Precision measurements at the LHC ATLAS and CMS

Precision measurements at the LHC (ATLAS and CMS) Loops and Legs in Quantum Field Theory Bastei/Konigstein, 12 April 2000 Monica Pepe Altarelli (INFN Frascati) 1

LHC luminosity • Lpeak 1033 cm-2 s-1 2005 -2008 (“low L”) 2009 (“ high L”) ultimate (beam-beam limited) • L dt 10 fb-1 L dt 100 fb-1 per year at low L per year at high L Lpeak 1034 cm-2 s-1 Lpeak 2. 5 1034 cm-2 s-1 • Bunch crossing : 25 ns • ~ 20 minimum-bias per crossing high L “pile-up” ~ 700 charged particles p. T > 150 Me. V per crossing detector speed (t 50 ns) radiation hardness 2

A few numbers …. . 1033 cm-2 s-1 Process s Events/s W e 15 nb 15 108 Z ee 1. 5 nb 1. 5 107 800 pb 0. 8 107 500 mb 105 1012 1 pb 0. 001 104 100 nb 102 109 Events/year (m =1 Te. V) H (m=0. 8 Te. V) QCD jets p. T > 200 Ge. V LHC is a B-factory, top factory, W/Z factory, Higgs factory, SUSY factory, etc. Mass reach: up to Te. V Precision measurements dominated by systematics 3

Energy dependence of some characteristic cross-sections ssig/stot 10 -12 CMS 100 fb-1 m. H=130 Ge. V 4

3 crucial parameters for precision measurements Uncertainties on : • Absolute luminosity : goal < 5% Main tools: machine, optical theorem, rate of known processes (W, Z, QED pp ) • energy scale : goal Main tool: Z 1‰ 0. 2‰ most cases W mass (1 event/ /s at low L) close to in mass to m. W, mh 1‰ achieved by CDF/D 0 despite small Z sample • Jet energy scale : goal 1% (mtop, SUSY) limited not only by calorimeter calibration but also by physics (fragmentation, gluon radiation, etc) Main tools : Z + 1 jet W jj 4% at Tevatron (Z ) from top decays tt Wb b -1 (10 events/s low L) Wb jjb Requirements: tracker material to 1%, overall alignment to 0. 1 mm, overall B-field to 0. 1‰, muon E-loss in calorimeters to 0. 25%, etc. 5

Measurements discussed here: W mass Drell-Yan production of lepton pairs Triple Gauge Couplings Top physics Higgs SUSY 6

W mass Year 2005 : Dm. W < 30 Me. V LEP 2+Tevatron Motivation to improve: f(m 2 top, log m. H) Dm. W 0. 7 10 -2 Dmtop to get similar errors Dmtop 2 Ge. V (LHC) requires Dm. W 15 Me. V -- if/when Higgs found: check consistency of theory -- constrains m. H to 25% 7

Dependence of MW on mt in SM and MSSM EW precision observables may be useful to distinguish between different models as candidates for the EW theory 8

Main method : transverse mass l = e, m (from transverse momentum imbalance in calorimeters) Edge of m. TW distribution sensitive to m. W smeared by : W width, detector resolution, pile-up ( technique probably limited to low L) m. W= 79. 8 Ge. V m. W= 80. 3 Ge. V m. TW (Ge. V) 9

W production and selection 300 106 events produced /exp/year at low L Selection cuts : • isolated charged lepton (e, m) with PT>25 Ge. V, |h|<2. 4 • Etmiss> 25 Ge. V • No jets with PT>30 Ge. V • Recoil momentum < 20 Ge. V Reject Ws at high PT: • worse m. T resolution • higher QCD background Expected efficiency ~ 20% 60 106 well measured W l (l =e, m) per experiment, per year at low L (~ 50 times more than Tevatron Run II) 10

Uncertainties on MW • Statistical error < 2 Me. V • Systematic error from Monte Carlo modelling of the data (physics & detector) Physics: p. TW and W pdfs W width Radiative decays Background Detector: Lepton E, p scale Lepton E, p resolution Recoil modelling Lepton ID cuts • Most uncertainties (lepton scale, detector resolution, p. TW, etc. ) controlled in situ with Z sample. • High statistics control sample : ~ 6 ´ 106 Z decays in one year of low L after all selection cuts ( factor 50 larger than event samples from Tevatron Run II) • Z close in mass to W small extrapolation 11

12 by combining both expts and both channels Recoil from Z evts with p. TZ p. TW

Comparison between Theory & Experiment LHC Stringent bound on MH to be confronted (hopefully!) with directly measured value 13

Drell-Yan production of lepton pairs Distinctive experimental signature: pair of isolated leptons with opposite charge Experimental backgrounds low (W+W-, t+t-, cc, bb, tt, fakes, etc. ) One channel/exp after h, p. T cuts Measure: Total cross section s(y, M) Forward-Backward Asymmetry AFB(y, M) y: rapidity of lepton pair M : invariant mass of lepton pair 14

• Main exp. systematics on s from L (known at few % level) • pdfs constrained experimentally by h distributions of leptons from W, Z decays • Precise measurement of s and AFB requires good knowledge of EW radiative corrections: Rel. exp. error on sll Complete one-loop (Baur, Brein, Hollik Schappacher, Wackeroth) M ll (Ge. V) Corrections can be probed up to 2 Te. V 15

From FB asymmetry AFB in di-lepton production near the Z pole sin 2 qefflept (M 2 Z) = 0. 23148± 0. 00017 LEP+SLD Measurement of AFB requires tag of q, q directions Only q can be valence quark Þ on average higher momentum wrt sea q AFB signed according to sign of yll • pe. T> 20 Ge. V • 85. 2<Mee<97. 2 Ge. V 2 • one e within | h| <2. 5 AFB(%) If very forward e± tagged • other e identified in forward calorimeters 2. 5< | h| <4. 9 significance of measurement increased For moderate jet rejection (~102) in fwd calo. dstatsin 2 qefflept (M 2 Z) 1. 4 10 -4 reachable (one exp, e channel, 100 fb-1) WHAT ABOUT SYSTEMATICS? Main effect: uncertainty on pdfs Agreement among different pdfs tested at 1% level (statistical power of the study). Needs another factor of 10. New measurements from HERA/Tevatron/LHC will improve understanding of pdfs 16 yll

17

Triple Gauge couplings W W , Z W W • Probe non-Abelian structure of SU(2) x U(1) and sensitive to New Physics • Year 2005 : g 1 Z, lg, kg, l. Z, k. Z known to better than 10 -2 from LEP 2+Tevatron • Some anomalous contributions increase with s high sensitivity at LHC • W l WZ l ll WW l l large tt background • Sensitivity from : -- cross-section measurements: l-type increase with s -- p. T, distributions: sensitive to different TGC’s, constrain k-type Need to know precisely Wg, WZ production from theory 18

WZ 30 fb-1 SM Dg 1 Z=0. 05 W l =0. 05 Systematics (background, NLO, pdf) under study (should be small, concentrated at low p. T) 19

L=10 Te. V Coupling 95% CL Dg 1 Z lg l. Z Dkg Dk. Z 0. 008 0. 0025 0. 006 0. 035 0. 07 ATLAS 30 fb-1 One coupling varied at a time LEP present precision: 0. 05 to 0. 10 Comparison of representative 95% C. L. upper limits on Dkg and lg for present and future accelerators 20

Top physics Most intriguing fermion (large mass, large width, radiative corrections, etc. ) precision measurements needed (limited by statistics at Tevatron) LHC: • s ( ) 830 pb 107 pairs per year at low L 7 pb at s = 1. 8 Te. V measure mt , stt, Vtb, rare decays, polarisation, single top, etc. Measurements and their interpretation dominated by exp. and th. systematic uncertainties • production is main background to New Physics (Higgs, SUSY, …) • W jj in of jet scale events: in situ calibration 21

Measurement of mtop • Year 2005 : Dmtop 3 Ge. V • Best channel: tt Wb l b Wb jjb (Tevatron) 29. 6% of all tt evts top mass determined from hadronic part of decay mt=mjjb leptonic top used to tag event: hight p. T lepton large ETmiss • After all cuts : 130 000 events 10 fb-1, S/B ~ 65 W jj t bjj Full sim. combinatorics Contribution statistics light-jet scale b-fragmentation ISR FSR background Total Dmtop (Ge. V) < 0. 07 0. 3 1. 2 0. 2 1. 5 Ge. V 10 fb-1 1 experiment dominated by knowledge of physics 22

mt in leptons plus jets channel • Special sub-sample where t and t have high p. T. • Cleaner topology lower combinatorics and background • Jets from top decay are close FSR sensitivity reduced by summing up calorimetric energy in cone around top direction mt in di-lepton channel • Complementary to single lepton + jets • Indirect: relies on relation between kinematic distribution of top decay products and mt mt from t J/Y + X decays pioneered by CMS • Correlation between mt and inv. mass of J/Y + lepton system (heavy object (J/Y) larger fraction of b momentum stronger correlation with mt) • BR 10 -5 but signal very clean • Completely indep. systematics ( b fragmentation) Very promising Different samples Important cross-checks 23

Other measurements • s ( ) to < 10 % (L uncertainty) • find X up to 3 Te. V (s x BR > 10 fb) • spin correlations (top decays fast no hadronic bound states information on top spin not diluted ) • Couplings tt. H to 10 % (stat. only) FCNC couplings t. Vc, t. Vu (V=g, , Z) BR (t Zq) 10 -4 In general at least factor 10 better than Tevatron BR (t q) 10 -5 BR (t gq) 10 -5 • Single top: s 300 pb (40% of ) Never observed so far!! 1/3 of tt - probe Wtb vertex sensitive to new physics - Vtb to 10% (stat. error < 1%) - top polarisation 24

Standard Model Higgs • s (pp H + X) 30 pb 100 fb m. H = 100 Ge. V m. H = 1 Te. V • Large QCD backgrounds look for final states with leptons and photons • Main channels: l = e, m m. H < 180 Ge. V: H ZZ* 4 l, H WW(*) l l m. H > 180 Ge. V: H ZZ 4 l, H ZZ ll (400 < m. H < 900 Ge. V) H WW l jj (m. H > 400 Ge. V) • Detector performance critical (often S/B << 1): b-tagging, EM energy resolution, muon momentum resolution, multi-jet mass resolution, /j separation, forward jet tag, electron reconstruction, ETmiss measurement, …. . ) 25

H 80 m. H 150 Ge. V s x BR 50 fb (BR 10 -3) Backgrounds (challenging for EM calorimeter): (irreducible) : sgg~ 3 pb j+jj Theory knowledge should be as accurate as possible need sm 1% : s j ~ 103 s need Rj 104 -jet o -jet backround over at 20% level after full -ID Excellent photon/jet discrimination required 80 m. H 130 Ge. V s x BR 300 fb Complex final state: H bb, t bjj, t bl (l=e , m for trigger) Combinatorial background from signal reduced by reconstructing both top quarks ( e 1%) b-tagging is crucial Backgrounds : continuum ttbb, ttjj (dominant but measurable), Wjjjjjj, etc. H ZZ 4 l Gold-plated channel 180 m. H < 700 Ge. V ME calculations for W/Z+n jets should be as complete as possible s x BR 1 -10 fb Background: ZZ continuum ( S/B > 2) GH >> 1 Ge. V (dominates exp mass resolution for m. H>300 Ge. V) detector performance not crucial 26

SM Higgs discovery potential LEP 2 limit • Higgs can be discovered ( signal > 5 s) over full mass range after 1 year of operation • In most cases > 1 channel available • No k-factors used. k~1. 3 for signal results optimistic only if k 2 for backgrounds 27

Measurement of the Higgs parameters Higgs mass • precision dominated by 4 l and 2 channels • 0. 1% precision up to m. H 500 Ge. V • still at 1% level for m. H 700 Ge. V • dominant syst. uncertainty from l/jet scale • no theory errors included (e. g. mass shift for large GH due to interference resonant/non-resonant production) 28

Higgs mass in MSSM LEP lower limit: 90 Ge. V upper theoretical limit: 115 -130 Ge. V MSSM Higgs h, A, H H 4 l H/A mm h bb H hh bb A Zh bbll H/A tt Dm/m (%) 300 fb-1 0. 2 -0. 4 0. 2 -1. 5 1 -2 1 -2 1 -10 Systematic error on abs. energy scale: 0. 1% for l/ 1% for jets Important to have a matching between the experimental error and theoretical precision of the relation between mh and the MSSM parameters 29

Higgs width • Measured directly from width of reconstructed peak • Only possible for m. H <200 Ge. V (GH detector resolution) 5% precision for 300 < MH <700 Ge. V (region where best discovery channel H ZZ 4 l) Detector resolution measured to 1. 5% from GZ Uncertainty from Z radiative decays 30

Higgs production rates: s x BR • typical precisions: 7% -20% (depending on m. H) • dominant syst. : luminosity (5 -10%) • rate of H ZZ(*) 4 l allows disentangling SM/MSSM (~10 times smaller in MSSM) • rate of A/H tt, mm allows measurement of tgb ATLAS m. A=150 Ge. V ATLAS m. A=300 Ge. V 31

Couplings and branching ratios: • Can be obtained from rate measurements if s (pp H+X) known from theory • Otherwise: measure ratios of rates for different channels ratios of couplings many constraints of theory • A few examples here 300 fb-1 From One measures Error 15 % (*) 80 -120 Ge. V 7% 120 -150 Ge. V 15 % (*) 80 -120 Ge. V (*) also in MSSM for m. A > 200 Ge. V Error dominated by statistics Many other possibilities under study in SM (e. g. ratio of WW to gg fusion) and MSSM (e. g. ratio of A/H to mm and to tt) 32

SUPERSYMMETRY Can be discovered up to m ~ 2 Te. V (~ independent of model parameters) using inclusive signatures m ~ 1 Te. V undergo cascade decays many jets, leptons, missing ET in final state CMS, 100 fb-1 m. SUGRA 5 s contours for various signatures with high p. T l 33

Can ATLAS and CMS perform precise measurements (masses, couplings, etc. ) extract fundamental parameters of theory ? Not obvious (two LSPs, not enough constraints to reconstruct mass peaks) Point 1 5 points of m. SUGRA studied m 0 m 1/2 A 0 tgb (Ge. V) 400 0 2. 0 sgn m + 2 400 0 10. 0 + 3 4 200 800 100 200 0 0 2. 0 10. 0 + 5 100 300 2. 1 + Point 1 mass (Ge. V) 1004 mass (Ge. V) 925 mass (Ge. V) 325 2 1008 933 321 431 115 3 4 298 582 313 910 96 147 207 805 68 112 5 767 664 232 157 93 mass (Ge. V) 430 h mass (Ge. V) 95 -- Reconstruct exclusive decay chains determine masses from kinematic distributions (often model-independent) -- Global fit of the model to all measurements determine parameters (à la LEP) 34

Example : Point 5 q 02 l l 01 m = 690 Ge. V l +l ll constrains 0 1, 2 llq constrains 0 1, 2 lq constrains 02 llq lq End-points can be measured with precision of 1‰ to 1% for 100 fb-1 35

Summary of measurements for Point 5 ATLAS Measured mass Value (Ge. V) Error (%) 30 fb-1 Error (%) 300 fb-1 h l+l- edge 92. 9 108. 7 157. 2 239 688 662 767 493 1 0. 4 1. 2 4 1. 7 3 2. 6 -- 0. 2 0. 3 1 1 1. 5 10 Particles directly observable: These experimental measurements used to constrain the model and its parameters: ATLAS 300 fb-1 Precision of order 1% 36

Conclusions • LHC has huge discovery potential for New Physics: -- SM Higgs : full mass range -- MSSM Higgs : cover m. A , tgb plane fully -- SUSY : up to m ~ 2 Te. V -- Beyond SUSY (LQ, W’, Z’, etc. ) : up to m ~ 5 Te. V • Great potential also for precise measurements: -- m. W to 15 Me. V, TGC to 10 -3 -- many measurements in top sector (precision ~ %) -- Higgs mass : 1 ‰ (SM, h) to 1% (A/H) -- many SUSY measurements fundamental parameters to % • Excellent multi-purpose detectors needed (b-tagging, l/j energy resolution, dynamic range, particle identification) under construction Many thanks to Fabiola Gianotti for help in preparing this talk! 37