Table Of ContentCharge asymmetry of top quark-antiquark pairs
Antoine Chapelain
CEA-Saclay, IRFU/SPP
email: [email protected]
Abstract
charge asymmetry at the Tevatron and LHC is there-
fore complementary.
Inthisnotewepresentthechargeasymmetrymeasure-
mentsoftopquark-antiquarkpairsathadroncolliders.
1 Introduction
4
1
0 Among the known twelve fermions, which are the fun-
2 damental bricks of matter, the top quark is the latest
to have been discovered at the Tevatron Fermilab Col-
n
a lider near Chicago by the CDF and D0 experiments in
J 1995 [1]. The top quark was found to be the heaviest
7 particle ever observed. Its mass is about the mass of
2 the gold atom, which is extremely heavy for a point-
like particle. Due to its large mass, studying the top
]
x quarkcouldbeawindowtowardsso-callednewphysics,
e i.e., physics that lies beyond the Standard Model of
-
p particle physics. The top quark can be scrutinized at
e hadron colliders since the Tevatron and LHC colliders
h produced numerous top quark-antiquark 1 pairs. The Figure 1: Rapidity distributions of the top quark and
[ chargeasymmetryisoneofthepropertiesoftopquark- antiquark at the Tevatron (top) and LHC (bottom).
1 antiquark pairs. Indeed the strong interaction predicts
v thatwhenproducedthroughquark-antiquarkcollisions Thechargeasymmetrycanalsobemeasureddirectly
6
the top quark and antiquark are not produced isotrop- using the leptons coming from the decay of the top
3
ically. The top quark is preferentially produced in the quark and antiquark since the flight direction of these
8
6 directionoftheincomingquarkwhilethetopantiquark leptonsiscorrelatedwiththeflightdirectionofthetop
. is preferentially produced in the direction of the in- quark/antiquark. Thismeasurementissimplerbecause
1
0 comingantiquarkintheincomingquark-antiquarkrest the flight direction of the leptons is directly measured
4 frame. As the top quark and antiquark have oppo- inthedetectorwhilethetopquarkflightdirectionneed
1 site electric charge, it will result in a charge asymme- tobereconstructedfromthedecayproductsofthetop
v: try (excess of positive/negative charge in the incoming quark.
i quark/antiquark direction). In 2011 the CDF and D0 collaborations reported
X
To quantify this effect we use the rapidity y measurements higher than the predictions as summa-
r
a (or pseudorapidity η). It is defined approximatively rized in Figure 2. The Tevatron stopped data taking
as y (cid:39)−ln(tan(θ/2)) where θ is the angle between inSeptember2011. Updatingtheasymmetrymeasure-
the flight direction of the top quark (antiquark) and ment with the full CDF and D0 recorded dataset is
the beam direction. Figure 1 shows how the charge underway.
asymmetry appears at the Tevatron and LHC. At the In Sec. 2 we will focus on the measurement per-
Tevatron, which is a proton-antiproton collider, there formed at D0 in the dilepton channel [2] and in Sec. 3
is a forward-backward (or right-left) asymmetry since onthemeasurementperformedatATLASinthedilep-
the incoming quark-antiquark collision frame is almost ton channel at 8 TeV. Section 4 summarizes the inclu-
equaltothequark-antiquarkrestframe. SincetheLHC sivechargeasymmetrymeasurementsathadroncollid-
is a proton-proton collider the quarks carry on average ers.
a higher momentum than the antiquarks which come
from the sea of the proton. The top quark will be
2 Dilepton measurement at D0
thus emitted more forward or backward and the top
antiquark will be emitted more central. Measuring the
Thett¯dileptonfinalstate(seeFig.3),ordileptonchan-
1Each fermion has a corresponding antifermion, which have nel, is characterized by two leptons with opposite elec-
thesamepropertiesbutoppositeelectriccharges. tric charge, at least two jets coming from the two b-
1
2
300
Top Quark Asymmetry DØ, L=9.7 fb 1 tt
Z
CDF L+jet (5.3 fb-1) 250 Instrum.
15.8 – 7.5 % Diboson
Data
D0 L+jet (5.4 fb-1) 19.6 – 6.5 % 0.4200
Lepton Asymmetry AFlB ents/150
D0 Dilepton (5.4 fb-1) 5.8 – 5.3 % Ev100
D0 L+jet (5.4 fb-1)
15.2 – 4.0 % 50
D0 combination (5.4 fb-1)
11.8 – 3.2 % 0 3 2 1 0 1 2 3
Lepton Asymmetry All q × η
D0 Dilepton (5.4 fb-1)
5.3 – 8.4 %
DØ, L=9.7 fb 1 tt
Bernreuther & Si, Phys.Rev., D86 (2012) 034026 100 Z
Instrum.
0 5 10 15 20 Diboson
Asymmetry (%) 80 Data
4
0.
s/ 60
nt
e
v
Figure 2: Summary of the asymmetry measurements E 40
at the Tevatron in 2011.
20
0
4 3 2 1 0 1 2 3 4
quarks,andmissingenergyduetothetwoneutrinoses- ∆η
capingthedetector(theneutrinosandthechargedlep-
tonsarebothcomingfromthedecayoftheW bosons).
Figure 4: Rapidity distributions used to compute the
asymmetry in the dilepton channel at D0 [2].
erage of the detector. Once these corrections are made
we can compare the measurements to the theoretical
predictions. Table1showsthemeasuredandpredicted
values. Both are agreement within the uncertainties.
It is interesting to look at the ratio A(cid:96) /A(cid:96)(cid:96) since the
FB
Figure 3: Topology of a dilepton tt¯ event produced two asymmetries are strongly correlated and because
through quark-antiquark annihilation at the Tevatron. the systematic uncertainty is reduced due to cancella-
tions. Figure 5 shows the measured value in black to-
gether with different predictions. The measured value
Thedileptonchannelsuffersfromsmallstatisticsbe-
of 0.36±0.20 is in agreement with the prediction of
cause of a small branching ratio but on the other hand
0.79±0.10attheleveloftwostandarddeviations. Fig-
haveasmallamountofbackground. Themeasurement
ure 6 summarizes all the current measurements at the
of the forward-backward asymmetries through leptons
Tevatron. We can see that the tensions between mea-
is performed using the two distributions in Fig. 4. The
surementsandpredictionsobservedin2011(seeFig.2)
single-lepton A(cid:96) asymmetry is defined with the q×η
FB vanished. Two measurements from D0 have still to be
distributionlookingateachleptonindependentlyifthe
releasedin2014. ThefocusisnowontheCDF-D0com-
leptongoesintheforward(η >0)orbackward(η <0)
bination of these different results to achieve the best
direction. The∆ηdistributionbuiltasthedifferenceof
possible precision.
lepton pseudorapidities is used to measure the lepton-
pairA(cid:96)(cid:96) asymmetry. A(cid:96) andA(cid:96)(cid:96) arecomputedasthe
FB
relative difference between the forward and backward
regionoftherelevantdistributionsusingthedatafrom Measured Predicted
which we subtracted the expected background. A(cid:96) 4.4±3.7±1.1 3.8±0.3
FB
In Fig. 4 the black dots represent the data, the col- A(cid:96)(cid:96) 12.3±5.3±1.5 4.8±0.4
oredhistogramsrepresentthepredictions: thett¯signal
in red, and the different backgrounds in grey, yellow
and blue. At this level we performed the asymmetry Table1: Measuredandpredictedvaluesofthetwolep-
measurements in the detector, i.e., distorted by detec- tonic asymmetries in the D0 dilepton channel. The
toreffectsthatneedtobecorrectedfor. Wefirstcorrect first uncertainty on the measured values is statistical
for the selection efficiency, i.e., for the fact that we do and the second is systematic.
not observe all the produced dilepton tt¯events in the
detector. We then correct for the limited spatial cov-
4. CURRENT STATUS AND CONCLUSION 3
%) DØ, L=9.7 fb 1 Top Quark Asymmetry
l (AFB 15 DMaCt@a NLO 3 σ CDD0 FL +Lj+ejte (t5 (.94. 4fb f-b1)-1) 19.6 – 6.5 %
Model 1 16.4 – 4.7 %
Model 2 2 σ Single-lepton Asymmetry AFlB
SM NLO D0 Dilepton (9.7 fb-1)
10 1 σ D0 L+jet (9.7 fb-1) |y|<1.5 4.4 – 3.9 %
l 4.7 – 2.6 %
CDF Dilepton (9.1 fb-1)
7.2 – 6.0 %
CDF L+jet (9.4 fb-1)
9.4 – 3.2 %
5 Lepton-pair Asymmetry All
D0 Dilepton (9.7 fb-1)
12.3 – 5.7 %
CDF Dilepton (9.1 fb-1)
7.6 – 8.1 %
Bernreuther & Si, Phys.Rev., D86 (2012) 034026
0 5 10 15 20 0 5 10 15 20
All (%) Asymmetry (%)
Figure 5: A(cid:96) versus A(cid:96)(cid:96) in the dilepton channel at Figure 6: Summary of the asymmetry measurement at
FB
D0[2]. Theblackdotrepresentsthemeasurementwith the Tevatron in 2013.
the uncertainty ellipses corresponding to 1, 2 and 3
standard deviation. “SM NLO” represents the most
recent theoretical prediction, “MC@NLO” is the event and detector effects. We see that the reconstructed
generatorusedtosimulatedthett¯signaland“Model1” distribution reproduces the behavior of the truth dis-
and “Model 2” are two new physics model that could tribution well.
explain the 2011 observed tension at the Tevatron be-
tween measurements and predictions.
3 Dilepton measurement at AT-
LAS
As explained earlier, measuring the charge asymme-
try at the LHC and the Tevatron is complementary.
This section is focusing on the charge asymmetry mea-
surement in the dilepton channel at ATLAS. Both
the asymmetry of the lepton coming from the top
quark/antiquarkandofthett¯pairsaremeasured. The
top quark is not directly observed in the detector due
to its very short lifetime (10−23 s). Thus we need to Figure 7: Rapidity distributions at truth (red) and re-
reconstructitskinematicfromitsobserveddecayprod- constructed (blue) level.
ucts. To do so we use the energy and momentum con-
versationateachdecayverticesofthedecaychain. We
Theasymmetryiscomputedusingtheobservablede-
obtain then a system of 16 equations and 22 unknowns
fined as ∆|y| = |y |−|y |. In the example of
top antitop
which cannot be solved. Making several assumptions
Fig. 7 we are able to reconstruct the correct sign of
and fixing the masses of the W bosons and the top
∆|y| in 70 % of the cases. This performance is rather
quarks to their measured values we finally end up with
satisfyingandverysimilartoperformancesofotherre-
18 equations and 18 unknowns. For a given event we
constructionmethod. Thismeasurementat8TeVwith
obtain several solutions. We define a weight for each
theATLASdetectorisstillongoingandwillbereleased
solution according to its probability to be a tt¯event.
soon.
This probability is computed using the matrix element
of the gg → tt¯process. The solution with the highest
weight is selected. This method is called the “Matrix 4 Current status and conclusion
Elementmethod” [3]. Weusethesimulationtotestthe
performances of this reconstruction method. The vari- The Tevatron and LHC are both the most powerful
able we are interested in to compute the asymmetry is proton-antiproton and proton-proton colliders, respec-
therapidityy ofthetopquarkandantiquark. Figure7 tively. They allow to conduct complementary studies
shows the y distribution at the so-called “truth” level on the charge asymmetry of the top quark-antiquark
andafterreconstruction. Thetruthleveliswhatisgen- pairs. In 2011 the Tevatron measurements showed ten-
erated with the simulation and the reconstructed level sionbetweenmeasurementsandpredictions. Thelatest
is what we reconstruct after the simulation of physics resultswiththefullstatisticsrecordedbytheCDFand
4 REFERENCES
D0experimentstendtoindicateabetteragreementbe-
tween predictions and measurements. At the LHC, so
far all the measurements are in good agreement with
the predictions. Some physics model beyond the Stan-
dard Model could explain the deviations observed at
the Tevatron in 2011 while still in agreement with the
observation at the LHC (see Fig. 8). We can see on
Fig.8thatasmallregionofphasespaceisstillallowed
forthesenewphysicsmodel. ThenewresultsfromD0,
as well as new results from ATLAS and CMS are ex-
pected to be able to make a conclusive statement. The
year 2014 is thus very promising to understand deeper
thechargeasymmetryofthetopquark-antiquarkpairs.
0.08
ATLAS G Models from:
µ
PRD 84 115013,
0.06 arXiv:1107.0841
W′ Ω4 ω4
0.04
φ
C
A
0.02
SM
ATLAS
0 CMS
-0.02 DF 0
C D
0 0.1 0.2 0.3 0.4 0.5
A
FB
Figure 8: Summary of the measurements at the Teva-
tron and LHC and predictions from different physics
models [4].
References
[1] The CDF Collaboration, Phys.Rev.Lett. 74.2626
(1995); The D0 Collaboration, Phys.Rev.Lett.
74.2632 (1995).
[2] The D0 Collaboration, Phys.Rev. D88 (2013)
112002.
[3] F. Fiedler, A. Grohsjean, P. Haefner, P. Schiefer-
decker, Nucl.Instrum.Meth. A624 (2010), 203-218.
[4] TheATLASCollaboration,arXiv:1311.6724(2013).
