1 First results from ATLAS experiment on production of W 2 Z s = 7 and bosons in proton-proton collisions at √ TeV ∗ † 3 PawelMalecki

4 Henryk Niewodniczanski Inst. of Nucl. Physics, PAN, Cracow, Poland

5 First results for the W and Z boson production cross-sections at √s = 6 7 TeV obtained with the ATLAS detector are presented. The measure- 7 ments are performed in electron and muon channels, with a data sample of −1 8 around 320 nb collected between March and July 2010. Methods of signal 9 extraction and background estimation are briefly presented. The measured 10 values of W production cross-sections in electron and muon channels are 11 σW BR(W eν)=10.51 1.45 nb and σW BR(W ν)=9.58 × → ± × → ± 12 1.20 nb. The Z cross-section measurements yield σZ BR(Z ee) = × → 13 0.75 0.14 nb and σZ BR(Z )=0.87 0.14 nb. The obtained 14 results± are in good agreement× with→ theoretical predictions.± A preliminary 15 measurement of the W boson charge asymmetry is also presented.

16 1. Introduction

17 A very important point in the ATLAS program of first physics measure- 18 ments is the observation of W and Z bosons and their properties. The- 19 oretical predictions for W/Z production cross-sections are nowadays very 20 precise [1], so a fine measurement is needed to test these predictions, which 21 are sensitive to higher-order (NNLO) QCD corrections as well as Parton Dis- 22 tribution Functions (P DF ) of the proton [2]. New constraints on the latter 23 are expected to be obtained with LHC measurements of W and Z bosons. 24 In addition, these bosons as first sources of high-transverse-momentum iso- 25 lated leptons are important for detector and algorithm commissioning. 26 The existence of W and Z bosons has been first proposed by Glashow, 27 Weinberg and Salam in the 1960’s [3]. After their discovery by the UA1 28 and UA2 experiments at the Super Proton Synchrotron (SPS) at CERN [4] 29 in proton-antiproton collisions, their properties have been thoroughly mea- 30 sured at LEP, SLC, RHIC and Tevatron, at different center-of-mass energies

∗ Presented at the Cracow Epiphany Conference on the first year of the LHC, Cracow, Poland, January 10-12, 2011 † On behalf of the ATLAS Collaboration

(1) 2 proceedings printed on March 31, 2011

31 varying from √s = 500 GeV up to √s = 1.96 TeV. Further confirmation 32 of previous results, improved accuracy and extension of measurements into 33 previously unexplored kinematic domains are the current goals for LHC 34 experiments. 35 The studies presented here are based on electron and muon final states, 36 either in a lepton-neutrino pair or in two oppositely charged leptons. This 37 ensures a clean experimental signature based on high-pT isolated charged 38 leptons, which can be easily identified and separated from the background 39 dominated by QCD processes. The following analyses have been published 40 by the ATLAS Collaboration in [5].

41 2. The ATLAS detector

42 The ATLAS detector [6] is the largest of the four detectors at the Large 43 Hadron Collider (LHC) and one of two so-called general-purpose detectors, 44 designed for most precision measurements, including Standard Model (SM) 45 tests, Higgs boson searches, SUSY searches, but also aspects of B-physics 46 and Heavy Ion collisions. It provides almost full solid-angle coverage around 47 the beam crossing point and allows for detailed and precise studies of colli- 48 sion events. 49 In the ATLAS coordinate system the z-axis is parallel to the beam-line 50 and points towards the anti-clockwise direction of the beam in LHC ring. 51 The x-axis points towards the ring center, setting the φ = 0 value of the 52 azimuthal angle and the y-axis points upwards. The transverse quantities 53 Miss (like transverse momentum pT or missing transverse energy ET ) are 54 defined in the x y plane. − 55 The detector is built of concentric layers around the beam-line. The 56 innermost subsystem is responsible for tracking of charged particles. It is 57 immersed in a 2T axial magnetic field parallel to the beam-line. Further 58 layers consist of calorimeters and muon chambers. The muon system has 59 its own toroidal magnetic field surrounding the detector. The Inner De- 60 tector (ID) provides coverage over full azimuthal angle and pseudorapidity 61 within η < 2.47. It consists of three subsystems, the two innermost be- | | 62 ing silicon detectors, namely the Pixel and Semiconductor Tracker (SCT), 63 and the outer being the gaseous Transition Radiation Tracker (TRT). The 64 calorimeters surround the ID and solenoid magnet and cover up to η < 4.9. | | 65 Liquid Argon (LAr) and scintillating tiles are used as active media, with 66 lead and steel being the respective absorber material. The calorimeters are 67 segmented, both longitudinally and transversely, to allow for shower-shape 68 studies, useful for particle identification. The electromagnetic (EM) part of 69 the ATLAS calorimeter system has a depth of at least 22 radiation-lengths, 70 whereas the hadronic part is approximately 10 interaction lengths deep. The proceedings printed on March 31, 2011 3

71 Muon Spectrometer (MS) provides the possibility to measure muon tracks 72 within η < 2.7. Tracking capabilities in the MS are assured by drift tube | | 73 detectors and cathode-strip chambers.

74 3. Data samples and Monte Carlo simulations

75 The presented analysis is based on the data sample collected by ATLAS 76 between March and July 2010. It corresponds to an integrated luminos- −1 77 ity of approximately 310 - 331 nb (depending on the channel). For the 78 electron channel, the sample was selected by a Level-1 calorimetric trigger, 79 requiring an electromagnetic cluster with transverse energy ET > 10 GeV. 80 Muon events were selected by comparing the signal pattern from the Muon 81 Spectrometer with reference tables based on muon tracks, starting at an 82 effective pT threshold of 6 GeV. 83 Monte Carlo samples were generated by PYTHIA [7] using MRSTLO* 84 [2] Parton Distribution Functions. Detector simulation [8] is performed by 85 the GEANT4 [9] package.

86 4. Object reconstruction

87 4.1. Electrons

88 The reconstruction of electrons in ATLAS is based on electromagnetic 89 clusters in the calorimeter. Such a cluster is matched to a track recon- 90 structed in the Inner Detector. The matched candidate object subsequently 91 undergoes the identification procedure, based on cuts applied on electromag- 92 netic shower shapes (both longitudinal and transverse), track quality and 93 the actual distance (in η φ) between track and cluster. Three sets of selec- − 94 tion criteria are used as a reference, namely the Loose, Medium and Tight 95 cuts. They provide strongly increasing rejection power against the back- 96 ground, with a small decrease of signal efficiency. Loose criteria are based 97 on the energy deposits in the second layer of electromagnetic calorimeter, 98 Medium ones add more restrictive criteria on the shower shape in the first 99 layer and on track-quality variables. The Tight requirements, in addition 100 to even tighter track selection and shower-shape restrictions reject electrons 101 from photon conversions.

102 4.2. Muons

103 In this analysis the combined muon objects are considered. They utilize 104 the information from both the Muon Spectrometer and Inner Detector. The 105 association between the tracks from these two subsystems is performed using 2 106 a χ test comparing the differences between track parameters weighted by 107 their covariance matrices. 4 proceedings printed on March 31, 2011

108 4.3. Missing transverse energy

109 To account for a neutrino that appears in W ℓν and is not re- 110 Miss → constructed, a missing transverse energy (ET ) is calculated, based on 111 calorimeter information. A vector sum of energy deposits in cells that 112 belong to reconstructed clusters is calculated in both x and y directions, 113 Miss and then combined into ET . This is an important tool to discriminate 114 against QCD background in W ℓν signal extraction procedures. In these → 115 studies, the missing transverse energy is reconstructed assuming all energy 116 is deposited by electromagnetic processes, and then corrected for different 117 calorimeter response for hadrons. In the analysis of the muon channel, the 118 missing transverse energy is corrected for the muon momentum.

119 5. Event selection

120 5.1. Preselection of high-pT leptons

121 Data samples containing high transverse momentum leptons are first 122 selected from the datasets available after trigger filtering. The lepton can- 123 didate is required to be contained within detector acceptance. In the case 124 of electrons the region η < 2.47 is considered (being the coverage of Inner | | 125 Detector) with the exception of 1.37 < η < 1.52 barrel - end-cap transition | | 126 regions in the Calorimeter. The acceptance region for muons is defined as 127 η < 2.4. In addition, the electron cluster transverse energy is required | | 128 to be ET > 20 GeV, and the muon transverse momentum needs to satisfy 129 pT > 20 GeV. 130 The selected samples are further refined by adding requirements on lep- 131 tons. The electrons selected for W eν analysis need to pass the tight → 132 identification criteria, whereas medium cuts are applied in Z ee selec- → 133 tion. In Z and W ν selections, lepton isolation criteria are → → 134 exploited. The sum of transverse momenta of tracks surrounding the candi- 135 date in the cone of ∆R< 0.4 must not exceed 20% of the muon candidate 136 transverse momentum. Fig. 1 shows the spectra of ET and pT of tight 137 electrons and muons respectively, with data and various Monte Carlo-based 138 contributions. Further selection procedure and background suppression is 139 needed and is based on event topologies.

140 5.2. Selection of W ℓν sample → 141 In addition to a high-pT lepton, transverse missing energy is present in 142 W ℓν events, accounting for a neutrino escaping detection. A cut of 143 Miss→ ET > 25 GeV is applied in both channels. A transverse mass of W proceedings printed on March 31, 2011 5

106 ATLAS Data 2010 ( s = 7 TeV ) 106 ATLAS Data 2010 ( s = 7 TeV) 5 W → eν W → νµ 10 -1 -1 L dt = 315 nb QCD 5 QCD ∫ ∫ L dt = 310 nb → ντ Z → ee 10 W Z → µµ 4 W → ντ 10 Z → ττ Z → ττ 4 10 tt Entries / 5 GeV Entries / 5 GeV tt 103 103 102 102

10 10

1 1

20 30 40 50 60 70 80 90 100 20 30 40 50 60 70 80 90 100 p [GeV] ET [GeV] T

Fig. 1: Left: Spectrum of calorimeter cluster ET of electron candidates after tight selection. Right: Transverse momentum of muon candidates. In both channels data and Monte Carlo are compared with the simulated sample broken down into various signal and background contributions.

144 boson, defined as

Miss m = 2pℓ EMiss [1 cos(φℓ φET )] T q T T − − 145 is also reconstructed, and a cut of mT > 40 GeV is applied. Spectra of 146 transverse missing energy and transverse mass in both channels are pre- 147 sented in Fig. 2. Clearly, the cuts mentioned above are crucial to reduce 148 the overall background rate. 149 The final number of events that fulfill the final selection criteria is found 150 to be 1069 in the electron channel and 1181 in the muon channel. The con- 151 tributions of various processes to these samples will be discussed in Section 152 6.

153 5.3. Z ℓℓ event selection → 154 The selection of Z ℓℓ events imposes a requirement of the presence → 155 of two oppositely-charged leptons, either medium electrons, or muons. In 156 addition, only events with a lepton-lepton system invariant mass window 157 of 66 < mℓℓ < 116 GeV are considered. Events containing more than two 158 leptons passing selection requirements are vetoed. Figure 3 shows the distri- 159 butions of Z-boson invariant mass in each channel. Discrepancies between 160 the shapes obtained from data and Monte Carlo predictions were investi- 161 gated and result from detector misalignments and imperfect description of 6 proceedings printed on March 31, 2011

105 105 ATLAS Data 2010 ( s = 7 TeV ) ATLAS Data 2010 ( s = 7 TeV ) W → eν W → eν 4 4 10 ∫ L dt = 315 nb-1 QCD 10 ∫ L dt = 315 nb-1 QCD W → ντ W → ντ Z → ee tt 3 3 10 Z → ττ 10 Z → ee

Entries / 5 GeV tt Entries / 5 GeV Z → ττ 102 102

10 10

1 1

0 20 40 60 80 100 120 0 20 40 60 80 100 120 miss ET [GeV] mT [GeV]

105 ATLAS Data 2010 ( s = 7 TeV) Data 2010 ( s = 7 TeV) ATLAS W → νµ 104 W → νµ 104 -1 QCD QCD -1 ∫ L dt = 310 nb → ντ → ντ ∫ L dt = 310 nb W W → µµ → µµ Z 3 Z 3 Z → ττ 10 Z → ττ 10 tt tt Entries / 5 GeV Entries / 5 GeV

102 102

10 10

1 1

0 20 40 60 80 100 120 0 20 40 60 80 100 120 miss m [GeV] ET [GeV] T

Fig. 2: Top row: electron channel, bottom: muon channel. Left: Transverse missing energy in the electron (muon) channel after tight electron (isolated Miss muon) selection, Right: Transverse mass after ET cut. In each plot data and Monte Carlo samples are shown, with the latter broken down into various signal and background components.

162 magnetic field. The total number of data events passing the Z selection 163 is 70 and 109 in the electron and muon channels respectively. Background 164 contributions for this dilepton selection are discussed in Section 6. proceedings printed on March 31, 2011 7

50 Data 2010 ( s=7 TeV) 45 Data 2010 ( s= 7 TeV) 70 40 Z→ee Z → µµ ATLAS 60 ATLAS 35 -1 -1

Entries / 5 GeV L dt = 331 nb L dt=316 nb Entries / 5 GeV 30 ∫ 50 ∫ 25 40 20 30 15 20 10 5 10

0 0 60 70 80 90 100 110 120 60 70 80 90 100 110 120

mee [GeV] m µµ [GeV]

Fig. 3: Invariant mass of Z boson in electron (left) and muon (right) chan- nels.

165 6. Background estimation

166 In each signal channel two classes of background processes are under 167 consideration, namely the non-QCD (sometimes referred to as electroweak) 168 processes, including decays of W and Z bosons different from the intended 169 signal, and the QCD di-jet background. The contribution from the former 170 class to the final signal sample is estimated from Monte Carlo simulations, 171 as these processes are well-modeled at NLO and NNLO, and the associated 172 uncertainty on number of background events is therefore small. The con- 173 tribution from the latter needs however to be estimated with data-driven 174 methods (with the exception of the Z process), as the associated un- → 175 certainty on the current (LO) Monte Carlo estimations is more than 100%. 176 A summary of the background contributions to the described channels is 177 given in Table 1.

178 6.1. Backgrounds for W ℓν → 179 The contribution of non-QCD background to this channel has been as- 180 sessed with Monte Carlo samples. W τν and Z ℓℓ decays are under → → 181 consideration, as well as the tt¯ process. In the electron channel, the most 182 significant non-QCD background process is W τν with τ decaying into → 183 an electron and a pair of neutrinos, constituting 77% of the non-QCD back- 184 ground. In the muon channel, the Z events are the most numerous → 185 (49%) among electroweak backgrounds. 186 The number of QCD background events in electron channel selection is 187 Miss assessed with a fit of signal and background templates to the ET dis- 188 tribution. The shape of the templates is either obtained from simulations 8 proceedings printed on March 31, 2011

189 (in case of W ℓν signal) or determined with data-driven methods in- → 190 volving reversal of certain electron identification cuts and rejecting events 191 with isolated leptons. Results of this fit are shown in Fig. 4. The num- 192 ber of events estimated with this method is 28 3(stat) 10(syst), with ± ± 193 systematic uncertainties obtained by varying the fit range and the cuts for template extraction. In the muon channel, the number of QCD events with

300 ATLAS 250 ∫ L dt =315 nb-1

200 Data 2010 ( s = 7 TeV)

Entries / 5 GeV W → eν + W → ντ

150 QCD (data template)

100

50

0 0 10 20 30 40 50 60 70 80 90 100 miss ET [GeV]

Miss Fig. 4: Result of ET template fits in number of QCD background events estimation

194 195 high missing transverse energy and well-isolated muons is estimated with 196 the knowledge of muon-isolation criteria efficiency measured with Z → 197 events, and from background-dominated low-pT muon sample. The estimate 198 gives 23 5(stat) 9(syst) events, with a systematic uncertainty dominated ± ± 199 by the uncertainty on the isolation efficiency for QCD events.

200 6.2. Z ℓℓ background estimation → 201 Even though the number of Z ℓℓ events is much less than the num- → 202 ber of W ℓν events, the background contributions are also much lower, → 203 resulting in cleaner samples after event selection. The total number of non- 204 QCD background events in the considered mass window of 66 < mℓℓ < 205 116 GeV is 0.27 0.00(stat) 0.03(syst) in the electron channel and 0.21 ± ± ± 206 0.01(stat) 0.01(syst) in the muon channel. ± 207 The QCD background for the Z ee process is estimated from data. → 208 Electron identification requirements are relaxed from medium to loose, and 209 a fit of a Breit-Wigner function convoluted with a Gaussian one (modelling 210 the signal) and a second-order polynomial (describing the background) is 211 performed in a broader range of invariant mass. The resulting number of proceedings printed on March 31, 2011 9

212 background events is then scaled down by a rejection factor matching the 213 medium electron identification performance, derived from data. The final 214 estimate is 0.91 0.11(stat) 0.41(syst) events. The systematic uncertainty ± ± 215 is estimated by varying the electron identification criteria, changing bin sizes 216 and replacing the second-order polynomial in the fit with a first-order one. 217 In the muon channel, the QCD background is also asessed with Monte 218 Carlo simulations. Background contributions from heavy quark decays are 219 estimated to be 0.04 events with a statistical uncertainty of 0.01. The sys- 220 tematic error on Monte Carlo rates is large, and it is conservatively assigned 221 to be 100% in this study.

Process Lumi #Evts Non-QCD bkg. QCD bkg. W eν 315 nb−1 1069 33.5 0.2 3.0 28 3 10 → ± ± ± ± W ν 310 nb−1 1181 77.6 0.3 5.4 23 5 9 → ± ± ± ± Z ee 316 nb−1 70 0.27 0.0 0.03 0.9 0.1 0.4 → ± ± ± ± Z 331 nb−1 109 0.21 0.01 0.01 0.04 0.01 0.04 → ± ± ± ± Table 1: Number of selected events and background expectations for the processes under consideration.

222 7. Acceptances and efficiencies

223 The acceptance and efficiency factors are needed for the cross-section 224 measurements presented in Section 8. They describe the kinematic and ge- 225 ometrical acceptance of the detector and selection procedures, as well as 226 discrepancies between truth-level and detector-level selection. The accep- 227 tance (AW/Z ) is calculated from PYTHIA generator-level quantities. The 228 selection efficiency (CW/Z ) is, in these studies, also obtained from Monte 229 Carlo (reconstruction-level quantities), but in further analyses with larger 230 datasets most of its components will be measured from data. A summary 231 on acceptance and efficiency factors is given in Table 2.

232 7.1. W acceptance and selection efficiency

233 The W ℓν acceptance is obtained by applying the following cuts at 234 → ℓ ν truth (generator) level: pT > 20 GeV, pT > 25 GeV. Pseudorapidity regions e e 235 of η < 2.47 without 1.37 < η < 1.52 or η < 2.4 are considered. A | | | | | | 236 cut on mT > 40 GeV is also applied. The systematic uncertainty of 3% for 237 each channel results from discrepancies in Monte Carlo models of PYTHIA 238 and MC@NLO [10] and also from differences in the PDF sets. 239 The efficiency of the W ℓν selection describes the influence of truth- → 240 level vs detector-level discrepancies and consists of trigger and event re- 10 proceedings printed on March 31, 2011

241 construction efficiencies, as well as the performance of lepton selection and 242 energy scale resolution. The total systematic uncertainty on CW is 7% in 243 the electron channel and 4% in the muon channel.

244 7.2. Z acceptance and selection efficiency

245 Similarily the acceptance for Z ℓℓ is obtained, by requiring (at truth 246 → ℓ level) the presence of two leptons with pT > 20 GeV. Pseudorapidity range e e 247 of η < 2.47 without the region of 1.37 < η < 1.52 or η < 2.4 is | | | | | | 248 considered. The events are required to lie within an invariant mass window 249 of 66 < mℓℓ < 116 GeV. A 3% systematic error is assigned to account 250 for PDF uncertainties and discrepancies between calculations at NLO and 251 NNLO. 252 The selection efficiency factor CZ contains dominant factors from lepton 253 selection and trigger efficiencies, both of which are squared to account for 254 two charged leptons that need to be observed in such events. Its systematic 255 uncertainty on CZ is 9.4% and 5.5% in the electron and muon channels 256 respectively.

Process Acceptance (AW/Z ) Efficiency (CW/Z ) W eν 0.462 0.014 0.659 0.046 W → ν 0.480 ± 0.014 0.758 ± 0.030 Z → ee 0.446 ± 0.018 0.651 ± 0.061 Z → 0.486 ± 0.019 0.773 ± 0.038 → ± ± Table 2: Acceptances and efficiencies for the considered signal channels.

257 8. Cross-section measurement

258 The cross-section for W/Z production times the branching fraction of 259 considered decay modes are expressed as follows:

Nobs Nbkg σW (Z) BR(W (Z) ℓν(ℓℓ)) = − (1) × → AW (Z)CW (Z)Lint

260 The components of the right-hand side of Equation 1 have been pre- 261 sented in the previous sections. Nobs is the number of observed events in 262 each channel, Nbkg denotes the estimated number of background events, 263 and AW (Z) and CW (Z) stand for event acceptance and selection efficiency, 264 respectively. Lint is the total integrated luminosity of each sample. Results 265 of this calculation are presented in Table 3. The dominant contribution to 266 the uncertainty results from the luminosity measurement error of 11%. proceedings printed on March 31, 2011 11

Process Cross-section times branching fraction [nb] W eν 10.51 0.34(stat) 0.81(syst) 1.16(lum) W → ν 9.58 ±0.30(stat) ±0.50(syst) ±1.05(lum) Z → ee 0.75 ± 0.09(stat)±0.08(syst)±0.08(lum) Z → 0.87 ± 0.08(stat)±0.06(syst)±0.10(lum) → ± ± ± Table 3: Cross-section values measured in each described channel.

267 The obtained results of cross-section measurements are in good agree- 268 ment with the theoretical predictions for W and Z production at NNLO, 269 which are 10.46 0.52 nb and 0.96 0.05 nb, respectively. Figure 5 shows ± ± 270 the obtained results in the context of Monte Carlo predictions and previ- 271 ous measurements from PHENIX (pp collisions), UA1, UA2, CDF and D0 experiments (pp¯ collisions).

ATLAS ATLAS

Data 2010 ( = 7 TeV)s Data 2010 ( = 7 TeV)s ) [nb] 1 ν -1 -1 10 ∫ L dt = 310-315 nb ll) [nb] ∫ L dt = 316-331 nb l

→ ν → W l γ → * Z/ * ll (66 < m < 116 GeV) → W+→ l+ν ll γ W-→ l-ν

Br(W 1 Br(Z/

× -1

CDF Z/γ*→ ee (66 < m < 116 GeV) × 10 ee

* W

γ D0 Z/γ*→ ee (70 < m < 110 GeV)

σ → ν ee

/ CDF W (l/e) Z/ NNLO QCD / CDF Z/γ*→ ee/ µµ (66 < m < 116 GeV) D0 W/ → (e/µ)ν σ ee W (p )p D0 Z/γ*→ ee (75 < m < 105 GeV) -1 UA1 W→ l ν NNLO QCD ee 10 W (pp) UA1 Z/γ*→ ee (m > 70 GeV) Z/γ* (p )p ee + UA2 W→ e ν W (pp) UA1 Z/γ*→ µµ (m > 50 GeV) ± + - γ µµ - / Phenix W → (e /e )ν Z/ * (pp) W (pp) -2 UA2 Z/γ*→ ee (m > 76 GeV) 10 ee 1 10 1 10 s [TeV] s [TeV]

Fig. 5: Cross-section measurement results in context of results from previous experiments and Monte Carlo predictions. Left: W ℓν , right: Z ℓℓ . → → 272

273 9. W charge asymmetry

274 In LHC proton-proton collisions an asymmetry between the numbers of 275 positively and negatively charged W bosons is expected. This is not seen 276 in charge-symmetric proton-antiproton collisions at SPS or Tevatron. This 277 asymmetry depends on the fractions of momenta carried by each colliding + − 278 parton, and therefore the ratio of W to W bosons should depend on 279 pseudorapidity. 280 The W charge asymmetry is measured with the following quantity A: + − σfid(W ) σfid(W ) A = + − − (2) σfid(W )+ σfid(W ) 12 proceedings printed on March 31, 2011

± 281 In the definition above, σfid(W ) is the so-called fiducial cross-section, 282 measured by assuming the acceptance equal to 1.0. This reduces the possible 283 bias from theoretical predictions on this measurement. 284 Results of this measurement is presented in Fig. 6. The data sample 285 used for this measurement is the same as for the measurement of W cross- 286 section described in previous sections. Good agreement with theoretical predictions is obtained with the current statistical sensitivity.

0.5 0.5 Data 2010 ( s=7 TeV) Data 2010 ( s=7 TeV) MCNLO, CTEQ 6.6 MCNLO, CTEQ 6.6 0.4 MCNLO, HERAPDF 1.0 0.4 MCNLO, HERAPDF 1.0 DYNNLO, MSTW 08 DYNNLO, MSTW 08 Asymmetry Asymmetry → ν → νµ 0.3 W e 0.3 W

0.2 0.2

0.1 ATLAS 0.1 ATLAS -1 -1 ∫ L dt = 315 nb ∫ L dt = 310 nb 0 0 0 0.5 1 1.5 2 0 0.5 1 1.5 2  η   η 

Fig. 6: W charge asymmetry with respect to pseudorapidity in electron (left) and muon (right) channels. The error bands show the theoretical uncertainties obtained with PDF error eigenvector set at 90% C.L.

287

288 10. Summary

289 The ATLAS experiment has observed the first W and Z bosons at 290 the center-of-mass energy of 7 TeV. The production cross-sections of these 291 bosons have been measured in both electron and muon decay channels yield- 292 ing results that are in good agreement with theoretical Standard Model pre- 293 dictions. The W charge asymmetry has also been measured to be consistent 294 with Monte Carlo predictions. The presented paper summarizes the results 295 published by the Collaboration in [5]. This analysis will be updated with 296 the full 2010 data, and a large reduction of both statistical and systematic 297 uncertainty is expected, including reduced luminosity uncertainty and error 298 on selection efficiencies.

REFERENCES proceedings printed on March 31, 2011 13

299 [1] K. Melnikov, F. Petriello, Electroweak gauge boson production at hadron collid- 2 300 ers through (αS ), Phys. Rev. D 74, 114017 (2006) O 301 [2] A. Martin, R. Roberts, J. Stirling, R. Thorne, Parton Distributions and 302 the LHC: W and Z Production, Eur.Phys.J. C14 (2000) 133-145, arXiv:hep- 303 ph/9907231v1 304 [3] S. Glashow, Partial-symmetries of weak interactions, Nucl. Phys. 22 (1961) 305 579-599 306 S. Weinberg, A model of leptons, Phys. Rev. Lett. 19 1264 (1967) 307 A. Salam, Gauge unification of fundamental forces, Nobel lecture, 8 Dec 1979 308 [4] G. Arnison et al. (UA1 Collaboration), Experimental observation of isolated 309 large transverse energy electrons with associated missing energy at √s = 310 540 GeV, Phys. Lett. B122, 103 311 M. Banner et al. (UA2 Collaboration, Observation of single isolated electrons 312 of high transverse momentum in events with missing transverse energy at the 313 CERN pp¯ Collider, Phys. Lett. B122, 476 314 G. Arnison et al. (UA1 Collaboration), Experimental observation of lepton pairs 2 315 of invariant mass around 95 GeV/c at the CERN SPS Collider, Phys. Lett. 316 B126, 398 − 317 P. Baganaia et al. (UA2 Collaboration), Evidence for Z0 e+e at the CERN 318 pp¯ Collider, Phys. Lett. B129, 130 → 319 [5] The ATLAS Collaboration, Measurement of the W ℓν and Z/γ → ∗ → 320 ℓℓ production cross sections in proton-proton collisions at (s) = 7 TeV p 321 with the ATLAS detector, JHEP 1012:060,2010, CERN-PH-EP-2010-037, 322 arXiv:1010.2130v1 [hep-ex] 323 [6] ATLAS Collaboration, The ATLAS Experiment at the CERN Large Hadron 324 Collider, JINST 3, S08003 (2008). 325 [7] T. Sjostrand, S. Mrenna and P. Skands, PYTHIA 6.4 physics and manual 326 JHEP05 (2006) 026. 327 [8] The ATLAS Collaboration, The ATLAS Simulation Infrastructure, Eur. Phys. 328 J. C 70, 823 (2010) [arXiv:1005.4568 [physics.ins-det]]. 329 [9] The GEANT4 Collaboration, S. Agostinelli et al., GEANT4: A simulation 330 toolkit, Nucl. Instrum. Meth. A506 (2003) 250. 331 [10] S. Frixione, P. Nason and B.R. Webber, Matching NLO QCD and parton 332 showers in heavy flavour production, JHEP 0308 (2003) 007.