arXiv:1209.0590v1 [hep-ex] 4 Sep 2012 aiso h htn n hte hycnetd h backgro The converted. depend they whether fraction, and signal expected the and of matics resolution mass differing with o-htnbcgon sls hn30%. than less is background non- civsams eouino bu 1 about of resolution mass a achieves good h ihtndcyo h ig oo srltvl ae(ihabranc a (with rare relatively is boson Higgs the of ∼ decay diphoton The 2.1 olwdby followed itntv nuht erhfrdrcl.FrlwrHgsbsnmas boson Higgs two lower into For is directly. decay for search to enough distinctive arof pair n fteguebsn so-hl,o o ig oo rdcdin produced boson Higgs a for or off-shell, is a bosons gauge the of one o ig oo ihmass with boson Higgs a For nesbet erhfrti ea iety n a okfrrrrb as rarer such for boson, look can Higgs One the directly. of decay decays this for search to infeasible h 01rno h ag arnClie LC tCR a eysu very was CERN at (LHC) Collider fb Hadron 5 nearly Large accumulating the experiments of with run 2011 The Introduction 1 USA 11973, NY, Upton, Collaboration ATLAS Laboratory the National of Brookhaven behalf on Snyder Scott h oiatpouto oefraHgsbsni lo-lo fusion gluon-gluon is state boson final Higgs different a many for for mode out production carried dominant is the search channels boson search Higgs The boson Higgs 2 erha h TA xeiet[]uigtefl 01dt e of set data 2011 full t the summarizes using note [1] This experiment ATLAS observed. comp the definitively remaining at the been search yet boson, Higgs not (SM) has Model that Standard a for search W 0 ootmz h xetdsniiiy h nlssi pi nonn subc nine into split is analysis the sensitivity, expected the optimize To . %.I re ohv odsniiiyi hscanl[] n utha must one [2], channel this in sensitivity good have to order In 2%). or m H W γγ Z → eouina ela odcnrloe o-htnbackgrounds non-photon over control good as well as resolution oo.Tebs estvt for sensitivity best The boson. or H γγ Z → ag oos uhdcy hticuelposi h nlsaea state final the in leptons include that decays such bosons; gauge channel b ig oo erha ATLAS at search boson Higgs ZZ qak.Tevr ag ev-aormlie akrudmake background multijet heavy-flavour large very The -quarks. ( ∗ ) → ℓℓℓ m ′ ℓ H ′ . bv bu 3 e,tedmnn ea sit a into is decay dominant the GeV, 135 about above H → . %at 7% γγ 1 m − H 1 or fdt.O atclritrs a the was interest particular Of data. of ≈ m H H 2 e sotie from obtained is GeV 125 → 2 e,adtefato of fraction the and GeV, 120 = ττ o ioo easi which in decays diboson for , n smdldwith modeled is und tmr distinctive more ut e,tedominant the ses, (4 n ntekine- the on ing esau fthis of status he nn fteSM the of onent soito with association .A h LHC, the At s. . 6–4 ( . gg igratio hing )fb 9) ATLAS . H ccessful, hannels every ve → → − 1 H γγ it s . re ). , an exponential fit to the data. Results are shown in Figure 1. The mH regions (113–115) GeV and (134.5–136) GeV are excluded at 95% CL. A small excess is seen around mH = 126.5 GeV, with a local significance of 2.8σ, or a global significance of 1.5σ when the look-elsewhere effect is taken into account over the mH range (110– 150) GeV. 0 800 Observed p Data 2011, s = 7 TeV ATLAS Data 2011 10 0 γγ→ 700 H Total background Local p SM H → γγ expected p 0 -1 1 ∫Ldt = 4.9 fb

Events / GeV 600 mH=125 GeV, 1 x SM 1σ 500 10-1 -1 s = 7 TeV, ∫ Ldt = 4.9 fb 2σ 400 10-2 ATLAS 300 3σ 10-3 200 Observed p (with energy scale uncertainty) 0 100 10-4 100 110 120 130 140 150 160 110 115 120 125 130 135 140 145 150

m γγ [GeV] mH [GeV]

Figure 1: Results from the H→γγ channel [2]. Left: Invariant mass distribution for the entire sample, overlaid with the fitted total background. The expectation for a mH = 125 GeV SM is also shown. Right: Local probability p0 for the background to fluctuate to the observed number of events or higher. The dashed line shows the expected median local p0 for the signal hypothesis when tested at mH .

2.2 H→ZZ(∗)→ℓℓℓ′ℓ′ channel In this channel [3], one looks for two pairs of opposite-sign, same-flavour leptons (e or µ), with one pair having an invariant mass close to the Z boson mass. This channel has a quite small background and a fully-reconstructed Higgs boson decay, which allows it to have good sensitivity over a wide range of Higgs boson masses, from 600 GeV down to 110 GeV. The background is primarily ZZ diboson production, with smaller contributions from Z + jets and tt. The results are shown in Figure 2, Row 1. The mH regions (134–156) GeV, (182– 233) GeV, (256–265) GeV, and (268–415) GeV are excluded at 95% CL. Small ex- cesses are seen around 125 GeV, 244 GeV, and 500 GeV with local significances of 2.1σ,2.2σ, and 2.1σ, respectively. The excess at 125 GeV corresponds to three events that cluster within the mH mass resolution of 2%: two 2e2µ events at 123.6 and 124.3 GeV, and one 4µ event at 124.3 GeV.

2.3 H→WW (∗)→ℓνℓν channel This channel [4] is also sensitive over a wide mass range. The event selection requires miss two isolated, opposite-sign leptons and a large missing transverse energy (ET ). The

2 102

14 SM

Data 2011 ATLAS ATLAS σ 12 Data 2011 / ATLAS Observed CLs → (*)→ → (*)→ σ (*) H ZZ 4l 12 H ZZ 4l H→ ZZ → 4l Expected CLs 10 Total background Total background ∫Ldt = 4.8 fb-1 ± 1 σ m =125 GeV, 1 x SM 10 ± σ H mH=125 GeV, 1 x SM 10 s=7 TeV 2 Events / 5 GeV 8 -1 Events / 10 GeV 8 s = 7 TeV, ∫ Ldt = 4.8 fb -1 6 s = 7 TeV, ∫ Ldt = 4.8 fb 6 on limit CL 95%

4 4 1

2 2

0 0 100 120 140 160 180 200 220 240 0 100 200 300 400 500 600 10-1 110 200 300 400 500 600 mllll [GeV] mllll [GeV] mH [GeV]

2 100 30 10 (*) SM ATLAS H→WW →lνlν ATLAS σ 90 ATLAS H→WW→lνlν+0j Data 2011 / 25 H→WW→lνlν+1j σ Obs. -1 m =125 GeV, 1xSM Ldt = 4.7 fb 80 Data 2011 H Exp. ∫ ± σ 70 Total background 1 m =125 GeV, 1 x SM 10 s = 7 TeV H 20 ±2 σ 60 Events / 10 GeV Events / 10 GeV -1 Total background s = 7 TeV, ∫ Ldt = 4.7 fb 50 15

40 s = 7 TeV, ∫ Ldt = 4.7 fb-1 on Limit CL 95% 10 1 30 20 5 10 -1 0 0 10 60 80 100 120 140 160 180 200 220 240 60 80 100 120 140 160 180 200 220 240 100 200 300 400 500 600 mH [GeV] mT [GeV] mT [GeV]

12 120 ATLAS Data 2011 ATLAS Data 2011 SM 450 Signal × 5 σ

/ ATLAS -1 Signal × 5 -1 -1 L dt = 4.7 fb , s = 7 TeV L dt = 4.7 fb , s = 7 TeV (m =120 GeV) σ s = 7 TeV, Ldt = 4.6-4.7 fb ∫ (m =120 GeV) ∫ H Observed (CLs) ∫ 400 Total BG 10 100 ZH → l+l-bb H WH → lνbb Total BG 350 Top Expected (CLs) VH(bb), combined Top Z+jets ± 1σ W+jets 80 300 8 ± 2σ Z+jets Diboson Diboson 250 Multijet

Events / 10 GeV 60 Events / 10 GeV 200 6

40 150 95% C.L. limit on 4 100 20 50 2 0 0 0 50 100 150 200 250 0 50 100 150 200 250 0 m [GeV] m [GeV] 110 115 120 125 130 bb bb

mH [GeV]

20 250 SM → ττ τ→ τ σ H ATLAS H τ→ τ + 1j 1800 ATLAS H lep had + 0/1j / 18 ATLAS lep lep σ Observed CL 1600 s 16 Expected CL ≈ -1 200 Data 2011 s ∫ Ldt 4.7 fb Data 2011 1400 14 ± 2σ m =125 GeV, 10xSM s = 7 TeV Total background 1200 H ± 1σ Events / 20 GeV 150 Events / 10 GeV 12 1000 Total background 10 mH=125 GeV, 10 x SM

800 on Limit CL 95% 8 100 -1 -1 s = 7 TeV, ∫ Ldt = 4.7 fb 600 s = 7 TeV, ∫ Ldt = 4.7 fb 6 50 400 4 200 2 0 0 0 50 100 150 200 250 300 0 50 100 150 200 250 300 350 400 100 110 120 130 140 150

m ττ [GeV] mMMC [GeV] mH [GeV]

Figure 2: Selected results. Left two columns: data compared with expected back- (∗) ′ ′ ground for (Row 1) m4ℓ from the H→ZZ →ℓℓℓ ℓ channel for 100 GeV < mH < (∗) 250 GeV and for the complete range [3]; (Row 2) mT from the H→W W →ℓνℓν channel with 0/1 jets [4]; (Row 3) mbb from the ZH and WH channels [5]; (Row 4) mττ from the H→τlepτlep +1j and H→τlepτhad +0/1j channels [6]. An example of the expected signal is also shown (scaled up for some channels to make it more visible). Right: Corresponding exclusion plots, showing the expected (dashed) and observed (solid) 95% CL upper limits on SM Higgs boson production. The dark (green) and light (yellow) bands indicate the expected limits with ±1σ and ±2σ fluctuations, respectively.

3 analysis is divided into subsamples with 0, 1, and ≥ 2 jets, as the background compo- sition is quite different for these cases: for the 0-jet case, the background is dominated by the W W and Z + jets processes, while for the 2-jet case, tt dominates. As there are two neutrinos in the final state, the Higgs boson mass cannot be reconstructed; the transverse mass mT is used instead. Results are shown in Figure 2, Row 2. No excess is seen, and the mH region (133–261) GeV is excluded at 95% CL.

2.4 Other low-mH channels Two additional channels are sensitive to a low-mass Higgs boson. In the V (H→bb) channel [5], one searches for a Higgs boson produced in association with a vector , with the Higgs boson decaying into a bb pair. The event selection for this channel requires a gauge boson decay to leptons/neutrinos (one of W →ℓν, Z→ℓℓ, or Z→νν) and exactly two b-tagged jets. The analysis is subdivided in bins of pT(V ). The H→ττ analysis [6] is divided into 2ℓ4ν, ℓτhad3ν, and 2τhad2ν subchannels; these are then further subdivided according to the number of extra jets in the event. Since there are multiple neutrinos in the final state, the invariant mass mττ is estimated either using the collinear approximation (assuming the neutrino to be collinear with the visible decay products) or the “missing mass calculator” technique (incorporating the probability distribution for the opening angle in τ decays). Selected results are shown in Figure 2, Rows 3 and 4. Exclusion limits for these channels range from about 2.5 to 12 times the SM Higgs boson production cross section over the mH range (100–150) GeV.

2.5 Other high-mH (diboson) channels The remaining diboson channels are sensitive to higher Higgs boson masses. For miss the H→ZZ→ℓℓνν channel [7], events with a Z→ℓℓ decay and large ET are se- lected. Different selection requirements are made for mH above or below 280 GeV. For H→ZZ→ℓℓqq [8], events are selected with one Z→ℓℓ decay, one Z→jj decay, miss and small ET . As the fraction of heavy-flavour in the final state is quite different for signal and background, this channel is divided into “tagged” (with two identified b-jets) and “untagged” subchannels. The principal backgrounds for these two chan- nels are Z + jets, tt, and ZZ. For the H→W W →ℓνqq channel [9], events are selected miss with an isolated lepton, large ET , and a W →jj decay. This analysis is divided into subchannels according to the number of additional jets in the event. Results are shown in Figure 3. The ZZ→ℓℓνν channel excludes at 95% CL the mH range (320–560) GeV, while the ZZ→ℓℓqq channel excludes the ranges (300– 322) GeV and (353–410) GeV. Exclusions from the W W →ℓνqq channel are 2–10 times the SM cross section.

4 8 ATLAS 120 90 -1 ATLAS Data 2011 ATLAS H→WW→lνqq+0j H→ZZ→ll νν s = 7 TeV, ∫ Ldt = 4.7 fb 7 80 H→ZZ→llbb Total background 100 6 Data 2011 70 Data 2011 mH=350 GeV, 1 x SM 60 5 80 Total background

Events / 50 GeV Total background Events / 18 GeV Events / 10 GeV s = 7 TeV, Ldt = 4.7 fb-1 50 4 ∫ mH=350 GeV, 1 x SM m =350 GeV, 1 x SM 60 40 H 3 s = 7 TeV, ∫ Ldt = 4.7 fb-1 30 40 2 20 20 10 1 0 0 200 300 400 500 600 700 800 900 200 300 400 500 600 700 800 300 400 500 600 700 800 900 1000

mT [GeV] mllbb [GeV] mlνjj [GeV]

102 SM SM SM σ σ ATLAS 2011 Observed σ / / 5 10 ATLAS Observed / Expected σ σ Expected σ → → νν Expected H→ ZZ→ llqq ± 1σ H ZZ ll ± 1 σ ± 1σ ∫Ldt = 4.7 fb-1, =7 TeVs ± σ ± 2σ 4 -1 8 2 ∫Ldt=4.7fb , =7TeVs ± 2σ 10 Observed

3 6 95% CL limit on on limit CL 95% 95% CL limit on on limit CL 95% 95% C.L. limit on 2 4 1

1 =7 TeVs 2 ATLAS -1 H+0/1/2j, H→ WW→ lν jj ∫ L dt=4.7 fb 0 10-1 200 250 300 350 400 450 500 550 600 0 300 350 400 450 500 550 600 200 300 400 500 600 MH[GeV] mH [GeV] mH [GeV] Figure 3: Selected results from the remaining diboson channels. Top: data and ex- pected background for H→ZZ→ℓℓνν, showing mT [7] (left); H→ZZ→ℓℓqq, showing mℓℓbb for the high-mH , tagged subchannel [8] (middle); and H→W W →ℓνqq, show- ing mℓνjj for the 0-jet subchannel [9] (right). The expectations for a Higgs boson of mH = 350 GeV are also shown. Bottom: Corresponding exclusion plots (see Fig- ure 2).

3 Combination of all channels

For the final results, all these channels are combined [10]. Systematic uncertainties are taken to be either 100% correlated or uncorrelated between channels. The results are shown in Figure 4. The mH ranges (111.4–116.6) GeV, (119.4–122.1) GeV, and (129.2–541) GeV are excluded at 95% CL, and the mH range (130.7–506) GeV is excluded at 99%. An excess is seen around mH = 126 GeV, with a local significance of 3.0σ. The global significance is approximately 15% over the full range (100–600) GeV and (5–7)% over the range (110–146) GeV, corresponding to the range at low-mH not excluded by the previous LHC SM Higgs boson combined search at 99% CL [11]. If the excess is interpreted as a SM Higgs boson, the corresponding cross section ratio at mH = 126 GeV is µ = σobs/σSM = 1.1 ± 0.4, consistent with the SM value of 1.0. The excess is observed only in the H→γγ and H→ℓℓℓ′ℓ′ channels; however, none of the other channels are inconsistent with a SM Higgs boson at this mass.

4 Summary

From the 2011 data, ATLAS excludes at 95% CL the mH range (111.4–541) GeV, except for the regions (116.6–119.4) GeV and (122.1–129.2) GeV. An excess is seen ′ ′ in the H→γγ and H→ℓℓℓ ℓ channels at mH ≈ 126 GeV which is consistent with a

5 10 0 SM ATLAS 2011 2011 Data SM ATLAS 2011 2011 Data ∫ L dt ~ 4.6-4.9 fb-1 = 7 TeVs

σ σ ATLAS 2011 / /

σ σ Exp. Comb. Exp. H → ZZ → llll Exp. H → bb Obs. -1 10 Obs. -1

Ldt = 4.6-4.9 fb Local p Exp. ∫ Exp. ∫ Ldt = 4.6-4.9 fb 10 Obs. Comb. Obs. H → ZZ → llll Obs. H → bb ±1 σ ±1 σ Exp. H → γγ Exp. H → WW → lνlν Exp. H → ττ s = 7 TeV ±2 σ ±2 σ s = 7 TeV Obs. H → γγ Obs. H → WW → lνlν Obs. H → ττ 1 1

1 σ 95% CL Limit on on Limit CL 95% on Limit CL 95% 1 10-1

2 σ 10-2 10-1 CLs Limits CLs Limits -1 -3 3 σ 100 200 300 400 500 600 10 110 115 120 125 130 135 140 145 150 10 110 115 120 125 130 135 140 145 150 mH [GeV] mH [GeV] mH [GeV]

Figure 4: Left, middle: Combined exclusion limits for the full mass range and for mH < 150 GeV [10] (see Figure 2). Right: Local probability p0 for the background to fluctuate to the observed number of events or higher, by channel, for mH < 150 GeV. Dashed lines show the expected median local p0 for the signal hypotheses at mH .

SM Higgs boson. The chance of this being due to a background fluctuation is 15% over the full mass range, or (5–7)% over (110–146) GeV.

This work is supported in part by the U.S. Department of Energy under contract DE-AC02-98CH10886 with Brookhaven National Laboratory.

References

[1] ATLAS Collaboration, JINST 3, S08003 (2008). [2] ATLAS Collaboration, Phys. Rev. Lett. 108, 111803 (2012) [arXiv:1202.1414]. [3] ATLAS Collaboration, Phys. Lett. B 710, 383 (2012) [arXiv:1202.1415]. [4] ATLAS Collaboration, arXiv:1206.0756. [5] ATLAS Collaboration, arXiv:1207.0210. [6] ATLAS Collaboration, arXiv:1206.5971. [7] ATLAS Collaboration, arXiv:1205.6744. [8] ATLAS Collaboration, arXiv:1206.2443. [9] ATLAS Collaboration, arXiv:1206.6074. [10] ATLAS Collaboration, Phys. Rev. D 86, 032003 (2012) [arXiv:1207.0319]. [11] ATLAS and CMS Collaborations, ATLAS-CONF-2011-157, CMS-PAS-HIG-11- 023 (2011) [https://cdsweb.cern.ch/record/1399599/].

6