Search for electroweak production of charginos, neutralinos and sleptons using leptonic final states in pp collisions at √s = 8 TeV

Abstract

(Note: This revision of the original 2012/11/27 document includes a correction in the same-sign analysis section.) Final states with exactly three leptons, four leptons, two same-sign leptons, two opposite-sign-same-flavor leptons plus two jets, or two opposite-sign leptons inconsistent with Z boson decay, are studied using a data sample consisting of an integrated luminosity of 9.2 fb^-1 of proton-proton collision data collected in 2012 with the CMS detector at √s = 8 TeV. The observed event rates are in agreement with expectations from the standard model. The results are used to set limits on the direct production of charginos, neutralinos, and sleptons. (The limits are unaffected by the same-sign analysis correction).

Detailed documentation can be found at SUS-12-022.

Approved Tables and Plots from SUS-12-022

( click on plot to get .pdf )

Introduction

Tables and Figures	Abbreviated Caption
	Figure 1a: Chargino-neutralino production at LHC leading to a three-lepton final state with missing transverse energy. The dotted and dashed lines indicate unstable particles and the LSP, respectively.
	Figure 1b: Chargino-neutralino production at LHC leading to a three-lepton final state with missing transverse energy. The dotted and dashed lines indicate unstable particles and the LSP, respectively.
	Figure 2a: Chargino-neutralino production leading on-shell W and Z bosons with missing transverse energy.
	Figure 2b: Neutralino-neutralino production leading to two on-shell Z bosons with missing transverse energy.
	Figure 3a: Chargino-pair production leading to opposite-sign lepton pairs with missing transverse energy. Each chargino can decay via either mode, giving four possible diagrams sharing the same final state.
	Figure 3b: Slepton-pair production leading to opposite-sign lepton pairs with missing transverse energy. In this diagram, the sleptons and thus the daughter leptons share the same flavor.

Search in the three-lepton final state

Tables and Figures	Abbreviated Caption
	Figure 4: MT versus Mll for the trilepton data events with an OSSF dilepton pair.
	Figure 5: Observed missing transverse energy distribution for the data events with an OSSF dilepton pair. Standard model expectations are also shown. ”Rare” refers to standard model processes such as ZZ, ttZ, ttW, and triboson production.
	Figure 6: MT versus Mll for the trilepton data events without an OSSF dilepton pair.
	Figure 7: Observed MET distribution for the data events without an OSSF dilepton pair or a hadronic tau-lepton. SM expectations are also shown. ”Rare” refers to standard model processes such as ZZ, ttZ, ttW, and triboson production.
	Figure 8: MT versus Mll for the trilepton data events containing a same-sign dilepton pair and a hadronically reconstructed tau-lepton.
	Figure 9: Observed MET distribution for the data events with a same-sign dilepton pair but with one hadronically reconstructed tau-lepton. SM expectations are also shown. ”Rare” refers to standard model processes such as ZZ, ttZ, ttW, and triboson production.
	Figure 8a: MT versus Mll for the trilepton data events containing an opposite-sign opposite-flavor dilepton pair and a hadronically reconstructed tau-lepton.
	Figure 9a: Observed MET distribution for the data events with an OSOF dilepton pair and a hadronic tau-lepton. SM expectations are also shown. ”Rare” refers to standard model processes such as ZZ, ttZ, ttW, and triboson production.
	Table 1: SM expectations and observations for events with an ee or mm OSSF pair, where the third lepton is either an electron or muon.
	Table 2: SM expectations and observations for eem and emm events without an OSSF pair or a hadronic tau lepton candidate.
	Table 3: SM expectations and observations for events with a same-sign ee, em or mm pair and a hadronic tau candidate.
	Table 4: SM expectations and observations for events with an opposite-sign em pair and a hadronic tau candidate.

Search in the four-lepton final state

Tables and Figures	Abbreviated Caption
	Figure 10: MET versus Mll for four-lepton data events with an on-Z OSSF pair and no taus. The legend indicates the flavor breakdown of events.
	Table 5: Expected SM backgrounds and observed yields for exclusive channels of four-lepton final states. All categories have four leptons including an OSSF pair consistent with a Z boson. The three sections refer, respectively, to events with only one OSSF dilepton pair with no taus, one OSSF dilepton pair with one tau, and two OSSF pairs with no taus.

Search in the same-sign two-lepton final state

Tables and Figures	Abbreviated Caption
	Figure 11: MET versus HT for same-sign dilepton candidate events. This plot tries to show how the candidate events distribute along the MET, HT plane. The horizontal line at MET 200 GeV divides the total sample into a control (120<MET<200 GeV) and a signal (MET > 200 GeV) region. The first is used to asses the validity of our background estimation method, the latter is used to search for any hint of new physics.
	Table 6: Summary of background predictions and observed yields for search regions: 120<MET<200 GeV, Njets ≤ 2 and Nbjets = 0, and MET >200 GeV, with and without 3rd lepton veto applied (including taus). Uncertainties include statistical and systematic contributions. This table has been revised since its original release to reflect a correction in the fake background estimation.The limits are unaffected by this correction.

Search in the WZ/ZZ + MET final state with two leptons and two jets

Tables and Figures	Abbreviated Caption
	Table 7: Summary of results in the Z + dijet analysis. The total background is the sum of the Z + jets background predicted from the MET templates method (Z + jets bkg), the same-flavor background predicted from eμ events, and the WZ, ZZ, and rare SM backgrounds predicted from MC. All uncertainties include both the statistical and systematic components.
	Figure 12: Results of the Z + dijet analysis. The observed MET distribution (black points) is compared with the sum of the predicted MET distributions from Z + jets, flavor-symmetric (FS), sum of WZ and ZZ (WZ+ZZ), and rare SM backgrounds. The ratio of observed to predicted yields in each bin is indicated. The error bars indicate the statistical uncertainty in the data and the shaded band indicates the total background uncertainty.

Search with non-resonant opposite-sign dileptons

Tables and Figures	Abbreviated Caption
	Figure 13a: The MCT⊥ distributions in data (black points) and background prediction (filled histogram) for same-flavor dileptons. The background estimate is normalized in the 5 to 100 GeV region. The number of observed events in the MCT⊥ > 100 GeV signal region is consistent with the data-driven background estimate. The large event yield in the lowest bin is an expected feature of the MCT⊥ variable, as a large region of phase space gives MCT⊥ = 0.
	Figure 13b: The MCT⊥ distributions in data (black points) and background prediction (filled histogram) for opposite (different)-flavor dileptons. The background estimate is normalized in the 5 to 100 GeV region. The number of observed events in the MCT⊥ > 100 GeV signal region is consistent with the data-driven background estimate. The large event yield in the lowest bin is an expected feature of the MCT⊥ variable, as a large region of phase space gives MCT⊥ = 0.
	Table 8: Results from shape fit to data in the low MCT⊥ control region and extrapolation to the high MCT⊥ signal region in both channels. Since the contributions of the individual backgrounds in the low MCT⊥ region are derived from a fit to data, they by definition add to the number of events in that region in data.

Interpretations of the searches

The lower plot in each case shows the same information in an alternate style

Tables and Figures	Abbreviated Caption
png version png version	Figure 14: The shading in the mχ0 versus mχ0 (= mχ± ) plane indicates the 95% CL upper limit on the chargino-neutralino production NLO cross section times branching fraction in the flavor-democratic scenario, for the three-lepton search. The contours bound the mass regions excluded at 95% CL for a branching fraction of 50%, as appropriate for the visible decay products in this scenario. The contours based on the observations are shown; in addition, the expected bound is shown.
png version png version	Figure 15a: The shading in the mχ0 versus mχ0 (= mχ± ) plane indicates the 95% CL upper limit on the chargino-neutralino production NLO cross section times branching fraction in the flavor-democratic scenario with x_slep = 0.05, for the three-lepton search. The contours bound the mass regions excluded at 95% CL for a branching fraction of 50%, as appropriate for the visible decay products in this scenario. The contours based on the observations are shown; in addition, the expected bound is shown. Red contours demonstrate separate mass exclusions for the three-lepton search and the same-sign dilepton search alone.
png version png version	Figure 15b: The plot corresponding to Figure 15a, but for x_slep = 0.95.
png version png version	Figure 16a: The exclusion contours for τ-enriched scenario corresponding to results in Figure 14 for x_slep = 0.05
png version png version	Figure 16b: The exclusion contours for τ-enriched scenario corresponding to results in Figure 14 for x_slep = 0.5
png version png version	Figure 16c: The exclusion contours for τ-enriched scenario corresponding to results in Figure 14 for x_slep = 0.95
png version png version	Figure 17: The exclusion contours for τ-dominated scenario corresponding to results in Figure 14.
png version png version	Figure 18: Interpretation of the dilepton WZ + MET and the three-lepton results. The dilepton observed, three-lepton observed, their combination, and combined expected contours are shown.
png version	Figure 19: Interpretation of the Z + dijet and four-lepton analyses in the GMSB model dis- cussed in the text. The observed and expected cross section upper limits are indicated as a function of the mass parameter μ, and are compared to the theory cross section. The region μ < 370 GeV is excluded at 95% confidence level.
png version png version	Figure 20: Limits on chargino pair production cross section times branching ratio for charginos decaying as in Fig. 3. These limits are set by the MCT⊥ based analysis using both the opposite and same flavor channels.
png version png version	Figure 21: Limits on slepton pair production cross section times branching ratio for sleptons decaying as in Fig. 3. These limits are set by the MCT⊥ based analysis using only the same flavor channel.

Summary

Tables and Figures	Abbreviated Caption
png version	Figure 22: A summary of results from four of the scenarios we investigated. Where applicable, the x value used to calculate the slepton mass is 0.5.

Additional Material

Search in the three-lepton final state

Tables and Figures	Abbreviated Caption
	Data-Monte Carlo comparison of the MET distribution in an on-Z, MT > 20 GeV, MET > 30 GeV control region
	Data-Monte Carlo comparison of the MT distribution in an on-Z, MT > 20 GeV, MET > 30 GeV control region
	Dilepton invariant mass distribution in a control region for llmu events
	Dilepton invariant mass distribution in a control region for lle events
	MT versus Mll for the trilepton data events with an OSSF dilepton pair and a hadronically-decaying tau.
	Figure 9b: Observed MET distribution for the data events with an opposite-sign same-flavor dilepton pair and with one hadronically reconstructed tau-lepton. SM expectations are also shown. ”Rare” refers to standard model processes such as ZZ, ttZ, ttW, and triboson production.
	Table 3b: Three lepton search results with an opposite-sign same flavor pair and one hadronically decaying tau.
	Observed yields and predicted backgrounds for a tri-lepton with an opposite sign same flavour lepton pair present as a function of transverse mass for MET > 50 GeV: M(ll) > 105 GeV.
	Observed yields and predicted backgrounds for a tri-lepton with an opposite sign same flavour lepton pair present as a function of transverse mass for MET > 50 GeV: M(ll) < 75 GeV.
	Observed yields and predicted backgrounds for a tri-lepton with an opposite sign same flavour lepton pair present as a function of transverse mass for MET > 50 GeV: 75 GeV < M(ll) < 105 GeV.
	Observed yields and predicted backgrounds for a tri-lepton without an opposite sign same flavour lepton pair present as a function of transverse mass for MET > 50 GeV.
	Observed yields and predicted backgrounds for a tri-lepton with a same sign di-lepton and a hadronically decaying tau as a function of transverse mass for MET > 50 GeV.
	Estimated from Monte Carlo backgrounds for tri-lepton channel with an opposite sign same flavor pair on Z, MET > 50 GeV.
	Estimated from Monte Carlo backgrounds for tri-lepton channel with an opposite sign same flavor pair off Z, MET > 50 GeV.
	Estimated from Monte Carlo backgrounds for tri-lepton channel without an opposite sign same flavor pair, MET > 50 GeV.
	Estimated from Monte Carlo backgrounds for tri-lepton channel with a same sign di-lepton and a hadronically decaying tau, MET > 50 GeV.
	Result of a Monte Carlo closure test for a method to estimate backgrounds from non-prompt or misidentified light leptons: comparison of the MC truth observed yield and MC predicted yield as a function of transverse mass.
	Comparison of MET distributions for internal and external conversions in data.
	Response of u1 component of the hadronic recoil in data and MC.
	Response of u2 component of the hadronic recoil in data and MC.
	Resolution of u1 component of the hadronic recoil in data and MC.
	Resolution of u2 component of the hadronic recoil in data and MC.

Search in the four-lepton final state

Tables and Figures	Abbreviated Caption
	Validation of the ZZ Monte Carlo simulation in the low MET region

Search in the same-sign two-lepton final state

Tables and Figures	Abbreviated Caption
	Cross-check of background prediction method in the baseline region.
	Cross-check of background prediction method in the baseline region.
	Data with at least one identified b-jet.
	Data with no identified b-jets.
	Fake ratios for electrons for the SS analysis.
	Fake ratios for electrons for the SS analysis.
	Fake ratios for muons for the SS analysis.
	Fake ratios for muons for the SS analysis.
	Same Sign predicted and observed yields in the two search regions with and without the search region. This figure has been revised since its original release to reflect a correction in the fake background estimation.The limits are unaffected by this correction
	Validation of WZ simulation in a WZ-enriched control region. This region has the Z veto inverted and MET > 120 GeV.

Search in the WZ/ZZ + MET final state with two leptons and two jets

Tables and Figures	Abbreviated Caption
	Dilepton invariant mass distribution for events in the preselection region
	Event yields for the preselection region
	Dilepton invariant mass after preselection and dijet requirement
	Event yields after preselection and dijet requirements
	Data-Monte Carlo comparison in dijet invariant mass
	Results in the ee channel
	Results in the mumu channel

Most important channels in limit setting

	Flavor-democratic x_slep = 0.05
	Flavor-democratic x_slep = 0.5
	Flavor-democratic x_slep = 0.95
	Tau-enriched x_slep = 0.05
	Tau-enriched x_slep = 0.5
	Tau-enriched x_slep = 0.95
	Tau-dominated x_slep = 0.5
	Decays with a W and a Z boson

-- FedorRatnikov - 10-Nov-2012