First Physics at the LHC M Cobal 1
First Physics at the LHC M. Cobal (1), E. Migliore (2) University of Udine and INFN Trieste (2) University of Torino and INFN Torino (1) IV Workshop Italia ATLAS-CMS Bologna, 23 -25 November 2006 IV Workshop Italiano ATLAS-CMS Thanks to: G. Polesello F. Giannotti R. Chierici R. Tenchini P. Bartalini
2008 should look something like… Hardware commissioning to 7 Te. V Machine Checkout 1 month Commissioning with beam 2 months l a on o Pr i s i v Pilot Physics 1 month Reach 1031 Running at 75 ns L~ 1032 cm-2 s-1 ~ 3 months of running +some optimism ~ 1 fb-1 IV Workshop Italiano ATLAS-CMS
LHC: physics roadmap Understand/calibrate detector and trigger in situ using “candles” samples e. g. - Z ee, tracker, ECAL, muon chamber calibration and alignment, etc. - tt bl bjj jet scale from W jj, b-tag performance, etc. Understand basic SM physics at s = 14 Te. V § measure cross-sections for e. g. minimum bias, W, Z, tt, QCD jets (to ~20 %), § start to tune Monte Carlo § measure top mass give feedback on detector performance Note : statistical error negligible with O(10 pb-1) Prepare the road to discovery: § measure backgrounds to New Physics : e. g. tt and W/Z+ jets § look at specific “control samples” for the individual channels: e. g. ttjj with j b “calibrates” ttbb irreducible background to tt. H ttbb Look for New Physics potentially accessible in first year(s) e. g. Z’, SUSY, Higgs ? IV Workshop Italiano ATLAS-CMS
How many events at the beginning ? Assumed selection efficiency: W l , Z ll : 20% tt l +X : 1. 5% (no b-tag, inside mass bin) similar statistics to CDF, D 0 today + lots of minimum-bias and jets (107 events in 2 weeks of data taking if 20% of trigger bandwidth allocated) 10 pb-1 1 month at 1030 and < 2 weeks at 1031, =50% 100 pb-1 few days at 1032 , =50% 1 fb-1 IV Workshop Italiano ATLAS-CMS
Which detector performance on day one ? Based on detector construction quality, test-beam results, cosmics, simulation Expected performance day 1 Physics samples to improve ECAL uniformity e/ scale ~ 1% ~2% Minimum-bias, Z ee Z ee HCAL uniformity Jet scale events ~3% < 10% Single pions, QCD jets Z ( ll) +1 j, W jj in tt Tracking alignment 20(100)-200 m in R ? Generic tracks, isolated , Z m Ultimate statistical precision achievable after few weeks of operation. Then face systematics…. E. g. : tracker alignment : 100 m (1 month) 20 m (4 months) 5 m (1 year) ? IV Workshop Italiano ATLAS-CMS
What we will know @ LHC start? W, Z cross-sections: to 3 -4% (NNLO calculation dominated by PDF) <Nch> at =0 for generic pp collisions (minimum bias) LHC ? Transverse < Nchg > tt cross-section to ~7% (NLO+PDF) Lot of progress with NLO matrix element MC interfaced to parton shower MC (MC@ NLO, Alp. Gen, . . ) PYTHIA 6. 214 - tuned LHC PHOJET 1. 12 x 3 — Alp. Gen Candidate to very early measurement: tuning of MC models understand basics of pp collisions, occupancy, pile-up, … IV Workshop Italiano ATLAS-CMS x 1. 5
Minimum Bias ¡ ¡ Non-single diffractive evts, ≈ 60 -70 mb Soft interactions n Low PT, low Multiplicity. § § § n Soft tracks: p. Tpeak~250 Me. V Approx flat distribution in to | |~3 and in Nch~30; | |<2. 5 Rate: R~700 k. Hz @ L=1031 cm-2 s-1, For d. N/d require ~10 k ¡ What we would observe with a fully inclusive detector/trigger. ¡ Several MB interactions can take place in a single beam crossing. n MB seen if “interesting” triggered interaction also produced. n Pile-up effect. n Tracking detectors help to separate the different primary vertices. n Possible overlap of clusters in calorimeters. Need energy flow. IV Workshop Italiano ATLAS-CMS
Initial Tracking & Alignment n n n Very first alignment will be based on: ¡ Mechanical precision ¡ Detailed survey data ¡ Cosmic data ¡ Minimum bias events and inclusive bb Studies indicate good efficiencies after initial alignment ¡ ~ 80% down to PT = 500 Me. V ¡ Precision will need Zs and resonances to fix energy scales, constrain twists, etc. Even lower PT accessible with reduced tracking ? ¡ PT = 400 Me. V - tracks reach end of TRT ¡ PT = 150 Me. V - tracks reach last SCT layer ¡ PT = 50 Me. V - tracks reach all Pixel layers p. T (Me. V) 150 Me. V IV Workshop Italiano ATLAS-CMS
Underlying Event ¡ All the activity from a single particle-particle interaction on top of the “interesting” process. High PT scatter n Initial State Radiation (ISR). Beam remnants n Final State Radiation (FSR). ISR n Spectators. n … Multiple Interactions ? (These models are certainly very successful!). ¡ The UE is correlated to its “interesting” process. n Share the same primary vertex. n Events with high PT jets or heavy particles have more underlying activity Pedestal effect. ¡ UE ≠ MB but some aspects & concepts are similar. n Phenomenological study of Multiplicity & PT of charged tracks. IV Workshop Italiano ATLAS-CMS
UE: measurement plan at the LHC From charged jet using MB and jet triggers) Topological structure of p-p collision from charged tracks The leading Ch_jet 1 defines a direction in the plane The transverse region is particularly sensitive to the UE Main observables: + d. N/d df, charged density + d(PTsum)/d df, energy density From D-Y muon pair production (using muon triggers) observables are the same but defined in all the plane (after removing the pairs everything else is UE) IV Workshop Italiano ATLAS-CMS
UE with charged jets d. Nch/d df VS PT_ch_jet 1 MC MB JET 60 JET 120 PT>0. 9, | |<1 Main observables: - d. N/d df, charged density - d(PTsum)/d df, energy density Njets > 1, |ηjet| < 2. 5, ETjet >10 Ge. V, PT_ch_jet 1 Good RECO/MC agreement in shape Differences compatible with the expected corrections from charged jet PT calibration, charged tracks innefficiencies and fake rate RECO/MC Differences absorb in the ratio, no need to apply corrections! Emphasis on the reconstruction of soft tracks… IV Workshop Italiano ATLAS-CMS
UE with Drell-Yan d. PTsum/d df d. N/d df M(m, m) Ratio PT>0. 9 Ge. V/PT>0. 5 Ge. V (PT tracks threshold) Ratio observables are sensitive to differences between models !!! IV Workshop Italiano ATLAS-CMS
Production of W and Z boson n n Large W (Z) cross section: 10 nb (1 nb) and clean leptonic signatures Compare to theo. prediction or assume prediction and use to measure luminosity ¡ Example : uncertainties with 1 fb-1 in the muon channel in detector fiducial volume IV Workshop Italiano ATLAS-CMS
PDFs: LHC Kinematic regime for LHC much broader than currently explored Test of QCD: n Test DGLAP evolution at small x: q Is NLO DGLAP evolution sufficient at so small x ? q Are higher orders important? n Improve information of high x gluon distribution At Te. V scale New Physics ’s predictions are dominated by high-x gluon uncertainty (not sufficiently well constrained by PDF fits) At the EW scale theoretical predictions for LHC are dominated by low-x gluon uncertainty (i. e. W and Z masses) => see later slides How can we constrain PDF’s at LHC? IV Workshop Italiano ATLAS-CMS
W en rapidity distributions HERWIG MC Simulations with NLO Corrections W production over |y|<2. 5 at LHC involves 10 -4 < x 1, 2 < 0. 1 region dominated by g qq Generator Level e- rapidity CTEQ 61 MRST 02 ZEUS-S e+ rapidity CTEQ 61 MRST 02 ZEUS-S Error boxes are the Full PDF Uncertainties ATLAS Detector Level with sel. cuts At y=0 the total PDF uncertainty is ~ ± 5. 2% from ZEUS-S ~ ± 3. 6% from MRST 01 E ~ ± 8. 7% from CTEQ 6. 1 M ZEUS-S to MRST 01 E central value difference ~5% GOAL: ZEUS-S to CTEQ 6. 1 central value difference ~3. 5% syst. exp. error ~4% IV Workshop Italiano ATLAS-CMS
Including ATLAS data on PDF fits Generate 1 M “data” sample with CTEQ 6. 1 PDF through ATLFAST detector simulation and then include this pseudo-data (with imposed 4% error) in the global ZEUS PDF fit (with Det. ->Gen. level correction). Central value of ZEUS-PDF prediction shifts and uncertainty is reduced: ZEUS-PDF BEFORE including W data ZEUS-PDF AFTER including W data e+ CTEQ 6. 1 pseudo-data | | low-x gluon shape parameter λ, xg(x) ~ x –λ BEFORE λ = -0. 199 ± 0. 046 AFTER λ = -0. 181 ± 0. 030 | | 40% error reduction In few day stat. of LHC at low Luminosity Systematics (e. g. e acceptance vs ) can be controlled to few % with Z ee (~ 30000 events for 100 pb-1) IV Workshop Italiano ATLAS-CMS
Top physics in the early phase The LHC will be a top-factory ! NLO~830 pb : 2 tt events per second ! more than 10 million tt /year §Measure total ttbar cross section: • test of p. QCD calculations (predicted at ~ 10%) • sensitive to top mass § Measure differential cross sections • sensitive to new physics § Make initial direct measurement of top mass § Measure single top production (t-channel) IV Workshop Italiano ATLAS-CMS
Top physics during commissioning § Several months to achieve pixel alignment § Study separation of top from background without b-tagging • Use high multiplicity in final states • High Pt cuts to clean sample • Use kinematical features W CANDIDATE § Even with a 5% efficiency 10 evts/hour at 1033 Hadronic top: Three jets with highest PT TOP CANDIDATE W boson: Two jets in hadronic top with highest PT in reconstructed jjj C. M. frame IV Workshop Italiano ATLAS-CMS
Top physics during commissioning m (top jjj) m(W jj) L=300 pb-1 S |mjj-m. W| < 10 Ge. V S/B = 1. 77 B S/B = 0. 45 S : MC @ NLO B : Alp. Gen x 2 to account for W+3, 5 partons (pessimistic) Expect ~ 100 events inside mass peak with only 300 pb-1 top signal observable in early days with no b-tagging and simple analysis W+jets background can be understood with MC+data (Z+jets) IV Workshop Italiano ATLAS-CMS
Luminosity (pb-1) W+jets x 4 W+jets x 8 Siginficance ( ) Luminosity (pb-1) W+jets x 2 Nominal W+jets Siginficance ( ) Fitted #signal events Top signal significance vs luminosity Luminosity IV Workshop Italiano ATLAS-CMS (pb-1)
What can you do with early tops? Calibrate light jet energy scale - impose PDG value of the W mass (precision < 1%) Estimate/calibration b-tagging e - From data (precision ~ 5%) - Study b-tag (performance) in complex events Study lepton trigger Calibrate missing transverse energy - use W mass constraint in the event - range 50 Ge. V < p T < 200 Ge. V Estimate (accuracy ~20%) of mt and tt. Events Use W boson mass to enhance purity Miscalibrated detector or escaping ‘new’ particle Perfect detector Missing ET (Ge. V) IV Workshop Italiano ATLAS-CMS
Systematic effects for top physics § Almost all SM measurements at LHC dominated by systematic errors. § Can be divided into instrumental and from theory/modeling § Dominant instrumental uncertainties for top physics: Luminosity: Reasonable goal is 3 -5% measure number of interactions/bunch crossing (HF) and (pp) (TOTEM) Reconstruction related: Jet energy scale need calibrated calorimetry (beam tests, MB, single particles, Z, W…) need jet energy calibration to a few % (with Z( )+jet) need excellent energy flow (association tracker+calo+muon system) b-tagging efficiency+fake rate use tt for calibration: to 4 -5% with 10/fb Lepton identification and energy scale use Z, other mesons. Less crucial than for the W mass measurement Theory related systematics are as important as instrumental ones ! IV Workshop Italiano ATLAS-CMS
Learning from data We model most of our description of reality at the LHC: Need to quote a confidence on the description of our simulations Need to avoid non realistic scenarios (ex: FSR OFF) Need to avoid to double count errors § pdfs PDFs MEs PS § hard scattered partons process description (signal AND backgrounds) scale dependency § final state radiation vary QCD and Q 2 max consistently ? Jets UE, MPI constrain using LHC data § hadronization b+light jets: is LEP good enough? vary QCD and Q 2 max consistently ? § minimum bias and underlying event energy dependent. Do our own tuning ! UE fragmentation § initial state radiation PDF must tuning fragmentation tuning UE radiation LHC learn the best way to use its own data to constrain models IV Workshop Italiano ATLAS-CMS
X-section of tt semileptonics at 1 fb-1 Single lepton selection Lepton with cuts on ET and Missing ET>40 Ge. V Hadronic activity (4 jets) b-tagging of 2 jets IV Workshop Italiano ATLAS-CMS
Top mass measurement at fb-1 ttbar semileptonics • Should be able to measure top mass at ~ 1% in both dileptonic and semileptonic channel • Need control of the jet energy scale ! • Larger error ~2 -3% in the hadronic channel IV Workshop Italiano ATLAS-CMS
Conclusions § LHC startup will require a long period of development and understanding § With first data measure § detector performance in situ physics § particle multiplicity in minimum bias § top signal with ~ 30 pb-1 § (tt) to 20% and Mtop to 7 -10 Ge. V with 100 pb-1 ? § PDF (low-x gluons !) with W/Z (O(100) pb-1 ? ) § first tuning of MC (MB, UE, tt, W/Z+jets, QCD jets, …) § Goal is to arrive @ high statistics (few fb-1) data-taking ready to go for early discovery physics § BUT…. as soon as interactions @ 14 Te. V happen interesting and new physics will appear in the data. Surprises? …. see discussion of Ernesto! IV Workshop Italiano ATLAS-CMS
IDEE PER LA DISCUSSIONE capacità di fare misure accurate/difficili n possibilità di fare scoperte n … con 100 -1000 pb-1 IV Workshop Italiano ATLAS-CMS
Cross-sections and rates At luminosity 1032 cm-2 s-1 Inelastic: n bb production: n W ℓ : n Z ℓℓ: n tt production: n 107 Hz 104 Hz 1 Hz 0. 1 Hz IV Workshop Italiano ATLAS-CMS
b-physics with 100 pb-1: control channels Sensitive tests of understanding of detector properties n ¡ ¡ ATLAS ¡ n learn how to do b-physics in exclusive channels w/o particle-ID test of tracking: alignment, material budget, B field test of trigger: L 1+HLT Statistics Stat. error on lifetime 100 pb-1 World today (stat+syst) 17000 1. 5 % 0. 4 % Reference/control channel (ie. B 0→ μ+μ-) Control channel for masses, lifetimes B+ B+→J/y K+ B 0→ J/y K 0* 8700 2. 2 % 0. 5 % Bs Bs→ J/y f 900 6% 2% b b→ J/y 260 8% 5% Need to be measured at low luminosity to fully exploit the high luminosity LHC run IV Workshop Italiano ATLAS-CMS
Bs J/ n Sensitivity to s/ s (SM 10%) in the “no-tag” mode n Bs J/ + -K+K- (SM: σ 75 pb) ¡ Trigger selection: n n L 1: di-μ (p. T>3 Ge. V) HLT: J/ and reconstruction using only 5 Tracker hits and w/o μ-ID ¡ ¡ ¡ CMS | M(J/ ) |<150 Me. V (L 2) Mass resolution: σ(J/ ) 50 Me. V Transverse decay length (Lxy/ >3) | M ( ) |<20 Me. V, | M(Bs )|<190 Me. V (L 3) Mass resolution: σ( ) 5 Me. V, σ(Bs) 60 Me. V sig 20% bgd=10 -5 HLT rate(sig+bgd): 0. 1 Hz @ 1033 cm-2 s-1, 15000 evts at 1 fb-1 Offline selection: n n μ-ID+Full Tracker reconstruction Kinematic fit ¡ sig 75% (Offline/HLT) ¡ Mass resolution: σ(Bs) 15 Me. V IV Workshop Italiano ATLAS-CMS
Bs J/ CMS Plots for 30 fb-1 Offline HLT n Likelihood for s/ s ¡ ¡ diff. decay rate as a function of (cos , , cos ; t) [PLB 369(1996) 144] + K+ Accurate knowledge of (t) “c distorsion” (ie. track seeds at HLT level) n n ¡ determine (t) from B 0 J/ K 0* + -K+ K/ mis-id: narrow MBs range (± 36 Me. V) low stat. included in the likelihood with higher stat. 20% accuracy at 1. 5 fb-1 IV Workshop Italiano ATLAS-CMS
Bs μ+μn n ¡ 8 10 -8 Trigger n L 1: 1 -μ (di-μ) 1031(1033) p. T>6 Ge. V L 2: di-μ (p. T>6 Ge. V) μ/tracker matching mass cut Offline n n n ATLAS Sensitivity to new physics MSSM: BR ~ (tan ) 6 Selection: p. T> 6 Ge. V, | |<2. 5, Rμμ<0. 9 m(μμ)=MBs+140 -70 Me. V Isolation Transverse decay length (Lxy/ >11) + pointing (<1°) flight direction SM: 3. 5 10 -9 90% upper limit S. Sivoklokov ICHEP 06 Affected by bgd ¡ Bs μ+ - (BR 10 -4) with / mis-id x 10 larger than SM predictions for Bs μ+μIV Workshop Italiano ATLAS-CMS
MW with “Scaled Observables Method” n R(x) different detector acceptance to leptons from W and Z CMS PRD 57 (1998) 4433 n Exploit large Z→ℓℓ statistics available at LHC (104 -105 evts with 100 pb-1) measurement of MW from the lepton p. T spectrum (no MT No MET) n Scaled lepton p. T “template” MT IV Workshop Italiano ATLAS-CMS
Diffractive physics Double Pomeron exchange: Single diffraction: X n n n 2 gluon exchange with vacuum quantum numbers “Pomeron” SD~ 15 mb DPE ~ 1 mb 1032 cm-2 s-1 → no/low pile-up → selection of events based on Large Rapidity Gap between scattered proton and X Measurement of SD and DPE cross-sections in presence of hard scale (dijets, vector bosons, heavy quarks) ¡ Test of the scale where the factorization is violated ¡ Handle for understanding parton-parton interactions (UE) IV Workshop Italiano ATLAS-CMS X
Di-lepton resonances n CMS Z’→ μ+ μ¡ ¡ ¡ single-μ OR di-μ trigger opposite charge μ brems. accounted 100 pb-1 Z’: M=1 Te. V σ*BR=150 fb (σ*BR=370 fb with /Z/Z’ interf. ) initial alignment RMS± 30% 120 Ge. V DY ¡ ¡ ± 1σ MC truth Unbinned max likelihood for bgd and sig+bgd Improve alignment: ie. CMS long term alignment (1 fb-1) RMS± 30% 60 Ge. V IV Workshop Italiano ATLAS-CMS
Z’ discovery reach (5 ) ATLAS CMS IV Workshop Italiano ATLAS-CMS
Di-jet resonances n Physics interest in the high mass tail ¡ ¡ ¡ W’, Z’ excited quarks RS gravitons n Limits from CDF and DØ up to 1 Te. V n Narrow resonances ¡ n CMS |jet |<1 dσ/d. M (pb/Ge. V)=exp (- M/486 Ge. V) 1 10 100 pb-1 Luminosity for 10 evts above threshold ie. Z’: /M 1 -3% Crucial knowledge of the physics of jets ¡ ¡ absolute jet energy scale modeling of jets IV Workshop Italiano ATLAS-CMS
Di-jet azimuthal decorrelation ATLAS Dfdijet = p Dfdijet ~ 2 p/3 Dfdijet p p/2 Dfdijet 2 p/3 dijet sensitive to higher orders QCD radiation w/o measuring the 3 rd or the 4 th jet IV Workshop Italiano ATLAS-CMS
Di-jet azimuthal decorrelation n Di-jet event selection: ¡ ¡ ATLAS Cone jet algorithm ( R=0. 7) Njets = 2 |ηjet| < 0. 5 ETjet #2 > 80 Ge. V 300 < ETMAX < 600 Ge. V 600 < ETMAX < 1200 Ge. V IV Workshop Italiano ATLAS-CMS
F. Gianotti ICHEP 06 The Higgs Boson IV Workshop Italiano ATLAS-CMS
H WW(*) ℓ ℓ CMS n σ 2. 4 pb n No narrow m(WW) peak ¡ exploit the spin-0 of SM Higgs W+ W- + n - Counting exp. in (ℓℓ) distribution ¡ n MH=165 Ge. V It requires low and well known background (gg tt (tt and qq WW) Selection: ¡ ¡ 2 isolated opposite charge ℓ S/B=1. 66 Jet veto (ET>15 Ge. V and | |<2. 5) σ/σ bgd=20% (syst) MET>50 Ge. V ℓ kinem. *: | |<2, 12 <m(ℓℓ)<40 Ge. V, (ℓℓ)<45˚ 30< p. Tmax< 55 Ge. V, 25< p. Tmin< 55 Ge. V * Optimized for 1 fb-1 IV Workshop Italiano ATLAS-CMS
SUSY with 100 -1000 pb-1 ATLAS n If s-particles have the same couplings as SM particles the production is dominated by s-quark and gluinos (if light enough) Dominant signatures: Jets and MET Hard at the beginning! DØ (V. Shary CALOR 04) IV Workshop Italiano ATLAS-CMS
Low Mass SUSY MET m(ll) High p. T jets n “Early discovery” selection ¡ ¡ ¡ 2 isolated leptons p. T>10 Ge. V 2 4 jets: p. Tjet 1>100 Ge. V, …, p. Tjet. N>50 Ge. V MET>200 Ge. V SM bgd: tt ( 10 -4), W/Z+jets ( 10 -5), WW/ZZ+jets n Same. Flavour. Opposite. Sign (ie. μ+μ-+ e+e- from ° 2 chain) ¡ n Bgd subtraction with Different. Flavour. Opposite. Sign (ie. eμ) Same. Flavour. Same. Sign (ie. μ±μ± + e±e± from two ±) ¡ No tt and DY bgd IV Workshop Italiano ATLAS-CMS
Low Mass SUSY: di-lepton endpoint CMS ATLAS 1 fb-1 Prelim. CMS “LM 1” m. SUGRA point n m(~g)= 600 Ge. V n = 50 pb ATLAS “SU 3” m. SUGRA point n m(~g)= 720 Ge. V n = 19 pb IV Workshop Italiano ATLAS-CMS
Outlook n Se siamo “bravi”… ¡ n Possibilità di fare misure precise Mtop, MW, canali esclusivi del b … ed anche “fortunati” ¡ Canali dileptonici (ma dopo avere capito tt !): n n Z’ Low Mass SUSY Higgs (ma solo per MH=165 Ge. V) Credits: M. Arneodo, P. Bartalini, U. De Sanctis, F. Gianotti, T. Lari, G. Polesello, R. Tenchini IV Workshop Italiano ATLAS-CMS
- Slides: 45