Search for a heavy particle decaying into an electron and a muon with the ATLAS detector in sqrt(s) = 7 TeV pp collisions at the LHC

This Letter presents the first search for a heavy particle decaying into an e\mu final state in sqrt(s)=7 TeV pp collisions at the LHC. The data were recorded by the ATLAS detector during 2010 and correspond to a total integrated luminosity of 35/pb. No excess above the Standard Model background expectation is observed. Exclusions at 95% confidence level are placed on two representative models. In an R-parity violating supersymmetric model, tau sneutrinos with a mass below 0.75 TeV are excluded, assuming single coupling dominance and the couplings lambda'_{311}=0.11, lambda_{312}=0.07. In a lepton flavor violating model, a Z'-like vector boson with masses of 0.70 to 1.00 TeV and corresponding cross sections times branching ratios of 0.175 to 0.183 pb is excluded. These results extend to higher mass RPV sneutrinos and LFV Z's than previous constraints from the Tevatron.

Events with e ± µ ∓ (eµ) in the final state, which played an important role in the discoveries of the tau lepton and the top quark, have a clean experimental signature and low background.Many new physics models allow an eµ signature.For example, in R-parity violating (RPV) Supersymmetric (SUSY) models [1] a sneutrino can decay to eµ.Models with additional gauge symmetry can accommodate an eµ signature through lepton flavor violating (LFV) decays of an extra gauge boson Z ′ [2].Standard Model (SM) processes that can produce an eµ signature typically have small cross sections and the eµ invariant mass (m eµ ) lies below the range favored for new physics signals.This Letter reports a search for a heavy particle decaying into the eµ final state using data taken with the ATLAS detector.The results are interpreted in terms of the production and decay of a tau sneutrino ντ and a Z ′ .Both the CDF and D0 Collaborations at the Tevatron collider have reported searches for the RPV production and decay of the ντ [3][4][5][6][7].The CDF Collaboration has also set limits on the LFV couplings as a function of the Z ′ mass [4].
The ATLAS detector [8] is a multipurpose particle physics apparatus with a forward-backward symmetric cylindrical geometry and near 4π coverage in solid angle [9].The inner tracking detector (ID) consists of a silicon pixel detector, a silicon microstrip detector, and a transition radiation tracker.The ID is surrounded by a thin superconducting solenoid providing a 2 T magnetic field, and by a finely-segmented, hermetic calorimeter.The latter covers |η| < 4.9 and provides threedimensional reconstruction of particle showers using leadliquid argon sampling for the electromagnetic (EM) compartment followed by a hadronic compartment which is based on iron-scintillating tiles sampling in the central region and on liquid argon sampling with copper or tungsten absorbers for |η| > 1.7.The muon spectrometer (MS) surrounds the calorimeters and consists of three large superconducting toroids, a system of pre-cision tracking chambers, and detectors for triggering.
The data used in this analysis were recorded in 2010 at a center-of-mass energy √ s = 7 TeV.Application of dataquality requirements results in a total integrated luminosity of 35 pb −1 with an estimated uncertainty of 11% [10].Events are required to satisfy one of the single lepton (e or µ) triggers, which have nominal transverse momentum, p T , thresholds up to 15 GeV for e and 13 GeV for µ.The trigger efficiency is measured to be 100%, with a precision of 1%, for eµ candidates containing two leptons with transverse momentum p T > 20 GeV.
To select eµ events, the electron candidate is required to have p T > 20 GeV and to lie inside the pseudorapidity regions |η| < 1.37 or 1.52 < |η| < 2.47.The event is rejected if the candidate cluster is located in a problematic region of the EM calorimeter.Electron identification and isolation requirements provide rejection against hadrons.A set of electron identification criteria based on the calorimeter shower shape, track quality and track matching with the calorimeter cluster, referred to as 'medium' in Ref. [11], is applied.In addition, a calorimeter isolation criterion E ∆R<0.2 T /E T < 0.1 is applied, where E ∆R<0.2 T is defined as the transverse energy deposited in the calorimeter within a cone of radius ∆R = ∆η 2 + ∆φ 2 < 0.2 around the electron cluster, and E T is the transverse energy of the electron.
The muon candidate must be reconstructed in both the ID and the MS.A good match of the parameters of the ID and MS tracks is required, and the p T values measured by these two systems must be compatible.Furthermore, the muon candidate must have p T > 20 GeV, |η| < 2.4, and be isolated in the ID with p ∆R<0.2 T /p T < 0.1, where p ∆R<0.2 T is defined as the sum of the p T of tracks with p T > 1 GeV within a cone of radius ∆R < 0.2 around the muon track.
Jets are reconstructed using the anti-k t jet clustering algorithm [12] with a radius parameter of 0.4.Only jets with p T > 20 GeV and |η| < 2.5 are considered.If such a jet and an electron lie within ∆R = 0.2 of each other, the jet is discarded.Leptons are only considered if they are separated from all of the remaining jets by ∆R > 0.4.Electrons and muons are also required to be separated from each other by ∆R > 0.2.
The eµ candidate events are required to have exactly one electron and one muon with opposite charge satisfying the above selection criteria.Furthermore, events have to contain at least one primary vertex reconstructed with more than four associated tracks with p T > 150 MeV.
The SM processes that can produce an eµ signature are predominantly t t, Z/γ * → τ τ , W/Z+jets, diboson, single top and QCD multijet events.Among the processes listed above, t t, Z/γ * → τ τ , single top, W W , W Z and ZZ produce electrons and muons in the final state and amount to ∼80% of the expected eµ data yield.The contributions from these processes are estimated using Monte Carlo (MC) samples generated at √ s = 7 TeV and processed with the standard chain of the ATLAS geant4 [13] simulation and reconstruction [14] using the ATLAS MC09 parameter tune [15].The event generators used are pythia 6.421 [16] (W and Z/γ * ), powheg 1.0 [17] (t t), madgraph 4 [18] (W/Z + γ), mc@nlo 3.4 [19] (single top) and herwig 6.510 [20] (W W , W Z and ZZ).The MC samples are normalized to cross sections with higher order corrections applied, as follows.The cross section is calculated to nextto-next-to-leading-order accuracy for W and Z/γ * [21], next-to-leading-order plus next-to-next-to-leading-log for t t [22], and next-to-leading-order for W W , W Z and ZZ [23].Single top and W/Z + γ cross sections come from mc@nlo and madgraph, respectively.Studies of Z/γ * → ℓℓ (ℓ = e, µ) events have shown that the lepton reconstruction and identification efficiencies, energy scale and resolution need to be adjusted in the MC to properly describe the data.The appropriate corrections are applied to the MC in order to improve the modelling of the backgrounds.
The processes W/Z + γ, W/Z+jets and pure QCD jet production give rise to background in addition to prompt leptons, which come from W and Z decay.Jets misidentified as leptons, electrons from photon conversions, and leptons from hadron decays (including b-and c-hadron decays) are classified as instrumental background.The instrumental background accounts for ∼20% of the expected eµ data yield.The dominant component of the instrumental background comes from events with one prompt lepton and one jet identified as a lepton, with an additional contribution from events with two misidentified jets.These sources are referred to as jet instrumental background, and are estimated using data.The background component initiated by prompt photons, referred to as the photon instrumental background, is estimated from MC.
The jet instrumental background is estimated using data in the following way.Two selections are defined, both requiring at least one lepton satisfying all lepton criteria.The selections are defined by the second lepton candidate: the "tight" selection requires the second candidate to pass all lepton criteria, while the "loose" selection does not enforce the lepton isolation requirement.The probability for a lepton to satisfy the lepton isolation requirement is estimated by applying the loose and tight selections on Z/γ * → ℓℓ events.The probability for a jet to satisfy the lepton isolation requirement is estimated by applying the loose and tight selections on a sample of QCD dijet events to which a cut on the missing transverse energy E miss T < 15 GeV is applied to remove W +jets events.These two probabilities, along with numbers of events selected in the loose and tight samples, are used to estimate the background in the final selected sample.The background is estimated to be 12 +10 −5 events for candidates with one electron and one jet satisfying the muon selection criteria, and 19 +28 −7 events for candidates with one muon and one jet satisfying the electron selection criteria.The dominant uncertainty on the jet instrumental background estimation comes from the p T dependence of the probability for a jet to pass the lepton isolation criterion.The background due to two jets satisfying the lepton requirements is estimated to be 1.3 +5 −1.3 events from same-sign dilepton events in data with the subtraction of contributions estimated from MC for all physics processes except the QCD multijet.The overall jet instrumental background is found to be 29 +30 −10 events.This background level has been checked in simulated samples, which agree with the data driven estimates.
The dominant photon instrumental background comes from the W (→ µν)γ and Z(→ µµ)γ processes where the photon is reconstructed as an electron.A photon can be reconstructed as an electron if it lies close to a charged particle track or the photon converts to e + e − after interacting with the material in front of the calorimeter.The photon instrumental background is found to be 4.0 ± 0.7 events.
Table I shows the number of events selected in data and the estimated background contributions with their uncertainties.A total of 160 eµ candidates are observed in data while the expectation from SM processes is 163 +34 −18 events.The dominant sources of systematic uncertainty for the SM prediction arise from the uncertainty on the probability for a jet to satisfy the lepton isolation requirement (70% for electron, 30% for muon), theoretical cross sections on the physics background processes (5% -15%), and the integrated luminosity (11%).Other systematic uncertainties from the lepton trigger, reconstruction and identification efficiencies, energy/momentum scale and resolution have been included and are small.Since no excess is observed in data, limits are set for the production of ντ in RPV SUSY models and an LFV Z ′ -like vector boson.
The RPV production of ντ by d d annihilation decay- ing into eµ is considered.With the assumption of single coupling dominance [24], fixing all RPV couplings but λ ′ 311 (ν τ to d d) and λ 312 (ν τ to eµ) to zero, and that ντ is the lightest supersymmetric particle, the contributions to the eµ final state originate from the ντ only.The signal cross section depends on the ντ mass (m ντ ), λ ′ 311 and λ 312 .The third-generation ντ is considered since stringent limits exist on the electron sneutrino and muon sneutrino [1].The couplings λ ′ 311 = 0.11 and λ 312 = 0.07, compatible with the current indirect limits [1], are chosen as a benchmark point.
An eµ resonance can be generated in models containing a heavy neutral gauge boson with non-diagonal lepton flavour couplings, Z ′ [25].Very stringent limits on the combination of the mass and the coupling to ee and eµ of such models have been inferred from searches for rare muon decay [2].Using the data presented in this Letter, a limit on the production cross section times branching ratio to eµ can be placed on a Z-like vector boson.To calculate the acceptance and efficiency, the Z ′ is assumed to have the same quark and lepton couplings as the SM Z.
MC events with ντ or Z ′ decaying into eµ are generated with herwig [20,26] or pythia, respectively.Samples are produced with sneutrino masses ranging from 0.1 to 1 TeV, and Z ′ masses from 0.7 to 1 TeV.
The eµ invariant mass distribution is presented in Fig. 1 for data, background contributions and two possible new physics signals: a ντ with m ντ = 650 GeV and a Z ′ with m Z ′ = 700 GeV.The cross section is 0.31 pb for m ντ = 650 GeV [27] and 0.61 pb for m Z ′ = 700 GeV [28].The corresponding overall acceptance times efficiency is 55% for ντ and 50% for Z ′ .
The m eµ spectrum is examined for the presence of a new heavy particle.For m ντ < 500 GeV, the search region for specific m ντ is defined to be (m ντ − 3σ, m ντ + 3σ), where σ is the expected m eµ resolution (e.g., σ ≃ 15 GeV for m ντ = 400 GeV).For higher m ντ , the region m eµ > 400 GeV is used.The expected and observed 95% C.L. upper limits on σ(pp → ντ ) × BR(ν τ → eµ) are calculated using a Bayesian method [29] with flat prior for signal cross section as a function of m ντ .Fig. 2a 0 100 200 300 400 500 600 700 800 900 1000 Events / 50 GeV shows the expected and observed limits, as a function of m ντ , together with the ±1 and ±2 standard deviation uncertainty bands.The expected exclusion limits are determined using simulated pseudo-experiments containing only SM processes by evaluating the 95% C.L. upper limits for each pseudo-experiment at each value of m ντ .The median of the distribution of limits is shown as the expected limit.The ensemble of limits is also used to find the 1σ and 2σ envelope of the expected limits as a function of m ντ .For a sneutrino with mass of 100 GeV (1 TeV), the limit on the cross section times branching ratio is 0.951 (0.154) pb.The theoretical cross sections for λ ′ 311 = 0.11, λ 312 = 0.07 and λ ′ 311 = 0.10, λ 312 = 0.05 are also shown.Sneutrinos with masses below 0.75 (0.65) TeV are excluded using λ ′ 311 = 0.11 and λ 312 = 0.07 (λ ′ 311 = 0.10 and λ 312 = 0.05).The results improve on the previous CDF 95% C.L. limit of 0.56 TeV assuming λ ′ 311 = 0.10 and λ 312 = 0.05.The 95% C.L. observed upper limits on λ ′ 311 as a function of m ντ are shown in Fig. 2b for three values of λ 312 , together with the exclusion region obtained from the D0 experiment [7].The limits derived here extend to higher mass regions.
A similar method is used to set limits on the LFV Z ′like vector boson, using only events with m eµ > 400 GeV.Finding no events in the data, the 95% C.L. upper limits on σ(pp → Z ′ )×BR(Z ′ → eµ) are set, as shown in Fig. 3.The expected limit is the same as the observed limit because the median background event count expectation is also zero.For a Z ′ with mass of 700 GeV (1 TeV), the limit on the cross section times branching ratio is 0.175 (0.183) pb.This result improves upon previous CDF limits by probing a higher mass range of Z ′ -like vector particles.In conclusion, a search has been performed for a heavy particle decaying into the e ± µ ∓ final state using pp collision data at √ s = 7 TeV recorded by the ATLAS detector.The data are found to be consistent with the SM prediction.Exclusions are placed on two representative models at 95% C.L. In an RPV SUSY model, tau sneutrinos with a mass below 0.75 TeV are excluded, assuming single coupling dominance and coupling values λ ′ 311 = 0.11, λ 312 = 0.07.Higher values of the RPV coupling are also excluded as a function of m ντ .In an LFV model, extra Z ′ -like gauge bosons are excluded with a cross section times branching ratio above 0.183 pb, assuming m Z ′ = 1 TeV.These results extend to higher mass RPV sneutrinos and LFV Z ′ s than previous constraints from the Tevatron.
We thank CERN for the very successful operation of the LHC, as well as the support staff from our institutions without whom ATLAS could not be operated efficiently.The crucial computing support from all WLCG partners is acknowledged gratefully, in particular from CERN and the ATLAS Tier-1 facilities at TRIUMF (Canada), NDGF (Denmark, Norway, Sweden), CC-IN2P3 (France), KIT/GridKA (Germany), INFN-CNAF (Italy), NL-T1 (Netherlands), PIC (Spain), ASGC (Taiwan), RAL (UK) and BNL (USA) and in the Tier-2 facilities worldwide.

FIG. 2 :
FIG. 2: (a) The observed 95% C.L. upper limits on σ(pp → ντ ) × BR(ντ → eµ) as a function of mν τ .The expected limits are also shown together with the ±1 and ±2 standard deviation uncertainty bands.The theoretical cross sections for λ ′ 311 = 0.11, λ312 = 0.07 and λ ′ 311 = 0.10, λ312 = 0.05 are also shown.(b) The 95% C.L. upper limits on the λ ′ 311 coupling as a function of mν τ for three values of λ312.Regions above the three curves represent ranges of λ ′ 311 values that are excluded.These results are compared to the exclusion region obtained from the D0 experiment.

TABLE I :
Estimated backgrounds in the selected sample, together with the observed event yield.