombinedcontribution.Thepredicted(s+t)channel
the top-quarkpair productionmechanismin afundamen- productioncrosssectionatthe Tevatronis 3.0-3.5pb [26,
tally new regime. Additionally, early LHC measurements 27,28,29,30,31,32]; although this is roughly half the in-
[72,73] of the tt production cross section were used to clusive tt production cross section, the task of extracting
demonstrate the health of the LHC and the general pur- thesingletopsignalismadesignificantlymorechallenging
pose experiments CMS and ATLAS. by the presence of significantly more backgrounds in the
jetmultiplicity bins inwhichthe signalresides,compared
Table2containsasummaryofthepublishedmeasure-
to the relevant jet bins for top-quark pair production.
ments of σ from CMS and ATLAS using the full 2010
tt
data sample [74,75,76,77]. New analyses exploiting the First observation of single top quark production was
fullstatisticsofthe2011LHCrunremainpreliminary[78, achievedin 2009by both CDF [11,86]and D0 [12]. These
79,80,81,82,83,84,85]. results were combined [87] yielding a measured CDF+D0
single top cross section of
Similar qualitative observations can be made regard-
ing these early LHC as were made for the summary of
TevatronRunII results.ThepredictionfromNLOtheory σt =2.76+00..5487(stat+syst) pb, (1)
indicates that σNLO = 157±24 pb [15,16,17,20]; each of −
the results in Tatbtle 2 is consistent with this NLO predic- assuming mt = 170 GeV/c2, completely consistent with
tion within 1σ total uncertainty. The best current mea- the prediction from the SM. This measured cross section
surement at the LHC comes from the ℓ+jets channel at corresponds to a measurement of
CMSinananalysisthatexploitedb-taggingineventclas-
sification. This technique extracted the tt content of the |V |=0.88±0.07(stat+syst), (2)
tb
10 Kevin Lannon et al.: Measurements of theProduction, Decay and Properties of theTop Quark:A Review
Table 2. Summary of published LHC results on thett production cross section. All measured σ assume a top-quarkmass of
tt
172.5 GeV/c2.
Channel Experiment (fb−1) σ (pb)
Lint tt
ℓ+jets - kinematics only CMS 0.036 173+39(stat.+syst.) 7(lumi)
32 ±
ℓ+jets - with b-tagging CMS 0.036 150 9(stat) 17(syst) 6(lumi)
± ± ±
dilepton CMS 0.036 168 18(stat) 14(syst) 7(lumi)
ATLAS 0.035 171± 20(stat)± 14(syst)±+8(lumi)
± ± −6
corresponding to a 95% C.L. lower limit of |V | > 0.77. 98] and dilepton channels[99,100], from the CDF mea-
tb
These results on |V | are also consistent with the expec- surements in the all hadronic[101] and MET+jets[102]
tb
tation from the SM (|VSM|∼1.0). channels, and a measurement that is largely independent
tb
D0 has also measured single top production using ex- ofthejetenergyscale[103].Theaverageoftheabovemea-
plicitly the τ+jets signature [64], extracted for the first surements using the 2001-2009 Tevatron dataset, and of
time the t-channel cross section separately with a model theearlierresultsusingthe1992-1996Tevatrondata,gives
independent technique [88], and updated the single top M =173.2±0.6 (stat)±0.8 (syst)GeV/c2,correspond-
top
cross section measurement to 5.4 fb 1 [89]. Each of these ing to a 0.56% uncertainty. The largest systematic source
−
resultsis consistentwith SMexpectationsforelectroweak comes from the uncertainty on the signal modeling. As
single top production. A review of Tevatron single top the systematics include a the JES term that scales with
quark results can be found in[90]. luminosity, a precision below 0.5% is achievable, once the
At the LHC, the early focus has been on establishing Tevatron data accumulated in the 2010-2011 running is
t-channel single top production directly. The CMS exper- incorporated.
imentperformedthefirstmeasurementoft-channelsingle The CMS collaborationmeasured the top quark mass
top production [91] at the LHC in the 36 pb−1 2010 data using the 2010 LHC dataset in the lepton+jets[104] and
sample obtaining a t-channel cross section—summing t dilepton[76]final states,finding goodagreementwith the
and t¯contributions—of σ = 83.6± 29.8 (stat.+syst.)± Tevatron result. Using 0.7fb 1 of data in the lepton +
−
3.3 (lumi.)pb.PreliminaryresultsfromCMSonasearch jetschannel,ATLASmeasuresM =175.9±0.9 (stat)±
top
forthetW channelproduction[92]andATLASonasearch 2.7 (syst)GeV/c2[105]. The precision is currently limited
forthe s-channelproduction[93]andameasurementoft- by the limited understanding of the jet energy scale, the
channelproduction[94]yieldresultsthataresofarconsis- initial and final state radiation, and tree level modeling
tentwiththestandardmodel.Withtheincreaseddatasets uncertainties. The Tevatron and LHC measurements and
availablein2011and2012the LHC experimentswillcon- their overall agreement can be seen in Fig.4.
tinue to expand the range of single top quark measure-
AllthementionedabovearecalibratedtoMonteCarlo
ments.
simulations. Thus the mass measured is effectively the
mass definition contained in the LO Monte Carlo used;
theoristsagreethat the Monte Carlomass shouldbe very
4 Measurement of the top quark mass close to the top quark pole mass. Beyond LO QCD, the
mass of the top quark is a convention-dependentparame-
The top quark mass is a free parameter in the standard ter,the other dominantconventionbeing the MS scheme.
model of particle physics and must thus be experimen- To probe further into this ambiguity, the D0 and ATLAS
tally determined. The top quark mass gives large contri- collaborations compared the measured inclusive tt¯ pro-
bution to electroweak radiative corrections, and can be duction cross section with fully inclusive calculations at
used together with other electroweak observables to in- higher-orderQCDthatinvolveanunambiguousdefinition
fer the Higgs boson mass in both SM and non-SM sce- of Mtop and compares the results to MC[106,107]. Both
narios[95]. The improvement in the precision on the top measurementsfavorthepolemasshypothesisovertheM¯S
quarkmassmeasurementintherecentyearshasnarrowed hypothesis.
significantly the mass range for the existence of the SM Due to the shortness of the top quark lifetime, the
Higgs boson. Also, the precise measurement of this pa- top quark is the only quark that can be studies before
rameter is of crucialimportance as it determines many of hadronization occurs; this fact allows a unique opportu-
the other properties of the top quark: as an example, the nity to measure directly the mass difference between a
dependenceofthetheoreticalcomputationofσtt¯fromthe quark and its antiquark as a test of the CPT symme-
top quark mass is ≈3%/(GeV/c2). try conservation[108]. The measurement of the difference
Measuring the top quark mass requires a large statis- between the top quark and antitop quark mass relies on
tics top quark sample, sophisticated analysis tools, and the same techniques that have been developed to mea-
excellent understanding of the detector response and of sure the top quark mass. The advantage here is that al-
the physics of tt¯events[96]. The current most precise es- mostallsystematicsaffectingthe M measurementcan-
top
timation[1] of the top quark mass comes from the combi- cel out in the ∆(M ) measurement as they affect the
top
nation of the CDF and D0 results in the lepton+jets[97, measurement of M and M in a highly correlated
top antitop