Search for R -Parity Violating Supersymmetry and Long Lived Particles with the ATLAS Detector

. We present the results of two recent searches for physics beyond the Standard Model in pp -collision data with a center-of-mass energy of 7 TeV collected by the ATLAS detector at the Large Hadron Collider. The ﬁrst analysis searches for heavy particles which decay a signiﬁcant distance from their production point into a ﬁnal state containing a high-momentum muon, and is based on a 33 pb − 1 data sample. The second is a search for a heavy neutral particle decaying into an e ± m ∓ ﬁnal state and is based on 0 . 87 fb − 1 of data. Both searches are motivated by models of R -partity violating supersymmetry, although alternative new physics interpretations are also considered in both cases.


SEARCH FOR EVENTS WITH DISPLACED VERTICES
One generic feature of RPV SUSY models for specific ranges of coupling constants is the prediction of massive longlived particles [2]. We present here the results of a search for a new heavy particle decaying into several charged particles in events containing a high-momentum muon, at a distance of order millimeters to centimeters from the pp interaction point. Such a signature might arise from a neutralino LSP decaying asχ 0 → µμ → µq i q j via an intermediate virtualμ and the RPV coupling λ ′ 2i j [3]. Events are triggered by the presence of a high-momentum muon. We identify the primary vertex (PV) in the event by requiring at least five well-reconstructed tracks; for events with multiple PV candidates, we select the one with the highest scalar sum of the momenta of associated tracks. We then select tracks which have a large impact parameter 1 Copyright CERN for the benefit of the ATLAS Collaboration 2 The long-lived particle search presented here is sensitive to lifetimes cτ ∼ 1-100 mm, but the LSP lifetime may be much shorter or much longer. relative to the PV and, from this set of tracks, use an iterative procedure to construct secondary vertex candidates. We require the selected tracks to have a good fit (χ 2 /degrees of freedom< 5) to the vertex. We require that the vertex be located within the central fiducial volume of the pixel detector, to ensure the most precise measurement of the secondary vertex location, and that the vertex be displaced at least 4 mm from the PV. The typical spatial resolution on the vertex position ranges from tens to hundreds of microns, depending on the decay position.
Dominant backgrounds to this search arise from combinatorial vertices and from material interactions. In order to suppress these backgrounds, we require that the number of tracks associated to the secondary vertex be at least four, and that the invariant mass of all associated tracks be greater than 10 GeV. We further veto vertices which are reconstructed in regions of high material density, using a material map derived from minimum-bias events.
Simulated event samples are used to estimate the expected backgrounds, with the efficiencies for events to pass the muon and trigger requirements evaluated independently of the secondary vertex selection. Based on simulation studies, we set a conservative limit on the expected background, N BG < 0.03 events in our 33 pb −1 sample. The combinatorial vertex background is independently estimated in a sample of non-diffractive events in data and found to agree with the simulation, and the efficiency of the vertex mass cut is independently validated in a jet-triggered event sample. The dominant systematic uncertainties are due to the track-reconstruction and muon identification efficiencies as functions of the track impact parameter relative to the PV. These are studied in control samples with displaced tracks, including cosmic-ray muons and pions from K 0 S decay, and the simulation is found to describe the data within 3-8%. Figure 1 shows the results of this search. The left figure shows the distribution of vertex mass versus number of tracks observed in data. No events are observed in the signal region. Based on this observation, we set upper limits on the production cross section times the branching fraction for various SUSY models. For a 150 GeV squark, this limit is below the expected 95 pb cross section for most of the sensitive range of cτ. We also set a model-independent limit on the cross section times branching fraction times acceptance of 0.09 pb at 95% C.L.

SEARCH FOR A RESONANCE DECAYING TO e ± µ ∓
R-parity violating SUSY includes couplings which violate lepton flavor conservation. We present the results of a search for a new heavy neutral particle decaying to an e ± µ ∓ pair [4]. In RPV SUSY, this may arise via the sneutrino decaỹ ν τ → e − µ + with theν τ produced via the RPV coupling λ ′ 311 and decaying via λ 312 [2]. The search is sensitive to other new physics processes as well, including new gauge bosons Z ′ with lepton-flavor-violating couplings [5].
Events are triggered by the presence of a high momentum electron or muon. An opposite charge eµ pair is reconstructed with both leptons well isolated from hadronic activity in the event. This search is an update of a previous ATLAS search [6] with a data sample 25 times larger, tighter lepton momentum and isolation requirements, and a more robust data-driven background estimation technique. Two types of background processes are considered. Events with two real prompt leptons, including tt, Z → ττ, and WW , are estimated using simulated event samples. Instrumental backgrounds, including fake leptons and secondary leptons from b-and c-hadron decays, are estimated in data using a matrix technique based on looser lepton identification criteria. We extrapolate backgrounds into the signal region by measuring signal and background efficiencies in dedicated control regions in data: Z → e ± e ∓ /µ ± µ ∓ for electron and muon efficiencies, and QCD dijet events for jet fake rates. After background estimation, good agreement is seen between data and simulation in a number of important kinematic variables, giving confidence that a potential high-mass resonance would also be well described. Figure 2 shows the results of the e ± µ ∓ resonance search. The left figure shows the e ± µ ∓ invariant mass spectrum observed in data and the expected backgrounds. No evidence for a resonant structure is seen. The figure on the right shows the observed 95% C.L. upper limits on theν τ cross section times branching ratio as a function of theν τ mass. These limits improve upon the previous ATLAS result by about one order of magnitude. We also set limits on Z ′ production; at 700 GeV (2 TeV) Z ′ mass, the cross section times branching ratio is less than 10.6 (10.0) fb.

SUMMARY AND OUTLOOK
We have presented the results of two recent searches for R-parity violating supersymmetry with the ATLAS detector, one for new heavy particles produced in association with a muon and decaying with a long lifetime, and the other a search for a new heavy resonance decaying to e ± µ ∓ . No excesses beyond the Standard Model backgrounds have been seen yet, but there is an active ongoing search program at the LHC and we continue to push the limits to constrain the allowable SUSY parameter space.