ATLAS NOTE - CERNcds.cern.ch/record/2139641/files/ATLAS-CONF-2016-007.pdf · 2016. 3. 16. · The...

ATL

AS-

CO

NF-

2016

-007

16M

arch

2016

ATLAS NOTEATLAS-CONF-2016-007

March 15, 2016

Search for top squarks in final states with one isolated lepton, jets,and missing transverse momentum in

√s = 13 TeV pp collisions of

ATLAS data

The ATLAS Collaboration

Abstract

A search for the stop, the supersymmetric partner of the top quark, is conducted in final stateswith one isolated electron or muon, jets, and missing transverse momentum using the 2015LHC pp collision data at a centre-of-mass energy of

√s = 13 TeV recorded by the ATLAS

detector and corresponding to an integrated luminosity of 3.2 fb−1. The analysis targetstwo types of signal models: gluino-mediated pair production of stops with a nearly massdegenerate stop and neutralino; and direct pair production of stops, decaying to the top quarkand the lightest neutralino. The experimental signature in both signal scenarios is similar tothat of a top quark pair produced in association with large missing transverse momentum, tt̄+EmissT . No significant excess over the Standard Model prediction is observed, and exclusionlimits on gluino and stop masses are set at 95 % CL. The results extend the LHC Run 1exclusion limit on the gluino mass up to 1460 GeV in the gluino-mediated scenario in thehigh gluino and low stop mass region, and for the direct stop model add an excluded stopmass region from 745 to 780 GeV for a massless lightest neutralino.

c© 2016 CERN for the benefit of the ATLAS Collaboration.Reproduction of this article or parts of it is allowed as specified in the CC-BY-4.0 license.

1 Introduction

Supersymmetry (SUSY) [1–6] is a natural solution [7, 8] to the hierarchy problem [9–12]. The superpart-ner of the top quark t̃ (top squark or stop) is expected to be relatively light due to its large contribution tothe Higgs boson mass radiative corrections [13, 14]. For reasons such as gauge unification [15] and the2-loop radiative corrections to the Higgs boson mass [16, 17], one may also expect a TeV scale for themass of the gluino, the superpartner of the gluon. A common theoretical strategy for avoiding strong con-straints from the non-observation of proton decay [18] is to introduce a multiplicative quantum numbercalled R-parity. If R-parity is conserved [19] SUSY particles are produced in pairs and the lightest super-symmetric particle (LSP) is stable. This note follows the typical assumption that the lightest neutralino1( χ̃01) is the LSP. Since the χ̃

01 interacts only weakly it can serve as a candidate for dark matter [20, 21].

This note presents a search targeting the lighter stop2 (t̃1) in two signal scenarios: gluino-mediated pairproduction of the t̃1 with a small t̃1-LSP mass splitting, and direct pair production of the t̃1, both illustratedby the diagrams in Fig. 1. The former scenario refers to pair production of gluinos, each decaying to thetop quark and the t̃1. In this scenario, the mass difference between the gluino and the t̃1 is assumed tobe well above the top quark mass, while the mass difference between the t̃1 and the LSP is assumedto be significantly smaller than the W boson mass. As a result, the visible t̃1 decay products have lowmomentum, typically below the reconstruction and identification thresholds. This scenario is motivatedby the dark matter relic density, which is generally too large in the Minimal Supersymmetric StandardModel [22, 23] but can be regulated by co-annihilation of the stop and the neutralino [24–26]. In thesecond scenario, the two directly produced t̃1 are each assumed to decay to the top quark and the LSP.This model is interesting as it is independent of the gluino mass, which is more weakly constrained bynaturalness arguments than the stop mass.

Experimentally, the final states of the two scenarios are similar, and the detector signature consists ofthe decay products of a pair of top quarks3 and large missing transverse momentum (~pmissT , where themagnitude is referred to as EmissT ) from the two LSPs: tt̄ + E

missT . The main difference between the two

scenarios is that the production cross-section for gluino pairs is about a factor 50 higher than for t̃1 pairsof the same mass. The results are also re-interpreted in a model of pair produced vector-like top quarksT (referred to as VLQ) [27–29], for which the decay mode T → tZ with Z → νν̄ has a phenomenologysimilar to that of direct stop pair production with t̃1 → t χ̃01.

The analysis presented here – which is based on previous ATLAS searches for the same signature [30, 31]– targets the one-lepton final state where the W boson from one of the top quarks decays to an electronor muon (either directly or via a τ) and the W boson from the other top quark decays hadronically. Thedominant background processes are: the production of tt̄ and the associated production of a top quark anda W boson (single top Wt); tt̄ + Z (→ νν̄); and the associated production of W bosons and jets (W+jets).The search uses the ATLAS data collected in proton-proton (pp) collisions in 2015 corresponding to anintegrated luminosity of 3.2 fb−1 at a centre-of-mass energy of

√s = 13 TeV. The ATLAS Run 1 searches

for gluino-mediated stop production and direct stop pair production are summarized in Ref. [32] and [33],

1 The charginos χ̃±1,2 and neutralinos χ̃01,2,3,4 are the mass eigenstates formed from the linear superposition of the charged and

neutral SUSY partners of the Higgs and electroweak gauge bosons (higgsinos, winos and binos).2 The superpartners of the left- and right-handed top quarks, t̃L and t̃R, mix to form the two mass eigenstates t̃1 and t̃2, where

t̃1 is the lighter one.3 Due to the Majorana nature of the gluino, in the gluino-mediated model, each of the two ‘visible’ top quarks can independently

be a top or an anti-top quark. Hereafter, the term tt̄ can be taken to refer to any combination of t and t̄.

2

t̃

t̃p

p

χ̃01

t

χ̃01

t

g̃

g̃

t̃

t̃

p

p

t

χ̃01

soft

t

χ̃01

soft

Figure 1: Diagrams illustrating the two considered signal scenarios. Left: stop pair production, where each stopdecays to the top quark and the lightest neutralino ( χ̃01). Right: gluino-mediated stop pair production, where eachstop decays to a low momentum (‘soft’) charm quark and the lightest neutralino.

respectively. The CMS collaboration has performed similar searches in Run 1 for gluino-mediated stopproduction [34] and direct stop pair production [35–39].

This document is organized as follows. The ATLAS detector, dataset and trigger are described in Section 2and the corresponding set of simulations are detailed in Section 3. Section 4 presents an overview of theassignment of detector-level measurements to physical objects and the construction of discriminatingvariables. These variables are used in Section 5 to construct the signal event selections. The backgroundestimation procedure (Section 6) and systematic uncertainties (Section 7) are described before the resultsare presented in Section 8. Section 9 contains concluding remarks.

2 ATLAS detector and dataset

The ATLAS detector [40] is a multi purpose particle physics detector with nearly 4π coverage in solidangle around the collision point.4 It consists of an inner tracking detector (ID), surrounded by a supercon-ducting solenoid providing a 2 T axial magnetic field, a system of calorimeters, and a muon spectrometer(MS) incorporating three large superconducting toroid magnets. The ID provides charged-particle track-ing in the range |η | < 2.5 using three technologies: silicon pixel and silicon microstrip tracking detectors,and a transition radiation tracker. During the LHC shutdown between Run 1 and Run 2, a new innermostlayer of silicon pixels has been added, which improves the track impact parameter resolution and vertexposition resolution performance [41]. High-granularity electromagnetic and hadronic calorimeters coverthe region |η | < 4.9. The central hadronic calorimeter is a sampling calorimeter with scintillator tilesas the active medium and steel absorbers. All the electromagnetic calorimeters, as well as the endcapand forward hadronic calorimeters, are sampling calorimeters with liquid argon as the active medium andlead, copper, or tungsten absorber. The MS consists of three layers of high-precision tracking chambers

4 ATLAS uses a right-handed coordinate system with its origin at the nominal interaction point (IP) in the centre of the detectorand the z-axis along the beam pipe. The x-axis points from the IP to the centre of the LHC ring, and the y-axis pointsupwards. Cylindrical coordinates (r, φ) are used in the transverse plane, φ being the azimuthal angle around the z-axis.The pseudorapidity is defined in terms of the polar angle θ as η = − ln tan(θ/2). Angular distance is measured in units of∆R ≡

√(∆η)2 + (∆φ)2.

3

with coverage up to |η | = 2.7 and dedicated chambers for triggering in the region |η | < 2.4. Events areselected by a two-level trigger system, the first level is a hardware-based system, while the second is asoftware-based system.

The 2015 LHC collision data used in this analysis has a mean number of additional pp interactions perbunch crossing (pileup) of approximately 14, and a bunch spacing of 25 ns. Following requirements basedon beam and detector conditions and data quality, the dataset corresponds to an integrated luminosity of3.2 fb−1 with an associated uncertainty of 5%. The uncertainty is derived following the same methodologyas that detailed in Ref. [42]. Events used for this search were recorded using a trigger logic that acceptsevents with EmissT , calibrated to the electromagnetic scale, above 70 GeV. The trigger is over 95% efficientfor events passing an offline computed EmissT > 200 GeV requirement and is > 99% efficient for eventspassing all signal selections. An additional data sample used to estimate one of the background processeswas recorded with a trigger requiring a photon with transverse momentum pT > 120 GeV.

3 Monte Carlo Simulations

Samples of Monte Carlo (MC) simulated events are used for the description of the background and tomodel the SUSY signals. Several matrix element (ME) generators are combined with parton shower (PS)generators. Signal SUSY samples are generated at leading order (LO) with MG5_aMC 2 [43] while VLQsignal samples are generated at LO with Protos v2.2 [44, 45]. All signal samples are interfaced withPythia 8.186 [46]. Background samples use one of three setups:

• MG5_aMC 2 interfaced with Pythia 8 [46] or Herwig++ using the CKKW-L [47] or the MC@NLOmethod for matching a LO or next-to-leading-order (NLO) ME to the PS, respectively.

• Powheg-Box [48–52] interfaced to Pythia 6 [53] or Herwig++ using the Powheg method [54, 55]for matching the NLO ME to the PS.

• Sherpa 2.1.1 [56] using Comix [57] and OpenLoops [58] ME generators interfaced with the Sherpaparton shower [59].

The CT10 [60] parton distribution function (PDF) is used for ME calculations with Sherpa and Powheg-Box and the NNPDF2.3 [61] PDF is used for samples generated with MG5_aMC, except for the NLOsamples which use either CT10 or NNPDF3.0 [62]. The CTEQ6L1 [63] LO PDF along with the P2012 [64]set of underlying event tuned parameters (UE tune) is used for Pythia 6, the NNPDF2.3 LO PDF and theA14 UE tune [65] is used for Pythia 8, and the CT10 PDF with the default author UE tune is used forthe Sherpa samples. The samples produced with MG5_aMC, Powheg-Box and Protos all use EvtGenv1.2.0 [66] for the modelling of b-hadron decays. The simulation setup is summarized in Table 1 andmore details can be found in Ref. [67–70] for tt̄ and single top, W/Z+jets, dibosons, and tt̄ + W/Z , re-spectively. Additional samples aside from those shown in Table 1 are used to assess theoretical modelinguncertainties and will be discussed in Section 7.

In the gluino-mediated production the stop is assumed to decay via t̃1 → c + χ̃01 with a 100% branchingratio and with a default mass splitting mt̃1 − m χ̃01 = 5 GeV. Alternative samples with larger mass splittingand/or replacing the two-body stop decay by a four-body stop decay t̃1 → b f f ′ χ̃01, where f f ′ is afermion-antifermion pair, are produced for additional studies. The gluinos and stops are assumed todecay promptly. In the direct stop pair production samples, the t̃1 is chosen to be mostly the partner of

4

Process ME Generator ME Fragmentation UE Cross-sectionPDF Tune Order

tt̄ Powheg-Box v2 CT10 Pythia 6 P2012 NNLO+NNLL [71–76]Single top Powheg-Box CT10 Pythia 6 P2012 NNLO+NNLL [77–79]W/Z+jets Sherpa 2.1.1 CT10 Sherpa Default NNLO [80]Diboson Sherpa 2.1.1 CT10 Sherpa Default NLOtt̄ + W/Z MG5_aMC 2 NNPDF2.3 Pythia 8 A14 NLO [43]tt̄ + γ MG5_aMC 2 CTEQ6L1 Pythia 8 A14 NLO [43]SUSY Signal MG5_aMC 2 NNPDF2.3 Pythia 8 A14 NLO+NLL [81]VLQ Signal Protos v2.2 NNPDF2.3 Pythia 8 A14 NNLO+NNLL

Table 1: Overview of the nominal simulated samples.

the right-handed top quark5 and the χ̃01 to be a pure bino. This choice is consistent with a large branchingratio for the given t̃1 decay. Different hypotheses for the left/right mixing in the stop sector and the natureof the neutralino lead to different acceptance values. The acceptance is affected because the polarizationof the top quark changes as a function of the field content of the supersymmetric particles, which impactsthe boost of the lepton in the top quark decay. Signal grids are generated for both the gluino and stoppair production models. The grid spacing on the gluino/stop and stop/neutralino masses varies between25 and 100 GeV.

All the MC samples are normalized to the highest order (in αs) cross-section available, as indicated inthe last column of Table 1. The cross-sections for the pair and single production of top quarks as wellas for the signal processes also include resummation of soft gluon emission to next-to-next-to-leadinglogarithmic NNLL and next-to-leading logarithmic NLL accuracy, respectively. As will be describedin Section 6.1.3, it is important that the simulated tt̄ + γ and tt̄ + Z events are as similar as possible.Therefore, a small 4% correction is applied to the tt̄ + γ cross-section to account for a different PDF,factorization/renormalization scale, and the number of partons from the matrix element.6 The same NLOQCD k-factor is then applied to the tt̄ +γ process as is used for the tt̄ + Z (→ νν̄) process [43]. This choiceis motivated by the similarity of QCD calculations for the two processes as well as empirical studies ofthe ratio of k-factors computed as a function of the boson pT. Further information about the k-factor andits uncertainty is given in Section 7. The cross-sections for the tt̄, W+jets, and Wt processes are used forcross checks and optimization studies while for the final results these processes are normalized to data incontrol regions.

All background samples, except for the tt̄+γ sample, are processed with the full simulation of the ATLASdetector [84] based on Geant 4 [85]. The signal samples and the tt̄ + γ sample are processed with a fastsimulation [86] of the ATLAS detector with parameterized showers in the calorimeters. All samplesare produced with varying numbers of simulated minimum-bias interactions generated with Pythia 8overlaid on the hard-scattering event to account for pileup from multiple pp interactions in the same ornearby bunch crossings. The average number of interactions per bunch crossing is reweighted to match the

5 The t̃R component is given by the the off-diagonal entry of the stop mixing matrix. The t̃1 decays in the direct stop pairproduction samples are performed by Pythia and produce unpolarized top quarks. The events are reweighted to obtain a stopmixing equivalent to a matrix with (off-) diagonal entries of approximately (±0.83) 0.55. The event weights depend on theangular distributions of the top decay products [82].

6 The tt̄ + γ sample uses a fixed factorization/renormalization scale of 2 × mtop with no extra partons in the ME, whereas tt̄ + Zis generated with up to two partons. The top decay is performed in MG5_aMC for tt̄ + γ to account for hard photon radiationfrom the top decay products that is a ∼ 15% effect for pγT ∼ 120 GeV [83].

5

distribution in data. Furthermore, the simulated samples are reweighted to account for small differences inthe performance of the reconstruction and identification of physics objects with respect to those measuredin data [87].

4 Event Reconstruction and Selection

All events must pass a series of quality criteria before being considered for further use. The reconstructedprimary vertex with the highest

∑tracks p2T must have at least two associated tracks. In this analysis, physics

objects are labeled as either baseline or signal depending on various quality and kinematic requirements,where the latter describes a tighter selection of the former. Baseline objects are used to distinguishbetween the physics objects in the event and to compute the missing transverse momentum. Baselineleptons are also used to apply a second-lepton veto to suppress dilepton tt̄ and Wt events.

Electron candidates are reconstructed from electromagnetic calorimeter cell clusters that are matched toID tracks. Baseline electrons are required to have pT > 7 GeV, |η | < 2.47, and pass ‘VeryLoose’ like-lihood identification criteria which are defined following the same methodology described in Ref. [88].Signal electrons must pass all baseline requirements and in addition have pT > 25 GeV, satisfy the ‘Loose’likelihood identification criteria of Ref. [88], and have impact parameters along the beam direction (z0)and in the transverse plane (d0) that satisfy |z0 sin θ | < 0.5 mm and |d0/σdo | < 5, where σdo is theuncertainty of d0. Furthermore, signal electrons must be isolated, where the criteria use track-based in-formation to obtain a flat 99% efficiency as derived from Z to two leptons MC samples and confirmed indata.

Muons are reconstructed from combined tracks that are formed from the ID and MS, ID tracks matchedto muon segments, stand-alone MS tracks, and ID tracks matched to an energy deposit in the calorimetercompatible with a minimum ionizing particle (referred to as calo-tagged muon) [89]. Baseline muons arerequired to have pT > 6 GeV, |η | < 2.7, and pass the ‘Loose’ identification criteria described in Ref. [89].Signal muons must pass all baseline requirements and in addition have pT > 25 GeV, and have impactparameters |z0 sin θ | < 0.5 mm and |d0/σdo | < 3. Furthermore, signal muons must be isolated accordingto isolation criteria similar to those used for signal electrons, yielding the same efficiency.

Photons are not used in the main event selection and give rise to extra jet or electron candidates, dependingon their properties and whether they pass the other identification criteria. Photons must be identified,however, for the tt̄ + γ sample that is used in the data-driven estimation of the tt̄ + Z background. Inthis case, photon candidates are reconstructed from calorimeter cell clusters and are required to pass the‘Tight’ identification criteria described in Ref. [90]. Furthermore, photons are required to have pT >125 GeV and |η | < 2.37 so that the photon trigger is fully efficient. Photons must further pass isolationcriteria based on both track and calorimeter information.

Jet candidates are built from topological clusters [91, 92] in the calorimeters using the anti-kt algorithmwith a jet radius parameter R = 0.4 [93]. Jets are corrected for contamination from pileup using the jetareas method [94–96] and then calibrated to account for the residual detector response [97, 98]. Jets indata are further calibrated based on in situ measurements of the jet energy scale. Baseline jets are requiredto have pT > 20 GeV. Signal jets must have pT > 25 GeV and reside within |η | < 2.5. Furthermore, signaljets with pT < 50 GeV are required to pass criteria, implemented in the jet vertex tagger algorithm [96],designed to reject jets originating from pileup. Events containing a jet that does not pass specific jetquality requirements are vetoed from the analysis in order to suppress detector noise and non-collision

6

Object 1 e e µ l γ γ τ

Object 2 µ j j j j e e

∆R < 0.01 0.2 0.2 min(0.4,0.04 + 10

plT

)0.2 0.1 0.1

Condition calo-tagged µ j not b-tagged j not b-tagged and – – – –

n jtrack < 3 orpµT

pjT

> 0.7

Resolution e e µ j γ e e

Table 2: Overlap removal procedure. The first two rows list the types of overlapping objects: electrons (e), muons(µ), electron or muon (l), jets ( j), photons (γ), and hadronically decaying τ (τ). All objects refer to the baselinedefinitions, except for γ and τ where no distinction between baseline and signal definition is made. The third rowspecifies when an object pair is considered as overlapping, the fourth row describes an optional condition, and thelast row lists which label is given to the ambiguous object. All units are in GeV. See text for more information.

backgrounds [99, 100]. Jets resulting from b-quarks (called b-jets) are identified using the MV2c20b-tagging algorithm, which is based on quantities such as impact parameters of associated tracks andreconstructed secondary vertices [101–103]. This algorithm is used at a working point that provides77% b-tagging efficiency in simulated tt̄ events. The choice of the working point was optimized forthis analysis and corresponds to a rejection factor of about 140 for light-quark flavours and gluons andabout 5 for charm jets. Jets and associated tracks are also used to identify hadronically decaying τ usingthe ‘Loose’ identification criteria described in Ref. [104, 105] which has a 60% and 50% efficiency forreconstructing τ decaying into one and three charged pions, respectively. These τ candidates are requiredto have one or three associated tracks, pT > 20 GeV and |η | < 2.5.

The missing transverse momentum vector is reconstructed from the negative sum of the transverse mo-menta of baseline electrons, muons, jets, and a soft-term built from tracks not associated to other recon-structed objects [106, 107]. For the event selections requiring photons, the calibrated photon is directlyincluded in the EmissT calculation. In all other cases, photons and hadronically decaying τ are not explicitlyincluded but enter as jets, electrons or via the soft-term.

To avoid labelling the same detector signature as more than one object, an overlap removal procedure isapplied. The procedure was tailored for this analysis and optimized using simulation. Table 2 summarizesthe procedure. Given the sets of baseline objects, the procedure checks for overlap based on the minimaldistance ∆R between pairs of objects. For example, if a baseline electron and a baseline jet are foundwith ∆R < 0.2, then the electron is preserved (‘resolution’) and the jet is discarded, unless the jet isb-tagged (‘condition’) in which case the electron is assumed to stem from a heavy-flavour decay and ishence discarded while the jet is preserved. The order of the procedure is given by the columns in Table 2which are executed from left to right. The steps involving a photon are not applied in the main eventselection, but only for the event selection where photons are identified. For the remainder of the note, allbaseline and signal objects are those that have survived the overlap removal procedure.

Large-radius jets are clustered from all signal (small-radius) jets using the anti-kt algorithm with R = 1.0or 1.2, and are groomed using re-clustered jet trimming with a pT fraction of 5% [108–111]. Electronsand muons are not included in the re-clustering, since it was found that including them increases thebackground acceptance more than the signal efficiency. Large-radius jets are not used in the overlapremoval procedure; however, the signal jets that enter the re-clustering have passed the overlap removal

7

procedure described above. The analysis also uses a large-radius jet mass, where the squared mass isdefined as the square of the four-vector sum of the constituent (small-radius) jets.

All events are required to have EmissT > 200 GeV, exactly one signal lepton and no additional baselineleptons, at least four signal jets, have a transverse mass7 of the signal lepton and the missing transversemomentum satisfying mT > 30 GeV, and have an azimuthal angle between leading or sub-leading jet andthe missing transverse momentum of |∆φ(jeti , ~pmissT ) | > 0.4 with i ∈ {1,2}. The H

missT,sig variable, based on

the identified lepton, jets and the per-event jet energy uncertainties, is also used (more details in Ref. [30,112]). The latter three criteria suppress multijet processes with mis-identified or non-prompt leptons andmis-measured EmissT to a negligible level. With the above event selection, the dominant backgrounds arett̄ events with at least one leptonically decaying W boson, and W+jets production. A powerful techniquefor suppressing these background processes is to require mT to be greater than the W boson mass.

Of the residual backgrounds, one of the dominant contributions is tt̄ production where both W bosonsdecay leptonically, or one W boson decays leptonically and the other via a hadronic τ. A series ofadditional variables, described in detail in Ref. [30], are used to discriminate between this background andthe signal processes. The mχtop variable is the invariant mass of the three jets in the event most compatiblewith coming from a hadronically decaying top quark, and where the three jets are selected by minimizinga χ2-distribution including the jet momenta and energy resolutions. The asymmetric mT2 (amT2) [113–116] and mτT2 are both variants of the variable mT2 [117], a generalization of the transverse mass appliedto signatures where two particles are not directly detected. The amT2 variable targets dileptonic tt̄ eventswhere one lepton is not reconstructed while the mτT2 variable targets tt̄ events where one of the two Wbosons decays via a hadronically decaying τ. The topness [118] variable is based on minimizing a χ2-typefunction quantifying the compatibility with a dileptonic tt̄ event where one lepton is not reconstructed.Furthermore, the mass of large-radius jets is useful when the boost of the top quark is significant.

An important change from the Run 1 suite of tools is the treatment of hadronically decaying τ candidatesin the mτT2 variable. Events are removed if one of the selected jets is additionally identified as a hadronicτ candidate, and the corresponding mτ

T2 < 80 GeV, where mτT2 uses the signal lepton and hadronic τ

candidate as the two visible objects. This τ veto removes approximately 40% of simulated tt̄ eventswhere at truth-level one W boson decays leptonically and the other proceeds via a hadronically decayingτ, for an event selection with a EmissT > 200 GeV requirement. For the considered signal models, theveto removes 1% of the events and has a negligible impact. The τ veto is applied in all following eventselections except those defining the tt̄ + Z control region (since the veto would remove only about 1% ofthe events in this region).

5 Signal Regions

Three signal event selections (called signal regions, or SR1–3) are constructed using the set of discrimi-nating variables described in the previous Section 4. The three signal regions are optimized before lookingat the data to maximize the discovery sensitivity using three benchmark signal models from the gluino-mediated stop models, each representing distinct phenomenology. The benchmark models for SR1 andSR2 are chosen to have production cross-sections and kinematic properties similar to those of direct stoppair production models near the edge of the Run 1 exclusion limit [30, 119], while the benchmark model

7 The transverse mass, mT, is defined as m2T = 2plepT E

missT (1 − cos(∆φ)), where ∆φ is the azimuthal angle between the lepton

and the missing transverse momentum direction. The quantity plepT is the transverse momentum of the charged lepton.

8

for SR3 cannot be mapped in these terms to any direct stop model. The SR1–3 target signal models withincreasing gluino masses, ranging from 1.1 TeV to 1.4 TeV, respectively, and t̃1 masses between 600 and800 GeV for SR1–2.

The three signal regions are characterized by increasing EmissT requirements. The SR1 benchmark has thesoftest EmissT spectrum and the available momentum of the hadronically decaying top quark is typicallynot sufficient to capture all of the decay products into a single large-radius jet. As a result, the resolved topreconstruction mχtop is useful for rejecting dileptonic tt̄ and other background events without a resonanttop quark with hadronic decay products. In contrast, the boost of the hadronically decaying top quarksin the SR2 and SR3 benchmarks is often sufficient to capture all decay products inside a single large-radius jet. The angular separation between the decay products scales with the inverse of the momentum.Therefore, the optimal large-radius jet cone size is found to be larger for SR2 (R = 1.2) than for SR3(R = 1.0). Additional requirements on topness and amT2 further reduce the dileptonic tt̄ background.Events without a high pT top quark that decays leptonically are suppressed by using a requirement onthe ∆R between the highest pT b-jet and the signal lepton. The regions have additional requirements onmT and HmissT,sig to further exploit the large genuine E

missT from undetected neutralinos. A requirement of at

least one b-tagged jet amongst the four highest pT jets is used in all SR1–3 in order to reduce the W+jetsand diboson backgrounds.

The signal region definitions are summarized in Table 3 and the expected yields in all three regions areshown in Table 4. The signal regions are not mutually exclusive.

6 Background Estimates

The dominant background processes are the production of tt̄ and single top Wt events where both Wbosons decay leptonically (one of which is ‘lost’, meaning it is either not reconstructed, not identified,or removed by the overlap removal procedure) or one W boson decays leptonically and the other via ahadronically decaying τ; tt̄ + Z (→ νν̄); and W+jets. Other background processes considered are semi-leptonic tt̄, dibosons, tt̄ + W , Z+jets and multijet events.

The main background processes are estimated by isolating each of them in a dedicated control region(CR), described in Section 6.1, normalizing simulation to match data in a simultaneous fit. This fit isperformed separately for each SR with the associated CRs. The accuracy of the fits is tested in a seriesof validation regions (VR), discussed in Section 6.2. Figure 2 schematically illustrates the setup for oneexample SR and its associated CRs and VRs. The CRs for Wt and tt̄ + Z are new with respect to the Run1 analysis.

The multijet background is estimated from data using a fake factor method [120]. The contribution isfound to be negligible. All other (small) backgrounds are determined entirely from simulation, normal-ized to the most accurate theoretical cross-sections available. The Z+jets background is found to benegligible.

6.1 Control Regions

A series of control regions are defined as event selections that are kinematically close to the signal regionsbut with a few key variable thresholds inverted to significantly reduce signal contamination and enhance

9

30 90 120100

100

175200

mT [GeV]

amT2

[GeV

]STCR SR

WCR WVR WVR-tail

TCR TVR

Nb-jets = 0

Nb-jets ≥ 1

Nb-jets ≥ 2

0TZCR

100

Ng ≥ 1

Figure 2: A schematic diagram for the various event selections used to estimate and validate the background modeland then search for stop production. Solid lines indicate kinematic boundaries while dashed lines indicate that theevents can extend beyond the boundary. CR, VR, and SR stand for control region, validation region, and signalregion, respectively. T, ST, TZ, and W stand for tt̄, single top, tt̄ + Z , and W+jets, respectively.

the yield and purity of a particular background. These control regions are then used to constrain the back-ground normalization. Each of the three signal regions has a dedicated control region for the tt̄ (TCR),for the W+jets (WCR), for the single top (STCR) and for the tt̄ + W/Z (TZCR) background processes.The general strategy in constructing the control regions is to invert the transverse mass requirement froma high threshold to a low window. The requirements on several variables are loosened to increase thestatistical power of the CR. The details of the TCR and the WCR are described in Section 6.1.1, while theSTCR and TZCR are documented in Section 6.1.2 and 6.1.3 respectively. Table 3 presents an overviewof the CR selections for the TCR, WCR and STCR corresponding to SR1, SR2, and SR3.

A likelihood fit is performed for each SR involving the SR and the associated CRs [121]. The expectednumber of events in each region is given by the sum over all background processes, and optionally a signalmodel. The normalizations of the tt̄, tt̄ + W/Z , single top, and W+jets backgrounds are controlled by fourfree parameters (normalization factors, NFs) in the fit. The electron and muon channels are always addedtogether. To obtain a set of background predictions that is independent of the observation in the SRs, thefit can be configured to use only the CRs to constrain the fit parameters: the SR bins are removed from thelikelihood and any potential signal contribution is neglected everywhere. This fit configuration is referredto as the background-only fit.

10

Common event selectionTrigger EmissT triggerLepton exactly one signal lepton (e, µ), no additional baseline leptons.Jets at least four signal jets, and |∆φ(jeti , ~pmissT ) | > 0.4 for i ∈ {1,2}.hadronic τ veto events with a hadronic τ and mτ

T2 < 80 GeV.

Variable SR1 TCR1 / WCR1 STCR1≥4 jets with pT > [GeV] (80 50 40 40) (80 50 40 40) (80 50 40 40)EmissT [GeV] > 260 > 200 > 200HmissT,sig > 14 > 5 > 5mT [GeV] > 170 [30,90] [30,120]amT2 [GeV] > 175 [100,200] / > 100 > 200topness > 6.5 > 6.5 > 6.5mχtop [GeV] < 270 < 270 < 270∆R(b, `) < 3.0 – –∆R(b1,b2) – – > 1.2number of b-tags ≥ 1 ≥ 1 / = 0 ≥ 2

SR2 TCR2 / WCR2 STCR2≥4 jets with pT > [GeV] (120 80 50 25) (120 80 50 25) (120 80 50 25)EmissT [GeV] > 350 > 250 > 200HmissT,sig > 20 > 15 > 5mT [GeV] > 200 [30,90] [30,120]amT2 [GeV] > 175 [100,200] / > 100 > 200∆R(b, `) < 2.5 – –∆R(b1,b2) – – > 1.2number of b-tags ≥ 1 ≥ 1 / = 0 ≥ 2leading large-R jet pT [GeV] > 200 > 200 > 200leading large-R jet mass [GeV] > 140 > 140 > 0∆φ(~pmissT ,2

ndlarge-R jet) > 1.0 > 1.0 > 1.0

SR3 TCR3 / WCR3 STCR3≥4 jets with pT > [GeV] (120 80 50 25) (120 80 50 25) (120 80 50 25)EmissT [GeV] > 480 > 280 > 200HmissT,sig > 14 > 8 > 5mT [GeV] > 190 [30,90] [30,120]amT2 [GeV] > 175 [100,200] / > 100 > 200topness [GeV] > 9.5 > 0 > 9.5∆R(b, `) < 2.8 – –∆R(b1,b2) – – > 1.2number of b-tags ≥ 1 ≥ 1 / = 0 ≥ 2leading large-R jet pT [GeV] > 280 > 200 > 200leading large-R jet mass [GeV] > 70 > 70 > 70

Table 3: Overview of the event selections for all SRs and the associated tt̄ (TCR), W+jets (WCR), and Wt (STCR)control regions. Round brackets are used to describe lists of values and square brackets denote intervals.

11

6.1.1 Top and W CRs

The TCRs and WCRs are constructed by modifying the mT selection in the SRs to be a window whoseupper edge is near the W boson mass. An additional upper requirement on amT2 is applied to the TCRsin order to make them orthogonal to the STCRs, described in the next section. Furthermore, some otherkinematic requirements are relaxed or removed to increase the event yields in the CRs. The resultingselections yield 238, 102, and 121 events in TCR1, TCR2, and TCR3, respectively, which are enrichedin semi-leptonic tt̄ events with purities that vary between 75% and 85%. The WCRs are built from theTCRs by changing the b-jet requirement to a b-jet veto, and relaxing the amT2 requirement. The b-jetveto suppresses tt̄ events and results in a W+jets purity of approximately 75% in all three regions. Theselections yield 558, 135, and 352 events in WCR1, WCR2, and WCR3, respectively.

6.1.2 Single Top CR

All of the expected single top contributions in the three SRs are in the Wt channel. This process canevade kinematic bounds from selections targeting the suppression of tt̄. Nonetheless, isolating a puresample of Wt events kinematically close to the SRs is challenging due to the similarity of Wt and tt̄.The Wt events that pass signal region-like event selections often have a second b-jet within acceptance.The amT2 variable is useful for discriminating tt̄ and Wt because the mass of the W b system not fromthe resonant top quark is typically higher than for an on-shell top quark in the phase space selected bythis analysis. Therefore, the STCRs are characterized by amT2 > 200 GeV. Furthermore, to increase thepurity of Wt and reduce the W+jets contamination, events are required to have two b-tagged jets. Topquark pair events often exceed the amT2 kinematic bound when one of the two b-tags used in the amT2calculation is a jet produced from a charm quark from the W decay. When this jet is from the same topquark as the other b-tagged jet then the ∆R between them tends to be smaller than for Wt events thathave two b-jets from b-quarks that are naturally well separated. Therefore, to further increase the Wtpurity, events in the STCRs are required to have ∆R(b1,b2) > 1.2 where b1 and b2 are the two highestpT b-tagged jets. Figure 3 shows distributions of the defining variables for STCR1 with all requirementsapplied but the one plotted. The expected purity for Wt events is approximately 40% in all three STCRs,and the selections yield 62, 71, and 45 events in STCR1, STCR2, and STCR3, respectively.

6.1.3 t t̄ + Z CR

Top quark pair production in association with a Z boson that decays into neutrinos is an irreduciblebackground. The expected contributions of tt̄ + W in the three SRs are less than 10% with respect to theexpected tt̄ + Z yields, and the two processes are lumped together in the analysis. A CR using chargedleptonic Z boson decays is not feasible given the small branching fraction to leptons and the limiteddataset available. However, a data-driven approach is still possible using a similar process: tt̄ +γ. Similartechniques have been used for estimating Z (→ νν̄)+jets from γ+jets [122] and the method was studiedas a VR using the Run 1 data [30]. An event selection is constructed requiring a high pT photon that isthen treated as EmissT in direct analogy to Z → νν̄.

The CR is designed to minimize the differences between the two processes, in order to reduce the theo-retical uncertainties in the extrapolation. The Feynman diagrams in the ME expansion for the productionof tt̄ + Z and tt̄ + γ are identical, except for a negligible contribution from the coupling of the Z bosonto neutrinos that is absent for photons. The main differences arise from the Z boson mass, which reduces

12

with photon added [GeV]missT = Emiss

TE~

0 100 200 300 400 500

Eve

nts

/ 20

GeV

10

20

30

40Data 2Ltt 1Ltt

Single TopW+jets+Vtt

VVTotal SM

PreliminaryATLAS-1 = 13 TeV, 3.2 fbs

STCR1

[GeV]T2am0 100 200 300 400 500

Dat

a / S

M

0.51

1.5

[GeV]T2am0 100 200 300 400 500

tW

t / t

00.5

1

1.5


TE~

0 1 2 3 4 5

Eve

nts

/ bin

100

200

300

400

500Data 2Ltt 1Ltt


VVTotal SM


STCR1

b-jet multiplicity0 1 2 3 4 5

Dat

a / S

M0.5

11.5

b-jet multiplicity0 1 2 3 4 5

Wt /

W+

jets

0123


TE~

0 1 2 3 4 5

Eve

nts

/ 0.3

10

20

30Data 2Ltt 1Ltt


VVTotal SM


STCR1

)2

,b1

R(b∆0 1 2 3 4 5

Dat

a / S

M

0.51

1.5

)2

,b1

R(b∆0 1 2 3 4 5

tW

t / t

00.5

1

1.5

Figure 3: Comparison of data with estimated backgrounds in the amT2 (top left), b-tagged jet multiplicity (top right),and ∆R(b1,b2) distributions with the STCR1 event selection except for the requirement (indicated by an arrow) ofthe variable shown. Furthermore, the ∆R(b1,b2) requirement is dropped in the shown b-tagged jet multiplicitydistribution. The predicted backgrounds are scaled with the NFs documented in Table 4. The uncertainty bandincludes statistical and all experimental systematic uncertainties. The last bin includes overflow. The middle panelshows the ratio of the data yield to the SM prediction, while the lower panel shows the ratio of the single top yieldto either the tt̄ prediction (top left and bottom) or the W+jets prediction (top right).

13


TE~

100 200 300 400

Eve

nts

/ 50

GeV

5

10

15

20

25

Data

1Ltt

γ+tt

Total SM


TZCR1


TE~

100 200 300 400

Dat

a / S

M

0.5

1

1.5 with photon added [GeV]missT = E

miss

TE~

0 100 200 300 400 500

Eve

nts

/ 50

GeV

5

10

15

20

25

Data

1Ltt

γ+tt

Total SM


TZCR1

with photon added [GeV]T = mTm~

0 100 200 300 400 500

Dat

a / S

M

0.5

1

1.5

Figure 4: Comparison of data with estimated backgrounds in the ẼmissT and m̃T distributions with the TZCR1 eventselection except for the requirement (indicated by an arrow) of the shown variable. The variables ẼmissT and m̃T areconstructed in the same way as EmissT and mT but treating the leading photon transverse momentum as invisible. Thepredicted backgrounds are scaled with the NFs documented in Table 4. The uncertainty band includes statisticaland all experimental systematic uncertainties. The last bin includes overflow.

the available phase space causing differences in kinematic distributions. In addition, the bremsstrahlungrate for Z bosons is highly suppressed at LHC energies, while there is a large contribution to the tt̄ + γcross-section from photons radiating from the top quark and its decay products. Both of these differencesare mitigated if the boson pT is larger than the Z boson mass. In this limit, the impact of the mass dif-ference on the available phase space is reduced and the rate of photon radiation from bremsstrahlung issuppressed [83]. In high EmissT tt̄ + Z (→ νν̄) events, the Z boson pT is the dominant source of E

missT and

so most tt̄ + Z events in the SRs have large Z pT.

Two tt̄ + γ CRs are designed to be kinematically close to SR1 and SR2/SR3. Both regions require atleast one signal photon, exactly one signal lepton and no additional baseline leptons, and at least foursignal jets from which at least one must be b-tagged. The two regions have the same jet pT thresholdsas the corresponding signal regions. To mimic the Z → νν̄ decay, the highest pT photon is vectoriallyadded to ~pmissT which is used to construct Ẽ

missT = |~p

missT + ~p

γT |, m̃T and H̃

missT,sig. Events entering the TZCRs

are required to satisfy ẼmissT > 120 GeV, m̃T > 100 GeV and H̃missT,sig > 5 in order to bring the region

kinematically closer to the SRs. Finally, EmissT < 200 GeV is imposed to ensure orthogonality betweenthe TZCR and the other CRs and SRs. The resulting regions have over 90% tt̄ + γ purity, and yield 43and 45 events in TZCR1, TZCR2 (=TZCR3), respectively. Figure 4 shows the distribution of ẼmissT andm̃T in the TZCR1 corresponding to SR1 before the requirement on the plotted variable is applied. Thecontribution from events not involving top quarks is negligible. The total number of events in data isabout 40% higher than in simulation, but there is no significant evidence for mis-modeling of the shapesof the various distributions within uncertainties.

14


TE~

0 0.1111 0.2222 0.3333 0.4444 0.5556 0.6667 0.7778 0.8889 1

Eve

nts

/ Reg

ion

1

10

210

Data ttSingle Top W+jets

+V tt VVTotal SM


TVR1 WVR1 TVR2 WVR2 TVR3 WVR3 SR1 SR2 SR3

tot

σ)

/ ex

p -

nob

s(n

-2

0

2

Figure 5: Comparison of the observed data (nobs) with the predicted background (nexp) in the validation and signalregions. The background predictions are obtained using the background-only fit configuration. The bottom panelshows the significance of the difference between data and predicted background, where the significance is based onthe total uncertainty (σtot).

6.2 Validation Regions

The background estimates are tested using validation regions, which are disjoint to both control andsignal regions. Background normalizations determined in the control regions are extrapolated to theVRs and compared with the observed data. Each signal region has associated validation regions for thett̄ (TVR) and W+jets (WVR) processes that are constructed with the same selection as the TCR/WCRexcept that mT is between 90 and 120 GeV.8 The validation regions are not used to constrain parametersin the fit, but provide a statistically independent test of the background estimates using the CRs. In Fig. 5background estimates in all the associated VRs are compared to the observed data. The potential signalcontamination in the VRs is studied for all considered signal models (and SUSY mass ranges) and foundto be negligible.

A second set of validation regions, not associated with any of the three signal regions, is used for generalmonitoring purposes. Some of the dominant backgrounds are dileptonic tt̄ and lepton+hadronic τ tt̄events. To pass the four-jet requirement, such events must have at least one hard jet that does not originate

8 A Wt VR is not defined since the mT range in the STCR is extended up to 120 GeV to increase the statistics.

15


TE~

2 3 4 5 6 7

Eve

nts

/ bin

50

100

150 Data

2Ltt

Other

Total SM


(OS)µVR 1e1

> 25 GeV)T

jet multiplicity (p2 3 4 5 6 7

Dat

a / S

M

0.5

1

1.5 with photon added [GeV]missT = E

miss

TE~

4 5 6 7 8

Eve

nts

/ bin

20

40

60

80

Data

τ 1L1tt

2Ltt

1Ltt

Single Top

Other

Total SM


τVR1L1

> 25 GeV)T

jet multiplicity (p4 5 6 7 8

Dat

a / S

M

0.5

1

1.5

Figure 6: Jet multiplicity distributions for events where exactly two signal leptons (left) or one lepton plus one τcandidate (right) are selected. No correction factors are included in the background normalizations. The uncertaintyband includes statistical and all experimental systematic uncertainties. The last bin includes overflow.

from the tt̄ decay (two hard jets for dileptonic tt̄). The modeling of these extra jets is validated in dedicatedVRs that require either two signal leptons (electron or muon) or one signal lepton and one hadronic τcandidate. In Fig. 6 the jet multiplicity distributions are shown for event selections of an electron-muonpair (left) and one lepton plus one τ candidate (right). Additional validation regions are constructed byconsidering (1) events with high EmissT , high mT and low amT2 for dilepton tt̄ events with a lost leptonor (2) high mT and a b-jet veto to probe the modeling of the resolution-induced mT tail in W+jets events(using the WVR-tail region in Fig. 2). There are no significant indications of mis-modeling in any of thevalidation regions.

7 Systematic Uncertainties

The systematic uncertainties on the signal and background estimates arise both from experimental sourcesand from the uncertainties in the theoretical predictions and modelling. Since the yields for the dominantbackground sources, tt̄, single top, tt̄V and W+jets, are all obtained in dedicated control regions, themodelling uncertainties for these processes affect only the extrapolation from the CRs into the signalregions (and between the various control regions), but not the overall normalization. The systematicuncertainties are included as nuisance parameters with Gaussian constraints and profiled in the likelihoodfits.

The dominant experimental uncertainties arise from imperfect knowledge of the jet energy scale (JES)and jet energy resolution (JER) [98], the modelling of the b-tagging efficiencies for b, c and light-flavourjets [123, 124] as well as the contribution to the EmissT soft-term i.e. from tracks not associated with anyreconstructed objects and from pileup. The resulting uncertainties from these sources on the extrapolationfactors from the four CRs to the SRs are 4–15% for JES, 0–9% for JER, 0–6% for b-tagging, and 0–3% for the EmissT soft-term. Other sources of experimental uncertainties are the modelling of lepton-and photon-related quantities (energy scales, resolutions, reconstruction and identification efficiencies,

16

isolation, hadronic-τ identification) and the uncertainty on the integrated luminosity. These uncertaintieshave a small impact on the final results.

The uncertainties on the modelling of the single top and tt̄ backgrounds include effects related to the MCevent generator, the hadronization and fragmentation modelling and the amount of initial and final stateradiation [67]. The MC generator uncertainty is estimated by comparing events produced with Powheg-Box+Herwig++ and with MG5_aMC+Herwig++. Events generated with Powheg-Box are hadronizedwith either Pythia or Herwig++ to estimate the effect from the modelling of the fragmentation andhadronization. The impact of altering the amount of initial- and final-state radiation is estimated fromcomparisons of Powheg-Box+Pythia samples with variable shower radiation, NLO radiation and modi-fied factorization and renormalization scales. One additional uncertainty stems from the modelling of theinterference between the tt̄ and Wt processes at NLO. The uncertainty is estimated using inclusive WW bbevents, generated using MG5_aMC, which are compared with the sum of the tt̄ and Wt processes [67].The resulting theoretical uncertainties on the extrapolation factors from the tt̄ and Wt CRs to the SRs are19–26% for tt̄, and 38–57% for Wt events, where the latter is dominated by the interference term.

The tt̄ + Z background is normalized using the tt̄ + γ CR and therefore there are uncertainties on boththe kinematic extrapolation to the SR and on the conversion between the two processes. As describedin Section 3, a small correction factor is applied to the tt̄ + γ cross-section to account for differences inthe generator setup, and the same k-factor is used for both processes. A first source of uncertainty isestimated by coherently varying the factorization and renormalization scales between tt̄ + Z and tt̄ + γevents generated at LO by a factor of two. The impact of the scale choice is slightly different betweentt̄ + Z and tt̄ +γ, leading to a 10% uncertainty for high pT bosons. An uncertainty due to NLO correctionsis estimated by studying the kinematic dependence of the ratio of tt̄ + Z and tt̄ + γ k-factors. Thisratio is studied by computing the k-factor for the tt̄ + Z and tt̄ + γ processes using MG5_aMC andOpenLoops+Sherpa as a function of the boson pT with a series of variations in the generator setup.Coherently varying the factorization and renormalization scale (set to HT =

∑pT for both LO and NLO)

by a factor of two results in a 5% uncertainty on the k-factor ratio. Comparing the results of the NNPDFand the CT14 [125] PDF sets changes the k-factor ratio by less than 2%. A final uncertainty of 5% is dueto the spread in k-factor ratios between the generators when the same scale and PDF are used, resultingfrom a different choice of electroweak scheme. The resulting theoretical systematic uncertainty on theextrapolation from the tt̄ + γ CR to the SR is 12%.

The uncertainty on the W+jets background from the merging of matrix elements and parton showers isstudied by varying the scales related to the matching scheme. In addition, the effects of varying the renor-malization, factorization and resummation scales are estimated. As the W+jets background is normalizedin a CR with a b-tagged jet veto, additional uncertainties on the flavour composition of the W+jets eventsin the signal region, based on the uncertainties on the measurement of Ref. [126] extrapolated to higherjet multiplicities, are applied in all regions requiring at least one b-tagged jet. The resulting theoreticaluncertainties on the extrapolation from the W+jets CR to the SR amounts to about 40%.

As the diboson backgrounds are not normalized in a CR, the analysis is sensitive to the uncertainty onthe total cross-section, estimated to be 6%. In addition, the estimate from the nominal Sherpa sample iscompared to that from a Powheg-Box+Pythia sample to account for differences related to the MC eventgenerator modelling. The resulting theoretical uncertainties for the diboson yields in the three SRs isabout 50%.

The SUSY signal cross-section uncertainty is taken from an envelope of cross-section predictions us-ing different PDF sets and factorization and renormalization scales, as described in Ref. [127], and the

17

[GeV]missTE

100 200 300 400 500 600 700 800 900

Eve

nts

/ 80

GeV

5

10

15

Data 2Ltt

Single TopW+jetsOther+Vtt

Total SM) = (1.1,0.7) TeV

1t~,m

g~(m

) = (1.1,0.8) TeV1

t~,m

g~(m

PreliminaryATLAS

-1 = 13 TeV, 3.2 fbs

SR1

[GeV]Tm

100 200 300 400 500 600 700 800 900

Eve

nts

/ 100

GeV

2

4

6

8

10

Data 2Ltt

Single TopW+jetsOther+Vtt

Total SM) = (1.1,0.7) TeV

1t~,m

g~(m

) = (1.1,0.8) TeV1

t~,m

g~(m

PreliminaryATLAS

-1 = 13 TeV, 3.2 fbs

SR1

Figure 7: The EmissT (left) and mT (right) distributions in SR1. In each plot, the full event selection in the correspond-ing signal region is applied, except for the requirement (indicated by an arrow) which is imposed on the variablebeing plotted. The predicted backgrounds are scaled with the NFs documented in Table 4. The uncertainty bandincludes statistical and all experimental systematic uncertainties. The last bin contains the overflow. Benchmarksignal models are overlaid for comparison.

resulting uncertainties range from 13% to 23%. The uncertainty on the VLQ signal cross-section is10% [76].

8 Results

Table 4 (top part) and Fig. 5 (right part) show the number of observed events together with the predictednumber of background events in the three SRs. The prediction is obtained using the background-only fitconfiguration described in Section 6. The SR2 and SR3 predicted yields agree well with the observed datain those regions. Table 4 (middle part) also lists the results of the four free fit parameters that control thenormalization of the four main backgrounds (normalization factors, NFs), together with the associated fituncertainties. To quantify the compatibility of the SM background-only hypothesis with the observationsin the SRs, a profile likelihood ratio test is performed. These fits are configured to include the SR binin the likelihood. Table 4 also reports the resulting p-values (p0), which are set to 0.5 for SR2 and SR3as the observation lies below the prediction. The data exceeds the background prediction in SR1 by 2.3standard deviations. Figure 7 shows the EmissT and mT distributions in SR1 for the data, for the backgroundprediction, as well as for two representative signal models.

The data are used to derive one-sided limits at 95% CL on generic beyond-SM yields and on the con-sidered signal models. The results are obtained from a profile likelihood ratio test following the CLsprescription [128]. Model-independent upper limits on beyond-SM contributions are derived separatelyfor each SR, where the fit is configured to include the SR and all its associated CRs. A generic signalmodel is assumed that contributes only to the SR and for which neither experimental nor theoretical sys-tematic uncertainties except for the luminosity uncertainty are considered. The resulting limits, expectedas well as observed, on the number of beyond-SM events are shown in the bottom rows of Table 4.

Exclusion limits are also derived for the gluino-mediated stop and direct stop pair production models.The signal uncertainties and potential signal contributions to all regions are taken into account. All un-certainties except those on the theoretical signal cross-section are included in the fit. Combined exclusion

18

Signal region SR1 SR2 SR3

Observed 12 1 1

Total bkg 5.50 ± 0.72 1.25 ± 0.26 1.03 ± 0.18tt̄ 2.21 ± 0.60 0.29 ± 0.10 0.20 ± 0.07Single top 0.46 ± 0.39 0.09 ± 0.08 0.10 ± 0.09W+jets 0.71 ± 0.43 0.15+0.19−0.15 0.20 ± 0.09tt̄ + W/Z 1.90 ± 0.42 0.61 ± 0.14 0.41 ± 0.10Diboson 0.23 ± 0.15 0.11 ± 0.07 0.12 ± 0.07tt̄ NF 1.10 ± 0.14 1.06 ± 0.14 0.80 ± 0.13Single top NF 0.62 ± 0.46 0.65 ± 0.49 0.71 ± 0.42W+jets NF 0.75 ± 0.12 0.78 ± 0.15 0.93 ± 0.12tt̄ + W/Z NF 1.42 ± 0.24 1.45 ± 0.24 1.46 ± 0.24p0 0.01(2.3σ) 0.50 (0.0σ) 0.50 (0.0σ)

N limitnon−SM exp. (95% CL) 6.4+3.2−2.0 3.6

+2.3−1.3 3.5

+2.2−1.2

N limitnon−SM obs. (95% CL) 13.3 3.4 3.4

Table 4: The numbers of observed events in the three SRs together with the expected numbers of background eventsas predicted by the background-only fits, the ratios of the post- to pre-fit background yields (NF), the probabili-ties (represented by the p0 values) that the observed numbers of events are compatible with the background-onlyhypothesis, as well as the expected and observed 95% CL upper limits on the number of beyond-SM events.

limits are obtained by selecting a priori the signal region with the lowest expected CLs value for eachsignal model.

Figure 8 shows the expected and observed exclusion contours for both gluino-mediated and direct pairproduction of stops. The ±1σexp (yellow) uncertainty band indicates the impact on the expected limit ofall uncertainties included in the fit. The ±1σth (dotted red) uncertainty lines around the observed limitillustrate the change in the observed limit as the nominal signal cross-section is scaled up and down by thetheoretical cross-section uncertainty. The gap in the observed exclusion between around 600 and 750 GeVin the direct stop model is due to the transition of signal regions (SR1 has the best expected sensitivity upto around 750 GeV for a massless χ̃01, beyond that SR2 has the best sensitivity) and the excess observedin SR1. The limits are sensitive to signal model assumptions. The gluino-mediated models have a 5 GeVmass splitting between the stop and the neutralino and 100% branching ratio t̃ → c + χ̃01. The impact ofvarying both of these assumptions is studied for SR2 with a benchmark model characterized by massesfor the gluino and the stop of (mg̃ ,mt̃1 ) = (1250,750) GeV. There is a small increase in the CLs valuewhen increasing the mass gap from 5 to 20 GeV and from switching between the two-body stop decay andthe four-body stop decay t̃ → b f f ′ χ̃01, each with 100% branching ratio, but under all of these variationsthe model is excluded. The direct stop pair production limits depend on the mixing of t̃L and t̃R in formingthe mass eigenstates t̃1 and t̃2. The nominal results assume that the t̃1 is mostly the t̃R. The stop masslimit for a massless neutralino is approximately 70 GeV weaker when the t̃1 is the t̃L.

The search for direct gluino and direct stop production can also be used to set limits on other models ofphysics beyond the SM that produce tt̄ + EmissT . Examples are third generation leptoquarks [129–135],which decay into a top quark and a neutrino (LQ → tν), and VLQ (T) models. For the former, limits

19

[GeV]g~m800 900 1000 1100 1200 1300 1400 1500 1600

[GeV

]1t~

m

300

400

500

600

700

800

900

1000

1100

1200

< 0t - m1t

~ - m

g~m

) = 5 GeV0

1χ∼,

1t~

m(∆+soft, 01

χ∼→1t~

, 1t~

t+→g~ production, g~g~

)thσ1±Observed limit ()expσ1±Expected limit (

-1ATLAS incl. 1L 8 TeV, 20.3 fbobs.exp.

ATLAS Preliminary-1 = 13 TeV, 3.2 fbs

Limit at 95% CL

[GeV]1t

~

m500 550 600 650 700 750 800 850

[GeV

]0 1χ∼

m

0

50

100

150

200

250

300

350

400

4500

1χ∼t+→1t

~ production, 1t

~1t

~

)thσ1±Observed limit ()expσ1±Expected limit (

-1ATLAS stop1L 8 TeV, 20.3 fbobs.exp.

ATLAS Preliminary-1 = 13 TeV, 3.2 fbs

Limit at 95% CL

Figure 8: Expected (black dashed) and observed (red solid) 95% excluded regions in the plane of mg̃ versus mt̃1 forgluino-mediated stop production (left), and in the plane of mt̃1 versus m χ̃01 for direct stop pair production (right).Both scenarios assume the SUSY decays shown on the plots, each with a branching ratio of 100%. The grey filledarea and grey dashed line shows the observed and expected exclusion limits, respectively, from ATLAS Run 1searches in the inclusive one lepton SUSY search [138] (left), and the stop search in the one lepton channel [30](right). The gap in the observed exclusion between around 600 and 750 GeV in the direct stop model is due to thetransition of signal regions and the excess observed in SR1. For any model point, the single signal region used forthe observed exclusion is chosen to be the one with the best expected CLs value.

on scalar LQ → tν are identical to limits on direct stop pair production with a massless neutralino andunpolarized top quarks. For the latter, simulated samples of pair produced T are used to re-interpret theresults. The T is assumed to decay in three possible ways: T → tZ , T → tH , and T → bW . The directT pair production cross-section is higher than for stops due to additional spin states, but after accountingfor the Z (→ νν̄) branching ratio, the models have a similar predicted yield. For a T quark with mass 800GeV (just beyond the Run 1 limit [136, 137]), a branching ratio B (T → tZ ) above 90% is excluded.

9 Conclusion

This document presents a search for pair production of gluino-mediated stops with a small mass splittingbetween the stop and the LSP, and direct pair production of stops in final states with one isolated lepton,jets, and missing transverse momentum. Three signal region selections are optimized for discoveringbenchmark models just beyond the exclusion limits of LHC Run 1 from searches with the same tt̄ + EmissTsignature. The search uses 3.2 fb−1of LHC data collected by the ATLAS experiment at a centre-of-mass energy of

√s = 13 TeV. The observed data are consistent with data-driven background estimates

in all three regions. The largest difference between data and the corresponding prediction is in the mostinclusive signal region (SR1) and corresponds to 2.3 standard deviations above the estimated background.In the absence of a significant excess, exclusion limits at 95% CL are derived on the target gluino andstop pair production models. These extend the LHC Run 1 exclusion limits on the gluino mass up to1460 GeV in the gluino-mediated stop pair production model in the high gluino and low stop mass region,and also add an excluded stop mass region from 745 to 780 GeV in the model of stop pair production witha massless lightest neutralino.

20

References

[1] Yu. A. Golfand and E. P. Likhtman,Extension of the Algebra of Poincare Group Generators and Violation of p Invariance,JETP Lett. 13 (1971) 323, [Pisma Zh. Eksp. Teor. Fiz.13,452(1971)].

[2] D. V. Volkov and V. P. Akulov, Is the Neutrino a Goldstone Particle?Phys. Lett. B 46 (1973) 109.

[3] J. Wess and B. Zumino, Supergauge Transformations in Four-Dimensions,Nucl. Phys. B 70 (1974) 39.

[4] J. Wess and B. Zumino, Supergauge Invariant Extension of Quantum Electrodynamics,Nucl. Phys. B 78 (1974) 1.

[5] S. Ferrara and B. Zumino, Supergauge Invariant Yang-Mills Theories,Nucl. Phys. B 79 (1974) 413.

[6] A. Salam and J. A. Strathdee, Supersymmetry and Nonabelian Gauges,Phys. Lett. B 51 (1974) 353.

[7] R. Barbieri and G. F. Giudice, Upper Bounds on Supersymmetric Particle Masses,Nucl. Phys. B 306 (1988) 63.

[8] B. de Carlos and J. A. Casas, One loop analysis of the electroweak breaking in supersymmetricmodels and the fine tuning problem, Phys. Lett. B 309 (1993) 320,arXiv: hep-ph/9303291 [hep-ph].

[9] N. Sakai, Naturalness in Supersymmetric Guts, Z. Phys. C 11 (1981) 153.

[10] S. Dimopoulos, S. Raby, and F. Wilczek, Supersymmetry and the Scale of Unification,Phys. Rev. D 24 (1981) 1681.

[11] L. E. Ibanez and G. G. Ross,Low-Energy Predictions in Supersymmetric Grand Unified Theories,Phys. Lett. B 105 (1981) 439.

[12] S. Dimopoulos and H. Georgi, Softly Broken Supersymmetry and SU(5),Nucl. Phys. B 193 (1981) 150.

[13] K. Inoue et al., Aspects of Grand Unified Models with Softly Broken Supersymmetry,Prog. Theor. Phys. 68 (1982) 927, [Erratum: Prog. Theor. Phys.70,330(1983)].

[14] J. R. Ellis and S. Rudaz, Search for Supersymmetry in Toponium Decays,Phys. Lett. B 128 (1983) 248.

[15] S. Raby, M. Ratz, and K. Schmidt-Hoberg, Precision gauge unification in the MSSM,Phys. Lett. B687 (2010) 342, arXiv: 0911.4249 [hep-ph].

[16] E. Hardy, Is Natural SUSY Natural? JHEP 10 (2013) 133, arXiv: 1306.1534 [hep-ph].

[17] A. Arvanitaki et al., The Last Vestiges of Naturalness, JHEP 03 (2014) 022,arXiv: 1309.3568 [hep-ph].

[18] K. Abe et al.,Search for proton decay via p→ νK+ using 260 kiloton-year data of Super-Kamiokande,Phys. Rev. D90 (2014) 072005, arXiv: 1408.1195 [hep-ex].

21

http://dx.doi.org/10.1016/0370-2693(73)90490-5http://dx.doi.org/10.1016/0370-2693(73)90490-5http://dx.doi.org/10.1016/0550-3213(74)90355-1http://dx.doi.org/10.1016/0550-3213(74)90355-1http://dx.doi.org/10.1016/0550-3213(74)90112-6http://dx.doi.org/10.1016/0550-3213(74)90112-6http://dx.doi.org/10.1016/0550-3213(74)90559-8http://dx.doi.org/10.1016/0550-3213(74)90559-8http://dx.doi.org/10.1016/0370-2693(74)90226-3http://dx.doi.org/10.1016/0370-2693(74)90226-3http://dx.doi.org/10.1016/0550-3213(88)90171-Xhttp://dx.doi.org/10.1016/0550-3213(88)90171-Xhttp://dx.doi.org/10.1016/0370-2693(93)90940-Jhttp://arxiv.org/abs/hep-ph/9303291http://dx.doi.org/10.1007/BF01573998http://dx.doi.org/10.1103/PhysRevD.24.1681http://dx.doi.org/10.1103/PhysRevD.24.1681http://dx.doi.org/10.1016/0370-2693(81)91200-4http://dx.doi.org/10.1016/0370-2693(81)91200-4http://dx.doi.org/10.1016/0550-3213(81)90522-8http://dx.doi.org/10.1016/0550-3213(81)90522-8http://dx.doi.org/10.1143/PTP.68.927http://dx.doi.org/10.1143/PTP.68.927http://dx.doi.org/10.1016/0370-2693(83)90402-1http://dx.doi.org/10.1016/0370-2693(83)90402-1http://dx.doi.org/10.1016/j.physletb.2010.03.060http://dx.doi.org/10.1016/j.physletb.2010.03.060http://arxiv.org/abs/0911.4249http://dx.doi.org/10.1007/JHEP10(2013)133http://arxiv.org/abs/1306.1534http://dx.doi.org/10.1007/JHEP03(2014)022http://arxiv.org/abs/1309.3568http://dx.doi.org/10.1103/PhysRevD.90.072005http://dx.doi.org/10.1103/PhysRevD.90.072005http://arxiv.org/abs/1408.1195

[19] G. R. Farrar and P. Fayet, Phenomenology of the Production, Decay, and Detection of NewHadronic States Associated with Supersymmetry, Phys. Lett. B 76 (1978) 575.

[20] H. Goldberg, Constraint on the Photino Mass from Cosmology,Phys. Rev. Lett. 50 (1983) 1419, [Erratum: Phys. Rev. Lett.103,099905(2009)].

[21] J. R. Ellis et al., Supersymmetric Relics from the Big Bang, Nucl. Phys. B 238 (1984) 453.

[22] P. Fayet, Supersymmetry and Weak, Electromagnetic and Strong Interactions,Phys. Lett. B 64 (1976) 159.

[23] P. Fayet, Spontaneously Broken Supersymmetric Theories of Weak, Electromagnetic and StrongInteractions, Phys. Lett. B 69 (1977) 489.

[24] A. De Simone, G. F. Giudice, and A. Strumia,Benchmarks for Dark Matter Searches at the LHC, JHEP 06 (2014) 081,arXiv: 1402.6287 [hep-ph].

[25] A. Delgado et al., The light stop window, Eur. Phys. J. C73 (2013) 2370,arXiv: 1212.6847 [hep-ph].

[26] S. Profumo and C. E. Yaguna,A Statistical analysis of supersymmetric dark matter in the MSSM after WMAP,Phys. Rev. D70 (2004) 095004, arXiv: hep-ph/0407036 [hep-ph].

[27] M Schmaltz and D Tucker-Smith, Little Higgs Review,Annu. Rev. Nucl. Part. Sci. 55 (2005) 229, url: https://cds.cern.ch/record/823954.

[28] K Agashe, R. Contino, and A Pomarol, The Minimal Composite Higgs Model,Nucl. Phys. B 719 (2004) 165, url: https://cds.cern.ch/record/809132.

[29] J. A. Aguilar-Saavedra et al., Handbook of vectorlike quarks: Mixing and single production,Phys. Rev. D88 (2013) 094010, arXiv: 1306.0572 [hep-ph].

[30] ATLAS Collaboration,Search for top squark pair production in final states with one isolated lepton, jets, and missingtransverse momentum in

√s = 8 TeV pp collisions with the ATLAS detector,

JHEP 1411 (2014) 118, arXiv: 1407.0583 [hep-ex].

[31] ATLAS Collaboration,Search for direct top squark pair production in final states with one isolated lepton, jets, andmissing transverse momentum in

√s = 7 TeV pp collisions using 4.7 fb−1 of ATLAS data,

Phys. Rev. Lett. 109 (2012) 211803, arXiv: 1208.2590 [hep-ex].

[32] ATLAS Collaboration, Summary of the searches for squarks and gluinos using√

s = 8 TeV ppcollisions with the ATLAS experiment at the LHC, JHEP 1510 (2015) 054,arXiv: 1507.05525 [hep-ex].

[33] ATLAS Collaboration, ATLAS Run 1 searches for direct pair production of third-generationsquarks at the Large Hadron Collider, Eur. Phys. J. C 75 (2015) 510,arXiv: 1506.08616 [hep-ex].

[34] CMS Collaboration, Search for supersymmetry in pp collisions at sqrt(s) = 8 TeV in final stateswith boosted W bosons and b jets using razor variables (2016), arXiv: 1602.02917 [hep-ex].

[35] CMS Collaboration, Searches for third-generation squark production in fully hadronic finalstates in proton-proton collisions at

√s = 8 TeV, JHEP 06 (2015) 116,

arXiv: 1503.08037 [hep-ex].

22

http://dx.doi.org/10.1016/0370-2693(78)90858-4http://dx.doi.org/10.1103/PhysRevLett.50.1419http://dx.doi.org/10.1103/PhysRevLett.50.1419http://dx.doi.org/10.1016/0550-3213(84)90461-9http://dx.doi.org/10.1016/0370-2693(76)90319-1http://dx.doi.org/10.1016/0370-2693(76)90319-1http://dx.doi.org/10.1016/0370-2693(77)90852-8http://dx.doi.org/10.1007/JHEP06(2014)081http://arxiv.org/abs/1402.6287http://dx.doi.org/10.1140/epjc/s10052-013-2370-5http://arxiv.org/abs/1212.6847http://dx.doi.org/10.1103/PhysRevD.70.095004http://dx.doi.org/10.1103/PhysRevD.70.095004http://arxiv.org/abs/hep-ph/0407036https://cds.cern.ch/record/823954https://cds.cern.ch/record/809132http://dx.doi.org/10.1103/PhysRevD.88.094010http://dx.doi.org/10.1103/PhysRevD.88.094010http://arxiv.org/abs/1306.0572http://dx.doi.org/10.1007/JHEP11(2014)118http://dx.doi.org/10.1007/JHEP11(2014)118http://arxiv.org/abs/1407.0583http://dx.doi.org/10.1103/PhysRevLett.109.211803http://dx.doi.org/10.1103/PhysRevLett.109.211803http://arxiv.org/abs/1208.2590http://dx.doi.org/10.1007/JHEP10(2015)054http://arxiv.org/abs/1507.05525http://dx.doi.org/10.1140/epjc/s10052-015-3726-9http://arxiv.org/abs/1506.08616http://arxiv.org/abs/1602.02917http://dx.doi.org/10.1007/JHEP06(2015)116http://arxiv.org/abs/1503.08037

[36] CMS Collaboration, Search for direct pair production of scalar top quarks in the single- anddilepton channels in proton-proton collisions at sqrt(s) = 8 TeV (2016),arXiv: 1602.03169 [hep-ex].

[37] CMS Collaboration, Search for top-squark pair production in the single-lepton final state in ppcollisions at

√s = 8 TeV, Eur. Phys. J. C 73 (2013) 2677, arXiv: 1308.1586 [hep-ex].

[38] CMS Collaboration, Search for supersymmetry using razor variables in events with b-tagged jetsin pp collisions at

√s = 8 TeV, Phys. Rev. D 91 (2015) 052018, arXiv: 1502.00300 [hep-ex].

[39] CMS Collaboration, Searches for supersymmetry using the MT2 variable in hadronic eventsproduced in pp collisions at 8 TeV, JHEP 1505 (2015) 078, arXiv: 1502.04358 [hep-ex].

[40] ATLAS Collaboration, The ATLAS Experiment at the CERN Large Hadron Collider,JINST 3 (2008) S08003.

[41] ATLAS Collaboration,Early Inner Detector Tracking Performance in the 2015 Data at

√s = 13 TeV,

ATL-PHYS-PUB-2015-051, 2015, url: http://cdsweb.cern.ch/record/2110140.

[42] ATLAS Collaboration, Improved luminosity determination in pp collisions at√

s = 7 TeV usingthe ATLAS detector at the LHC, Eur. Phys. J. C 73 (2013) 2518, arXiv: 1302.4393 [hep-ex].

[43] J. Alwall et al., The automated computation of tree-level and next-to-leading order differentialcross sections, and their matching to parton shower simulations, JHEP 07 (2014) 079,arXiv: 1405.0301 [hep-ph].

[44] J. A. Aguilar-Saavedra, Identifying top partners at LHC, JHEP 11 (2009) 030,arXiv: 0907.3155 [hep-ph].

[45] J. A. Aguilar-Saavedra, “Protos - PROgram for TOp Simulations,”url: https://jaguilar.web.cern.ch/jaguilar/protos/.

[46] T. Sjostrand, S. Mrenna, and P. Z. Skands, A Brief Introduction to PYTHIA 8.1,Comput. Phys. Commun. 178 (2008) 852, arXiv: 0710.3820 [hep-ph].

[47] L. Lonnblad and S. Prestel, Matching Tree-Level Matrix Elements with Interleaved Showers,JHEP 03 (2012) 019, arXiv: 1109.4829 [hep-ph].

[48] S. Alioli et al., A general framework for implementing NLO calculations in shower Monte Carloprograms: the POWHEG BOX, JHEP 06 (2010) 043, arXiv: 1002.2581 [hep-ph].

[49] E. Re,Single-top Wt-channel production matched with parton showers using the POWHEG method,Eur. Phys. J. C71 (2011) 1547, arXiv: 1009.2450 [hep-ph].

[50] S. Frixione, P. Nason, and G. Ridolfi,A Positive-weight next-to-leading-order Monte Carlo for heavy flavour hadroproduction,JHEP 09 (2007) 126, arXiv: 0707.3088 [hep-ph].

[51] R. Frederix, E. Re, and P. Torrielli, Single-top t-channel hadroproduction in the four-flavourscheme with POWHEG and aMC@NLO, JHEP 09 (2012) 130, arXiv: 1207.5391 [hep-ph].

[52] S. Alioli et al.,NLO single-top production matched with shower in POWHEG: s- and t-channel contributions,JHEP 09 (2009) 111, [Erratum: JHEP02,011(2010)], arXiv: 0907.4076 [hep-ph].

[53] T. Sjöstrand, S. Mrenna, and P. Z. Skands, PYTHIA 6.4 Physics and Manual,JHEP 0605 (2006) 026, arXiv: hep-ph/0603175.

23

http://arxiv.org/abs/1602.03169http://dx.doi.org/10.1140/epjc/s10052-013-2677-2http://arxiv.org/abs/1308.1586http://dx.doi.org/10.1103/PhysRevD.91.052018http://arxiv.org/abs/1502.00300http://dx.doi.org/10.1007/JHEP05(2015)078http://arxiv.org/abs/1502.04358http://dx.doi.org/10.1088/1748-0221/3/08/S08003http://dx.doi.org/10.1088/1748-0221/3/08/S08003http://cdsweb.cern.ch/record/2110140http://dx.doi.org/10.1140/epjc/s10052-013-2518-3http://arxiv.org/abs/1302.4393http://dx.doi.org/10.1007/JHEP07(2014)079http://arxiv.org/abs/1405.0301http://dx.doi.org/10.1088/1126-6708/2009/11/030http://arxiv.org/abs/0907.3155https://jaguilar.web.cern.ch/jaguilar/protos/http://dx.doi.org/10.1016/j.cpc.2008.01.036http://dx.doi.org/10.1016/j.cpc.2008.01.036http://arxiv.org/abs/0710.3820http://dx.doi.org/10.1007/JHEP03(2012)019http://dx.doi.org/10.1007/JHEP03(2012)019http://arxiv.org/abs/1109.4829http://dx.doi.org/10.1007/JHEP06(2010)043http://arxiv.org/abs/1002.2581http://dx.doi.org/10.1140/epjc/s10052-011-1547-zhttp://dx.doi.org/10.1140/epjc/s10052-011-1547-zhttp://arxiv.org/abs/1009.2450http://dx.doi.org/10.1088/1126-6708/2007/09/126http://dx.doi.org/10.1088/1126-6708/2007/09/126http://arxiv.org/abs/0707.3088http://dx.doi.org/10.1007/JHEP09(2012)130http://arxiv.org/abs/1207.5391http://dx.doi.org/10.1007/JHEP02(2010)011, 10.1088/1126-6708/2009/09/111http://dx.doi.org/10.1007/JHEP02(2010)011, 10.1088/1126-6708/2009/09/111http://arxiv.org/abs/0907.4076http://dx.doi.org/10.1088/1126-6708/2006/05/026http://dx.doi.org/10.1088/1126-6708/2006/05/026http://arxiv.org/abs/hep-ph/0603175

[54] P. Nason, A New method for combining NLO QCD with shower Monte Carlo algorithms,JHEP 11 (2004) 040, arXiv: hep-ph/0409146 [hep-ph].

[55] S. Frixione, P. Nason, and C. Oleari,Matching NLO QCD computations with Parton Shower simulations: the POWHEG method,JHEP 11 (2007) 070, arXiv: 0709.2092 [hep-ph].

[56] T. Gleisberg et al., Event generation with SHERPA 1.1, JHEP 0902 (2009) 007,arXiv: 0811.4622 [hep-ph].

[57] T. Gleisberg and S. Höche, Comix, a new matrix element generator, JHEP 0812 (2008) 039,arXiv: 0808.3674 [hep-ph].

[58] F. Cascioli, P. Maierhofer, and S. Pozzorini, Scattering Amplitudes with Open Loops,Phys. Rev. Lett. 108 (2012) 111601, arXiv: 1111.5206 [hep-ph].

[59] S. Schumann and F. Krauss,A Parton shower algorithm based on Catani-Seymour dipole factorisation,JHEP 0803 (2008) 038, arXiv: 0709.1027 [hep-ph].

[60] H.-L. Lai et al., New parton distributions for collider physics, Phys. Rev. D82 (2010) 074024,arXiv: 1007.2241 [hep-ph].

[61] R. D. Ball et al., Parton distributions with LHC data, Nucl. Phys. B867 (2013) 244,arXiv: 1207.1303 [hep-ph].

[62] R. D. Ball et al., Parton distributions for the LHC Run II, JHEP 04 (2015) 040,arXiv: 1410.8849 [hep-ph].

[63] J. Pumplin et al.,New generation of parton distributions with uncertainties from global QCD analysis,JHEP 07 (2002) 012, arXiv: hep-ph/0201195.

[64] P. Z. Skands, Tuning Monte Carlo Generators: The Perugia Tunes,Phys. Rev. D82 (2010) 074018, arXiv: 1005.3457 [hep-ph].

[65] ATLAS Collaboration, ATLAS Pythia 8 tunes to 7 TeV data, ATL-PHYS-PUB-2014-021, 2014,url: http://cdsweb.cern.ch/record/1966419.

[66] D. J. Lange, The EvtGen particle decay simulation package,Nucl. Instrum. Meth. A462 (2001) 152.

[67] ATLAS Collaboration, Studies on the MC generator modelling of top pair and single top (Wt)processes as used in Run2, ATL-PHYS-PUB-2016-004, 2016,url: http://cdsweb.cern.ch/record/2120417.

[68] ATLAS Collaboration, Monte Carlo Generators for the Production of a W or Z/γ∗ Boson inAssociation with Jets at ATLAS in Run 2, ATL-PHYS-PUB-2016-003, 2016,url: http://cdsweb.cern.ch/record/2120133.

[69] ATLAS Collaboration, Studies on the MC generator modelling of ttX (X=W,Z,gamma,Higgs orheavy flavor jets) processes as used in Run2, ATL-PHYS-PUB-2016-005, 2016,url: https://cds.cern.ch/record/2120826.

[70] ATLAS Collaboration, Multi-boson simulation for 13 TeV ATLAS analyses,ATL-PHYS-PUB-2016-002, 2016, url: http://cdsweb.cern.ch/record/2119986.

24

http://dx.doi.org/10.1088/1126-6708/2004/11/040http://dx.doi.org/10.1088/1126-6708/2004/11/040http://arxiv.org/abs/hep-ph/0409146http://dx.doi.org/10.1088/1126-6708/2007/11/070http://dx.doi.org/10.1088/1126-6708/2007/11/070http://arxiv.org/abs/0709.2092http://dx.doi.org/10.1088/1126-6708/2009/02/007http://arxiv.org/abs/0811.4622http://dx.doi.org/10.1088/1126-6708/2008/12/039http://arxiv.org/abs/0808.3674http://dx.doi.org/10.1103/PhysRevLett.108.111601http://dx.doi.org/10.1103/PhysRevLett.108.111601http://arxiv.org/abs/1111.5206http://dx.doi.org/10.1088/1126-6708/2008/03/038http://dx.doi.org/10.1088/1126-6708/2008/03/038http://arxiv.org/abs/0709.1027http://dx.doi.org/10.1103/PhysRevD.82.074024http://arxiv.org/abs/1007.2241http://dx.doi.org/10.1016/j.nuclphysb.2012.10.003http://arxiv.org/abs/1207.1303http://dx.doi.org/10.1007/JHEP04(2015)040http://arxiv.org/abs/1410.8849http://arxiv.org/abs/hep-ph/0201195http://dx.doi.org/10.1103/PhysRevD.82.074018http://dx.doi.org/10.1103/PhysRevD.82.074018http://arxiv.org/abs/1005.3457http://cdsweb.cern.ch/record/1966419http://dx.doi.org/10.1016/S0168-9002(01)00089-4http://dx.doi.org/10.1016/S0168-9002(01)00089-4http://cdsweb.cern.ch/record/2120417http://cdsweb.cern.ch/record/2120133https://cds.cern.ch/record/2120826http://cdsweb.cern.ch/record/2119986

[71] M. Czakon, P. Fiedler, and A. Mitov,Total Top-Quark Pair-Production Cross Section at Hadron Colliders Through O(α 4S ),Phys. Rev. Lett. 110 (2013) 252004, arXiv: 1303.6254 [hep-ph].

[72] M. Czakon and A. Mitov,NNLO corrections to top pair production at hadron colliders: the quark-gluon reaction,JHEP 01 (2013) 080, arXiv: 1210.6832 [hep-ph].

[73] M. Czakon and A. Mitov, NNLO corrections to top-pair production at hadron colliders: theall-fermionic scattering channels, JHEP 12 (2012) 054, arXiv: 1207.0236 [hep-ph].

[74] P. Bärnreuther, M. Czakon, and A. Mitov, Percent Level Precision Physics at the Tevatron: FirstGenuine NNLO QCD Corrections to qq̄ → tt̄ + X , Phys. Rev. Lett. 109 (2012) 132001,arXiv: 1204.5201 [hep-ph].

[75] M. Cacciari et al., Top-pair production at hadron colliders with next-to-next-to-leadinglogarithmic soft-gluon resummation, Phys. Lett. B710 (2012) 612,arXiv: 1111.5869 [hep-ph].

[76] M. Czakon and A. Mitov,Top++: A Program for the Calculation of the Top-Pair Cross-Section at Hadron Colliders,Comput. Phys. Commun. 185 (2014) 2930, arXiv: 1112.5675 [hep-ph].

[77] N. Kidonakis, Next-to-next-to-leading-order collinear and soft gluon corrections for t-channelsingle top quark production, Phys. Rev. D83 (2011) 091503, arXiv: 1103.2792 [hep-ph].

[78] N. Kidonakis,Two-loop soft anomalous dimensions for single top quark associated production with a W- or H-,Phys. Rev. D82 (2010) 054018, arXiv: 1005.4451 [hep-ph].

[79] N. Kidonakis, NNLL resummation for s-channel single top quark production,Phys. Rev. D81 (2010) 054028, arXiv: 1001.5034 [hep-ph].

[80] S. Catani et al.,Vector boson production at hadron colliders: a fully exclusive QCD calculation at NNLO,Phys. Rev. Lett. 103 (2009) 082001, arXiv: 0903.2120 [hep-ph].

[81] C. Borschensky et al.,Squark and gluino production cross sections in pp collisions at

√s = 13, 14, 33 and 100 TeV,

Eur. Phys. J. C74 (2014) 3174, arXiv: 1407.5066 [hep-ph].

[82] I. Low, Polarized charginos (and top quarks) in scalar top quark decays,Phys. Rev. D88 (2013) 095018, arXiv: 1304.0491 [hep-ph].

[83] K. Melnikov, M. Schulze, and A. Scharf,QCD corrections to top quark pair production in association with a photon at hadron colliders,Phys. Rev. D83 (2011) 074013, arXiv: 1102.1967 [hep-ph].

[84] ATLAS Collaboration, The ATLAS Simulation Infrastructure, Eur. Phys. J. C 70 (2010) 823,arXiv: 1005.4568 [hep-ex].

[85] S. Agostinelli et al. (GEANT4 Collaboration), GEANT4: A Simulation toolkit,Nucl. Instrum. Meth. A506 (2003) 250.

[86] ATLAS Collaboration, The simulation principle and performance of the ATLAS fast calorimetersimulation FastCaloSim, ATL-PHYS-PUB-2010-013, 2010,url: http://cds.cern.ch/record/1300517.

25

http://dx.doi.org/10.1103/PhysRevLett.110.252004http://dx.doi.org/10.1103/PhysRevLett.110.252004http://arxiv.org/abs/1303.6254http://dx.doi.org/10.1007/JHEP01(2013)080http://dx.doi.org/10.1007/JHEP01(2013)080http://arxiv.org/abs/1210.6832http://dx.doi.org/10.1007/JHEP12(2012)054http://arxiv.org/abs/1207.0236http://dx.doi.org/10.1103/PhysRevLett.109.132001http://arxiv.org/abs/1204.5201http://dx.doi.org/10.1016/j.physletb.2012.03.013http://arxiv.org/abs/1111.5869http://dx.doi.org/10.1016/j.cpc.2014.06.021http://dx.doi.org/10.1016/j.cpc.2014.06.021http://arxiv.org/abs/1112.5675http://dx.doi.org/10.1103/PhysRevD.83.091503http://arxiv.org/abs/1103.2792http://dx.doi.org/10.1103/PhysRevD.82.054018http://dx.doi.org/10.1103/PhysRevD.82.054018http://arxiv.org/abs/1005.4451http://dx.doi.org/10.1103/PhysRevD.81.054028http://dx.doi.org/10.1103/PhysRevD.81.054028http://arxiv.org/abs/1001.5034http://dx.doi.org/10.1103/PhysRevLett.103.082001http://dx.doi.org/10.1103/PhysRevLett.103.082001http://arxiv.org/abs/0903.2120http://dx.doi.org/10.1140/epjc/s10052-014-3174-yhttp://dx.doi.org/10.1140/epjc/s10052-014-3174-yhttp://arxiv.org/abs/1407.5066http://dx.doi.org/10.1103/PhysRevD.88.095018http://dx.doi.org/10.1103/PhysRevD.88.095018http://arxiv.org/abs/1304.0491http://dx.doi.org/10.1103/PhysRevD.83.074013http://dx.doi.org/10.1103/PhysRevD.83.074013http://arxiv.org/abs/1102.1967http://dx.doi.org/10.1140/epjc/s10052-010-1429-9http://arxiv.org/abs/1005.4568http://dx.doi.org/10.1016/S0168-9002(03)01368-8http://dx.doi.org/10.1016/S0168-9002(03)01368-8http://cds.cern.ch/record/1300517

[87] ATLAS Collaboration,Electron and photon energy calibration with the ATLAS detector using LHC Run 1 data,Eur. Phys. J. C 74 (2014) 3071, arXiv: 1407.5063 [hep-ex].

[88] ATLAS Collaboration, Electron identification measurements in ATLAS using√

s = 13 TeV datawith 50 ns bunch spacing, ATL-PHYS-PUB-2015-041, 2015,url: http://cdsweb.cern.ch/record/2048202.

[89] ATLAS Collaboration, Muon reconstruction performance in early√

s = 13 TeV data,ATL-PHYS-PUB-2015-037, 2015, url: http://cdsweb.cern.ch/record/2047831.

[90] ATLAS Collaboration, Expected photon performance in the ATLAS experiment,ATL-PHYS-PUB-2011-007, 2011, url: http://cdsweb.cern.ch/record/1345329.

[91] Lampl, W and others, Calorimeter Clustering Algorithms: Description and Performance,ATL-LARG-PUB-2008-002. ATL-COM-LARG-2008-003, 2008,url: cds.cern.ch/record/1099735.

[92] ATLAS Collaboration,Jet energy measurement with the ATLAS detector in proton–proton collisions at

√s = 7 TeV,

Eur. Phys. J. C 73 (2013) 2304, arXiv: 1112.6426 [hep-ex].

[93] M. Cacciari, G. P. Salam, and G. Soyez, The Anti-k(t) jet clustering algorithm,JHEP 04 (2008) 063, arXiv: 0802.1189 [hep-ph].

[94] M. Cacciari and G. P. Salam, Pileup subtraction using jet areas, Phys. Lett. B659 (2008) 119,arXiv: 0707.1378 [hep-ph].

[95] M. Cacciari, G. P. Salam, and G. Soyez, The Catchment Area of Jets, JHEP 04 (2008) 005,arXiv: 0802.1188 [hep-ph].

[96] ATLAS Collaboration, Performance of pile-up mitigation techniques for jets in pp collisions at√s = 8 TeV using the ATLAS detector (2015), arXiv: 1510.03823 [hep-ex].

[97] ATLAS Collaboration, Jet energy measurement and its systematic uncertainty in proton–protoncollisions at

√s = 7 TeV with the ATLAS detector, Eur. Phys. J. C 75 (2015) 17,

arXiv: 1406.0076 [hep-ex].

[98] ATLAS Collaboration, Jet Calibration and Systematic Uncertainties for Jets Reconstructed inthe ATLAS Detector at

√s = 13 TeV, ATL-PHYS-PUB-2015-015, 2015,

url: https://cds.cern.ch/record/2037613.

[99] ATLAS Collaboration, Characterisation and mitigation of beam-induced backgrounds observedin the ATLAS detector during t

ATLAS NOTE - CERNcds.cern.ch/record/2139641/files/ATLAS-CONF-2016-007.pdf · 2016. 3. 16. · The...

Documents

Transcript of ATLAS NOTE - CERNcds.cern.ch/record/2139641/files/ATLAS-CONF-2016-007.pdf · 2016. 3. 16. · The...