HIGGS + 2 JET PRODUCTION AT THE LHC

Transcription

1 IGGS + 2 JET PRODUCTION AT TE LC Dieter Zeppenfeld Universität Karlsruhe, Germany Seminar at KITP, Santa Barbara, March 13, 2008 LC goals Vector boson fusion Measurement of iggs couplings jj production via gluon fusion WW study ττ study Probing CP properties Summary

2 Goals of iggs Physics iggs Search = search for dynamics of SU(2) U(1) breaking Discover the iggs boson Measure its couplings and probe mass generation for gauge bosons and fermions Fermion masses arise from Yukawa couplings via Φ (0, v+ 2 ) L Yukawa = Γ i j d = f Q L i Φd j R Γ i j d m f f f ( 1 + v d R i Φ Q j ) L +... = Γ i j d v + 2 d i L d j R +... Test SM prediction: f f iggs coupling strength = m f /v Observation of f f Yukawa coupling is no proof that v.e.v exists Dieter Zeppenfeld j j production 1

3 iggs coupling to gauge bosons Kinetic energy term of iggs doublet field: [ (D µ Φ) (D µ Φ) = 1 ( gv ) 2W 2 µ µ + µ+ W µ W, Z mass generation: m 2 W = ( gv 2 WW and ZZ couplings are generated ) 2, m 2 Z = ( g 2 +g 2 )v 2 4 ( g 2 + g 2) ] ( v 2 Z µ Z µ v ) 2 iggs couples propotional to mass: coupling strength = 2 m 2 V /v g2 v within SM Measurement of WW and ZZ couplings is essential for identification of as agent of symmetry breaking: Without a v.e.v. such a trilinear coupling is impossible at tree level Dieter Zeppenfeld j j production 2

4 Feynman rules f f i m f v W ν - ig m W g µν W µ + Z ν i g 1 cosθ W m Z g µν Z µ Verify tensor structure of VV couplings. Loop induced couplings lead to V µν V µν effective coupling and different tensor structure: g µν q 1 q 2 g µν q 1ν q 2µ Dieter Zeppenfeld j j production 3

5 Total cross sections at the LC gg (NNLO) σ(pp + X) [pb] s = 14 TeV NLO / NNLO t t t qq _ ' W qq qq gg/qq _ tt _ (NLO) qq _ Z MRST [Krämer ( 02)] M [GeV] q q q _ q q _ q V V W, Z _ t t

6 Vector Boson Fusion (VBF) W p V q W τ m > 120 GeV p V τ + γ m < 140 GeV q W m < 150 GeV γ [Eboli, agiwara, Kauer, Plehn, Rainwater, D.Z.... ] Most measurements can be performed at the LC with statistical accuracies on the measured cross sections times decay branching ratios, σ BR, of order 10% (sometimes even better). Dieter Zeppenfeld j j production 5

7 VBF signature p + µ J 2 θ 1 J 1 ϕ θ 2 J 1 p µ + ϕ ϕ jj e - J 2 e - η Characteristics: energetic jets in the forward and backward directions (p T > 20 GeV) large rapidity separation and large invariant mass of the two tagging jets iggs decay products between tagging jets Little gluon radiation in the central-rapidity region, due to colorless W/Z exchange (central jet veto: no extra jets between tagging jets) Dieter Zeppenfeld j j production 6

8 Example: Parton level analysis of WW Kauer, Plehn, Rainwater, D.Z. hep-ph/ Near threshold: W and W almost at rest in iggs rest frame = use m ll m νν for improved transverse mass calculation: E T,ll = p 2 T,ll + m2 ll /E T = p/ 2 T + m2 νν p/ 2 T + m2 ll M T = (/E T + E T,ll ) 2 (p T,ll + p/ T ) 2 Observe Jacobian peak below M T = m dσ/dm T (WW) [fb/gev] ττ bbjj iggs WW tt+jets M T (WW) [GeV] Transverse mass distribution for m = 115 GeV and WW e ± µ p/ T

9 tau reconstruction in collinear approximation, the decay lepton has the same direction as the τ, i.e. p µ l,i = x i p µ τ,i the energy fractions x 1, x 2 of the decay leptons can be reconstructed by solving the equation: p T = the invariant tau pair mass is then given by ( ) ( ) p 1,T + 1 p 2,T x 1 x 2 m 2 ττ = 2p l,1 p l,2 x 1 x 2 = m2 ll x 1 x 2 Dieter Zeppenfeld j j production 8

10

11 iggs discovery potential Signal significance 10 2 L dt = 30 fb -1 (no K-factors) ATLAS γ γ tt ( bb) ZZ (*) 4 l WW (*) lνlν qq qq WW (*) qq qq ττ Total significance S B m (GeV/c 2 )

12 Statistical and systematic errors at LC Assumed errors in fits to couplings: QCD/PDF uncertainties - ±5% for VBF - ±20% for gluon fusion luminosity/acceptance uncertainties - ±5%

13 Measuring iggs couplings at LC LC rates for partonic process pp xx given by σ(pp ) BR( xx) σ() BR( xx) = σ()sm Γ SM p ΓpΓ x Γ, Measure products Γ p Γ x /Γ for combination of processes (Γ p = Γ( pp)) Problem: rescaling fit results by common factor f Γ i f Γ i, Γ f 2 Γ = obs f Γ i + Γ rest leaves observable rate invariant = no model independent results at LC Loose bounds on scaling factor: f 2 Γ > obs. f Γ x = f > obs. Γ x Γ = BR( xx)(= O(1)) obs. Total width below experimental resolution of iggs mass peak ( m = GeV) f 2 Γ < m = f < m Γ < O(10 40)

14 Fit LC data within constrained models g ττ g bb = SM value g WW g ZZ = SM value no exotic channels With 200 fb 1 measure partial width with 10 30% errors, couplings with 5 15% errors

15 Distinguishing the MSSM iggs sector from the SM Alternative: compare data to predictions of specific models Example: m max Consider modest m A : scenario of LEP analyses decoupling almost complete for hww and hγγ (effective) vertices enhanced hbb and hττ couplings compared to SM increases total width of h = tan β σ-effects or more at small m A 2 * 30 fb -1 2 * * 100 fb -1 2 * 300 fb -1 m h max scenario 3σ m h = 130 GeV 125 GeV SM rates for h ττ in VBF suppressed h γγ and h WW rates in VBF M A (GeV) Dieter Zeppenfeld j j production 14

16 ow to distinguish VBF and gluon fusion? W, Z vs. W, Z (a) Double real corrections to gg can fake VBF = we need to investigate the phenomenology of these two processes and understand the differences that can be exploited to distinguish between gluon fusion and VBF = derive cuts to be applied to enhance VBF with respect to gluon fusion. Measure WW and ZZ coupling = derive cuts to be applied to enhance gluon fusion with respect to VBF. Measure effective gg coupling or tt coupling Dieter Zeppenfeld j j production 15

17 Diagrams for gg fusion with finite m t effects (a) (b) (c) q Q q Q q g q g g g g g plus crossed processes. In total 61 independent diagrams. [DelDuca, Kilgore, Oleari, Schmidt, DZ (2001)]

18 Applied cuts for LC predictions The cross section diverges in collinear and soft regions INCLUSIVE cuts to define + 2 jets p T j > 20 GeV η j < 5 R jj = (η j1 η j 2 ) 2 + ( φj1 φ j2 ) 2 > 0.6 VBF cuts to enhance VBF over gluon fusion In addition to the previous ones, we impose η j1 η j2 > 4.2 η j1 η j2 < 0 m jj > 600 GeV the two tagging jets must be well separated in rapidity they must reside in opposite detector hemispheres they must possess a large dijet invariant mass. LC cross sections below calculated with CTEQ6L1 pdfs and fixed α s = 0.12 Expect factor 1.5 to 2 scale uncertainty due to σ α 4 s Dieter Zeppenfeld j j production 17

19 Total cross section with cuts as function of m Large top mass limit ok for total cross section provided m < m t Dieter Zeppenfeld j j production 18

20 Distributions and m t limit Transverse momentum: Large top mass limit ok provided p T,j < m t Dijet invariant mass: Large top mass limit ok throughout Dieter Zeppenfeld j j production 19

21 New calculation: pseudoscalar iggs production pp A j jx including top and bottom loops + interference [Michael Kubocz, diploma thesis] A A A (a) (b) A A A (c) Dieter Zeppenfeld j j production 20

22 New elements in the calculation AQQ vertices given by m b v γ 5 tanβ and m t v γ 5 Interference of top and bottom loops 1 tanβ Can simulate CP violation in the iggs sector: a + ibγ 5 coupling to top and bottom Inclusive cuts VBF cuts σ [pb] 2 10 tanβ=1, m =175 GeV t tanβ=50, m =4.4 GeV b tanβ=7, t+b σ [pb] 1 tanβ=1, m =175 GeV t tanβ=50, m =4.4 GeV b tanβ=7, t+b m A [GeV] m A [GeV]

23 Gluon Fusion as a signal channel eavy quark loop induces effective gg vertex: CP even : CP odd : i m Q v L e f f = α s 12πv Ga µνg µν,a m Q v γ 5 L e f f = α s 8πv A Ga µν G µν,a = α s 16πv A Ga µνg a αβ εµναβ Azimuthal angle between tagging jets probes difference Use gluon fusion induced Φ j j signal to probe structure of gg vertex Measure size of coupling (requires NLO corrections for precision) Find cuts to enhance gluon fusion over VBF and other backgrounds = Study by Gunnar Klämke in m Q limit (hep-ph/ ) Dieter Zeppenfeld j j production 22

24 Gluon fusion signal and backgrounds Signal channel (LO): pp j j in gluon fusion with W + W l + l ν ν, (l = e, µ) m = 160 GeV dominant backgrounds: W + W -production via VBF (including iggs-channel): pp W + W j j top-pair production: pp t t, t t j, t t j j (N. Kauer) QCD induced W + W -production: pp W + W j j applied inclusive cuts (minimal cuts): 2 tagging-jets p T j > 30 GeV, η j < identified leptons p Tl > 10 GeV, η l < 2.5 separation of jets and leptons η jj > 1.0, R jl > 0.7 process σ [fb] GF pp + j j VBF pp W + W + j j 75.2 pp t t 6832 pp t t + j 9518 pp t t + j j 1676 QCD pp W + W + j j 363

25 Characteristic distributions Separation of VBF j j signal from QCD background is much easier than separation of gluon fusion j j signal tagging jet rapidity separation dijet invariant mass jj 1/σ dσ/d η signal VBF tt+jets QCD-WW [1/GeV] 1/σ dσ/dm jj signal VBF tt+jets QCD-WW η jj m jj [GeV] Dieter Zeppenfeld j j production 24

26 Selection continued b-tagging for reduction of top-backgrounds. (CMS Note 06/014) (η, p T ) - dependent tagging-efficiencies (60% - 75%) with 10% mistagging - probability selection cuts: p Tl > 30 GeV, M ll < 75 GeV, M ll < 0.44 M WW T, R ll < 1.1, M WW T < 170 GeV, p T > 30 GeV M WW T = (E T + E Tll ) 2 ( p Tll + p T ) 2 1/σ dσ/d R ll [1/GeV] WW 1/σ dσ/dm T signal VBF t t+jets QCD-WW R ll WW [GeV] m T

27 Results process σ [fb] events/ 30 fb 1 GF pp + j j VBF pp W + W + j j pp t t pp t t + j pp t t + j j QCD pp W + W + j j Σ backgrounds S/ B 16.2 for 30 fb 1 Dieter Zeppenfeld j j production 26

28 iggs + 2 Jets in Gluon Fusion, ττ l + l ν ν this channel has not been studied so far q q interesting for SM iggs ( 120 GeV) and SUSY scenario with large tanβ (m m A > 150 GeV) x-section times branching ratio of 50 fb looks promising (SM) g g t has potential for study of iggs CP-properties q q Studied so far (by Gunnar Klämke): Study of signal and SM backgrounds for m = 120 GeV case (simple cut based analysis) same for one MSSM scenario m A = 200 GeV, tanβ = 50 Questions: ow many signal and background events are there after cuts (what s the statistical significance) What are the prospects of CP-measurements via jet-jet azimuthal angle correlation

29 finite detector resolution The detector has a finite resolution. The measured jet energy and missing transverse energy have large uncertainties. Parameterization (from CMS NOTE 2006/035, CMS NOTE 2006/036): Jets : Leptons : Missing p T : E j E j = ( a E T j ) b c ET j E l E l = 2% a b c η j < < η j < η j > p x = 0.46 E T j Dieter Zeppenfeld j j production 28

30 SM iggs with 120 GeV mass inclusive cuts p T,jets > 30 GeV, p T,l > 10 GeV, η j < 4.5, η l < 2.5, η jj > 1.0, R jl > 0.7, cross sections for inclusive cuts for signal and background process σ [fb] events / 600 fb 1 GF pp + j j ττ j j GF pp A + j j ττ j j VBF pp + j j ττ j j QCD pp Z + j j ττ j j VBF pp Z + j j ττ j j pp t t pp t t + j pp t t + j j QCD pp W + W + j j VBF pp W + W + j j

31 Distributions [1/GeV] 1/σ dσ/dm ll Signal VBF- QCD-Z EW-Z tt+jets QCD-WW EW-WW [1/GeV] 1/σ dσ/dm ττ Signal VBF- QCD-Z EW-Z tt+jets QCD-WW EW-WW m ll dilepton invariant mass [GeV] m ττ [GeV] reconstructed ττ invariant mass Dieter Zeppenfeld j j production 30

32 selection cuts a b-veto was applied to reduce the top backgrounds. R ll < 2.4, p T > 30 GeV, m ll < 80 GeV, 110 GeV < m ττ < 135 GeV, 0 < x i < 1 process σ [fb] events / 600 fb 1 GF pp + j j ττ j j GF pp A + j j ττ j j VBF pp + j j ττ j j QCD pp Z + j j ττ j j VBF pp Z + j j ττ j j pp t t pp t t + j pp t t + j j QCD pp W + W + j j VBF pp W + W + j j backgrounds for cp-even higgs: S/ B 17 ( 600 fb 1 ) this corresponds to: S/ B 5 ( 50 fb 1 ) for cp-odd higgs: S/ B 40 ( 600 fb 1 ) this corresponds to: S/ B 5 ( 10 fb 1 )

33 Tensor structure of the VV coupling Most general VV vertex T µν (q 1, q 2 ) g µ µ q q q q q V q 1 1 q 2 V q 2 Q Q Q Q ν ν (a) (b) Physical interpretation of terms: SM iggs L I V µ V µ a 1 loop induced couplings for neutral scalar CP even L e f f V µν V µν a 2 T µν = a 1 g µν + a 2 ( q1 q 2 g µν q ν 1 qµ 2) + a 3 ε µνρσ q 1ρ q 2σ CP odd L e f f V µν Ṽ µν a 3 Must distinguish a 1, a 2, a 3 experimentally The a i = a i (q 1, q 2 ) are scalar form factors Dieter Zeppenfeld j j production 32

34 Azimuthal angle correlations Tell-tale signal for non-sm coupling is azimuthal angle between tagging jets Dip structure at 90 (CP even) or 0/180 (CP odd) only depends on tensor structure of VV vertex. Very little dependence on form factor, LO vs. NLO, iggs mass etc.

35 Azimuthal angle distribution and iggs CP properties Kinematics of j j event: Define azimuthal angle between jet momenta j + and j via ε µνρσ b µ + jν +b ρ jσ = 2p T,+ p T, sin(φ + φ ) = 2 p T,+ p T, sin φ jj φ jj is a parity odd observable φ jj is invariant under interchange of beam directions (b +, j + ) (b, j ) Work with Vera ankele, Gunnar Klämke and Terrance Figy: hep-ph/

36 Signals for CP violation in the iggs Sector 1/ σ dσ / d φ jj CP-even, CP-odd CP-even CP-odd SM mixed CP case: a 2 = a 3, a 1 = 0 pure CP-even case: a 2 only pure CP odd case: a 3 only Position of minimum of φ jj distribution measures relative size of CP-even and CP-odd couplings. For = Minimum at α and π α φ jj a 1 = 0, a 2 = d sinα, a 3 = d cosα,

37 Φ j j -Distribution in gluon fusion: WW case Fit to Φ jj -distribution with function f( Φ) = N(1 + A cos[2( Φ Φ max )] B cos( Φ)) events events Signal VBF t t+jets QCD-WW Φ jj Φ jj L = 300 fb 1 ( η jj > 3.0) CP-even CP-odd A = ± A = ± Φ max = 5.8 ± 15.3 Φ max = 93.7 ± 5.1 fit of the background only : A = ± and Φ max = 64 ± 25 ( mean values of 10 independent fits of data for L = 30 f b 1 each)

38 Φ j j -Distribution: CP violating case events Φ jj CP-mixture: equal CP-even and CP-odd contributions A = ± Φ max = 45.6 ± 7.3 Dieter Zeppenfeld j j production 37

39 ττ case: Φ j j -distribution with backgrounds Fit to Φ jj -distribution with function f( Φ) = N(1 + A cos[2( Φ)] B cos( Φ)) CP-even CP-odd A = ± A = ± fit of the background only : ± significance for CP-even vs. CP-odd 8 Signal VBF- QCD-Z EW-Z t t+jets L = 600 fb 1 ( η jj > 3.0) Dieter Zeppenfeld j j production 38

40 Summary iggs + 2 Jet events at the LC provide very useful information on iggs couplings Order 200 fb 1 of LC data allow to probe iggs couplings at the 10% level Beside VBF, gluon fusion is a second copious source of Φ j j events at the LC Full one-loop calculations are available for quark-loop induced j j and A j j production, including CP-even CP-odd interference and finite quark mass effects For m = 160 GeV and dominant decay WW the gluon fusion induced signal at the LC is visible above backgrounds. ττ is somewhat more challenging CP-violation in the iggs sector is observable via the shape of the azimuthal angle distribution dσ/d φ jj Dieter Zeppenfeld j j production 39