Heavy Quarks and τ. CVC test Michel parameters. Charm Mixing f D s

Transcription

1 Heavy Quarks and τ University of Illinois CVC test Michel parameters Charm Mixing f D s B f B Two mysteries: N c and η K CKM: ππ, Kπ, V cb, V ub Lifetime and mixing CP search

2 Belle and Babar* A very brief description of the Belle and Babar data sets: Both run at: L ~ 2 x cm -1 s -2 (CLEO s peak: ~8 x ) Integrated luminosities: Babar: 9.9 fb -1 for Osaka, 7.7 fb -1 on 4S (>11 in the can, I think) Belle: 6.8 fb -1 for Osaka, 6.2 fb -1 on 4S CLEO: 14 fb -1, 9 fb -1 on 4S 1 fb -1 is about 1 million BB pairs. * Beauty and the Beast

3 Two τ results, briefly τ ππ o ν goes via vector current. CVC (e + e - ) predicts: BR = (24.52±0.33)% CLEO measures: (25.32±0.15)% Fractional difference is (3.2±1.5)%. Michel parameters: DELPHI has new measurements. Still x5 worse than with muons, but B factories can improve this (not yet systematics limited). CLEO DELPHI ρ Standard model η CLEO ξ ξδ DELPHI

4 DD Mixing The standard box diagram is GIM suppressed, because the top quark does not appear (only d,s,b). However, theoretical predictions vary: Parameters x and y are the imaginary and real parts of the mixing matrix. Two methods: Compare the lifetime of decays to different CP states (eg, D KK,Kπ). Only sensitive to y. Look for interference between mixed and direct decays to the same state (eg, D o K + π - ). H. Nelson hep-ex/ SM x SM y New physics

5 Lifetime method: Belle measures the lifetimes separately. Focus performs a combined fit. Belle: D = ( ± 18) fs D o = (414±3.8 ± 3.4)fs D s = ( ) fs D o KK D o K y CP = K KK = (408.9 ± 14.3) fs = (412.9 ± 3.8)fs = ( )% Focus: D o K = (409.2 ± 1.3 ±?) fs y CP = (3.42 ± 1.39 ± 0.74)% K K K Belle s measurements verify their time calibration for B J/ΨK o

6 Interference method: CLEO measures wrong sign Kπ only. Look for interference between mixing and DCSD channels. Measure both x and y. (Primes result from final state interactions.) -5.8% < Y < 1.0% x < 2.9%

7 Comparisons to CLEO, E791 and BELLE FOCUS CLEO The comparison to CLEO is vaild only if one assumes a small strong phase difference δ. About the same sensitivity to the CLEO CP constrained fit, but the opposite sign! FOCUS y CP = % E791 BELLE 95% CL Recent Measurements E791: y CP = % CLEO: -5.8 % < y < 1% ( 95% CL) BELL Eprelim.: y CP = /10/00 FOCUS MIXING 17 x splits masses, y splits lifetimes. What if δ = 40, the estimated maximum of the model of Falk, Nir & Petrov (99)? We see some overlap... CLEO and FOCUS would be more consistent if δ > Bergmann, Grossman et al(00). Phase ambiguity FOCUS CLEO E791 BELLE 95% CL 8/10/00 FOCUS MIXING 18

8 Decay Constants (f D, f Ds, f B ) Leptonic decay rates depend on the meson decay constants: b V* µ + ub B + f W + B u ν BR(B l ) = G F 2 m B m l m l 2 m2 B 2 f B 2 Vub 2 B Similarly for D µν, D s µν, etc. Many other processes depend on them also. For example, in BB mixing: M d = 0.50ps 1 B Bd f Bd 200MeV 2 m t ( m t ) 170GeV 1.52 V td B 0.55 This is non-perturbative QCD - measure or calculate on lattice. B τν is difficult to measure; the approach for now is to use charm results to test the lattice calculations.

9 Recent Results ALEPH D s ->τν: (5.86±1.18±2.09)% (e) (5.78±0.85±1.76)% (µ) D s ->µν: (0.68±0.11±0.18)% τ eνν µν f Ds 275±28±49 MeV 273±20±41 291±25±38 285±20±40 (combined)

10 BEATRICE (WA92): PL B478, 31(2000) Fixed target: πw, πcu Assume D s µν = (0.83 ± 0.23 ± 0.06 ± 0.18(D s φπ))% = 323±44±12±34 MeV f D s f D f Ds 2 = 0.82 ± 0.09 cc Ds µν D µν

11 f Ds summary WA75 ( 93) CLEO ( 94) BES ( 95) E653 ( 96) L3 ( 97) CLEO ( 98) Beatrice ( 00) ALEPH ( 00) 238 ± 47 ± 21± ± 30 ± 43 ± ± ± 34 ± 20 ± ± 58 ± 33± ±17 ± 25 ± ± 44 ±12 ± ± 20 ± 40 Systematic errors dominate Lattice UKQCD ( 00) Draper98(q) (unq) f Ds (MeV)

12 B τν Predict: (4.08 ± 0.24) x 10-4 x V ub /V td 2 ~ few x 10-5 (Use m B to eliminate f B ) L3 B τν < 5.7x10-4 (the best LEP result) CLEO At the Υ(4S) one can suppress background by reconstructing the other B. B τν < 8.4x10-4 B K ± νν < 2.4x10-4 Data Fit to Monte Carlo shapes The Standard Model may be reachable.

13 Charm Counting It is possible to calculate both the inclusive semileptonic BR and the number of charmed particles per decay in terms of fundamental quantities. In the past, the measurements have disagreed, making interpretation difficult. New ALEPH & DELPHI results change the picture. There is now experimental consistency. N c µ/m b 0.5 CLEO LEP (97) m c /m b LEP (00) N c = ± B SL = ± 0.25 (Barker & ICHEP) Using new lifetimes B SL CLEO: B SL = (10.49 ± 0.17 ± 0.43)% Nc = 1.10 ± 0.05 Note: LEP Vcb WG says: B SL = ± 0.11 ± 0.18

14 B η K CLEO observed (1997) B ± η K ± to have a surprisingly large BR. This is now confirmed, and the K o mode is also seen. BR(B ± η K ± ) = ( ± 0.7) x BR(B o η K o ) = ( ± 0.9) x BR(B ± η π ± ) < 1.2 x 10-5 ργ ηππ The equality of these BRs and the suppression of η π disfavors the spectator process: d,s u π -,K - B - b u Intrinsic charm or glue content would enhance penguin diagrams: b u glue u u uds uds B - K - s u η, η,π o η, η (no π o )

15 How to measure the CKM Matrix b ulν B DK,ππ,Kπ η V ud V ub * γ B π + π,ρ + π α V cd V cb * V td V* tb BB mixing, B ργ β b clν B J/Ψ K s ρ

16 B J/ψK s sin(2β) vs B π + π sin(2α) b d b d B J/ψK s V* cb V* ub V cs glue W W V us c c s d Penguin Aλ 4 (ρ-iη) ~ Tree Aλ 2 (1-λ 2 /2) ~ λ 2 ~ 0.05 c c s d b d b d B π + π V* W ub V* ub glue Penguin Tree V ud ~ 1 V ud u d u d d u u d Is this correct? B o K + π is dominated by penguins, so: B o π + π 1 ~ B o K + π 4 tells us that P/T ~ 0.5. The presence of penguins requires an isospin analysis. (π 0 π 0 BR < ) It may be easier to use ρπ, because the different charge states interfere. (Snyder-Quinn)

17 B K + π, π + π Data K ± π m ± K o π m ±1.6 < K ± π o K o π o CLEO Babar Belle A complete set! π ± π m ± BRs are ± 3.4 < 16.5 Particle ID is important: π ± π o < 12.7 π o π o < 5.7 K ± K m <1.9 < 6.6 < 6 K ± K o < ± < 8 K o K o < ULs are 90% Kπ bkgd ππ signal

18 V cb The differential decay rate, for B D*lν is: 2 dγ dw = G F 48 3 V cb 2 F(w) 2 G(w) w is the Lorentz γ factor of the recoiling D*. It is a function of masses and q 2. (1 w 1.5) w = 1 is the zero recoil point. G(w) is a known kinematic function. F(w) is the form factor. HQET constrains it. As m b,c, F(1) 1. For finite m, the corrections are O(1/m 2 ). Two methods: Extrapolate differential rate to w = 1. Integrate the total B(b X c lν). The first method is less sensitive to F(w) systematics, but has worse statistics.

19 New CLEO result using w = 1 method. Smallest w bin (worst S/B) dγ/dw Similar plot from OPAL:

20 CLEO: F(1) V cb = (42.4 ± 1.8 ± 1.9) x 10-3 using: F(1) = ± BR(D* + lν) = (5.66 ± 0.29 ± 0.33)% yields: V cb = (46.4 ± 2.0 ± 2.1 ± 2.1) x 10-3 There appears to be a problem: F(1) uncertainty LEP (VCB WG) (42.4 ± 1.8 ± 1.9) x 10-3 (34.5 ± 0.7 ± 1.5) x 10-3 This result is strongly correlated with the measured slope, ρ 2, of F(w): CLEO: ρ 2 = 1.67 ± 0.11 ± 0.22 LEP: ρ 2 = 1.01 ± 0.08 ± 0.16 As we saw, LEP and CLEO now agree about the total semileptonic BR. Preliminary: Belle: BR(D*lν) = (4.74 ± 0.25 ± 0.51)% BR(Dlν) = (2.07 ± 0.21 ± 0.31)%

21 V ub B X u lν B ρlν, πlν inclusive exclusive Signal New inclusive method: Look at M had as well as E lep B X c lν BR(B X u lν) = (1.57 ± 0.35 ± 0.48 ± 0.27) x 10-3 V ub / V cb = ± ± LEP V ub WG (ALEPH+DELPHI+L3): BR(B X u lν) = (1.74 ± 0.37 ± 0.88 ± 0.21) x 10-3 V ub = ( ± 0.02 ± 0.20) x

22 B Meson Lifetime & m d unmixed - mixed = cos( m t) unmixed + mixed τ(b ± ) ps τ(b0) ps Ratio m d h ps -1 PDG ± ± ± ±0.01 New: ALEPH 1.648±0.049± ±0.053± ±0.059±0.018 OPAL 1.528±0.025± ±0.018±0.015 Babar 1.602±0.049± ±0.052± ±0.044± ±0.017±0.022(recon) 0.507±0.015±0.022 (ll) Belle 1.70 ±0.06 ± ± 0.05 ± ±0.06 ± ±0.008±0.030 (ll) CLEO 0.522±0.029±0.031 * CDF 0.495±0.026±0.025 OPAL *from χ d = ± ± assuming: Γ = 0 using τ B o = 1.55 ps Comparison of systematic errors: LEP τ: B momentum Babar τ: z recon Babar m: Misid (B + & cascade) Belle m: B + contamination These must be improved to make progress

23 Babar and Belle time measurement: B s D K rec (4S) Z 0.56 B z ~260 m e,, K tag tag The time resolution is about τ B /2. The primary vertex is not used (z V is poorly known), except in dilepton analyses. One B decay defines t = 0.

24 Dileptons Enriched dileptons (soft π ± selects B o ) Belle Babar Babar Recon + tag Babar

25 m s 10.1 ps ps -1 Anticipated CDF sensitivity in Run-II: Expect 2 fb -1 x s ~ 65 ( m s ~ 40 ps -1 ).

26 CP violation in B system Direct CP violation (without mixing): Γ( B f ) Γ( B f ) CLEO: pp,pv: Predictions require assumptions about strong phase shifts. K*γ: Use self-tagging modes. B + : ( 38 ± 20)% B o : (-13 ± 17)% A both: ( 8 ± 13)% CP = Γ 2Γ = Γ b Γ b Γ b + Γ b SM predicts ~1%. J/ψK ± : A CP = (-1.8±4.3)% Direct CP violation does not complicate the J/ψK o measurement.

27 CP in mixing (as in K o system) CLEO: Look for Υ(4S) BB Υ(4S) B B A CP = 4Re(ε B ) = (χ + - χ - )/(χ + + χ - ) Re(ε B ) =0.004 ± < 3.4% (95% c.l.) Expect about 10-3 OPAL s time dependent measurement is similar: Re(ε B ) = ± ± 0.003

28 B J/ψ K o Interference between decay and mixing Only one decay diagram contributes significantly. Interference is provided via BB mixing. B o (') K s,l B o (') K s, L A(t) Bo B o B o o = ±sin(2 )sin( m t) + B Decay rates for B tag (+) and B tag (-): f ± = Γe Γ t 4 [ 1± Dsin( 2 )sin( m t) ] Decay probability Babar sin2β = 0.6 f - f t/τ

29 The signal is quite clean (94% for all J/ψK s ), despite the small branching fraction. B J/ψK s ( π + π - )

30 Belle s time resolution: σ = 1.11 ps (main) 2.24 ps (tail) Babar s is somewhat better: 0.6 ps (main) 1.8 ps (tail)

31 Babar 120 CP odd events: 85 J/ψK s (π + π - ) 12 J/ψK s (π o π o ) 23 ψ K s (π + π - ) Belle 52 CP odd events (mostly J/ψK s (π + π - )) 46 CP even events (mostly J/ψK L ) sin2β = 0.12 ± 0.37 ± 0.09 CDF: sin2β =

32 ±1 σ regions: Babar CDF m s / m d m d 0.6 η V ub /V cb ε K ρ 1 Babar

33 Nature 405, 722 (6/15/2000):