Measurements of the Masses, Mixing, and Lifetimes, of B Hadrons at the Tevatron

Transcription

1 Measurements of the Masses, Mixing, and Lifetimes, of B Hadrons at the Tevatron Mike Strauss The University of Oklahoma The Oklahoma Center for High Energy Physics for the CDF and DØ Collaborations

2 Outline B Physics at the Tevatron B Resonances B 0 oscillations B Lifetimes Exclusive Decays Lifetime Ratios and Differences Mike Strauss The University of Oklahoma 2

3 Tevatron Luminosity ~0.5 fb 1 delivered this year Detectors collect data at typically 85% efficiency These analyses use pb 1 About 150 pb 1 of data has been recorded and is currently being analyzed Mike Strauss The University of Oklahoma 3

4 Why B Physics? Understanding the structure of flavour dynamics is crucial 3 families, handedness, mixing angles, masses, any unified theory will have to account for it Weak decays, especially mixing, CP violating and rare decays, provide insight into short-distance physics Short distance phenomena are sensitive to beyond-sm effects CKM matrix determines the charged weak decays of quarks, tree level diagrams, one-loop transitions Lifetime measurements probe QCD at large distances and give information on form factors and quark models Mike Strauss The University of Oklahoma 4

5 B Physics at the Tevatron Large production cross sections All B Hadrons produced (Best B s and Λ b ) Larger inelastic cross section (S/B 10-3 ) Specialized Triggers: Single lepton triggers Dilepton triggers (e.g. J/ψ µ + µ ) L1 Track triggers L2 displaced track trigger for CDF σ(pp bb) 150 µb at 2 TeV σ(e + e bb) 7 nb at Z 0 σ(e + e bb) 1 nb at ϒ(4S) Mike Strauss The University of Oklahoma 5

6 Detectors Silicon vertex tracker, Axial solenoid, Central tracking, High rate trigger/daq, Calorimeter, Muon system CDF DØ L2 trigger on displaced vertexes Low p particle ID (TOF and de/dx) Excellent mass resolution Excellent muon ID; η < 2 Excellent calorimetry Tracking acceptance η < 2-3 L3 trigger on impact parameter Mike Strauss The University of Oklahoma 6

7 DØ Tracker SMT + CFT η max = 1.65 SMT region η max = 2.5 SMT Mike Strauss The University of Oklahoma 7 η = - ln (tan(θ/2))

8 Silicon Microstrip Tracker (SMT) 12 F-disks 4 H-disks Multi-layer barrel cross-section 6 Barrels Barrels F-disks H-disks Dble+sngle sided Double sided Single sided Stereo angle 0 o, 2 o, 90 o ±15 o ±7.5 o Channels ~400K ~250K ~150K Inner radius 2.7 cm 2.6 cm 9.5 cm 3m 2 of silicon Outer radius 9.4 cm 10.5 cm 26 cm Mike Strauss The University of Oklahoma 8

9 Tracking Performance Muon η in J/ψ events p T spectrum of soft pion candidate in D* D 0 π Coverage of Muon system is matched by L3/offline tracking Tracks are reconstructed starting from p T = 180 MeV Mike Strauss The University of Oklahoma 9

10 SMT Resolution and de/dx Impact Parameter Resolution SMT de/dx MC Data σ(dca) 53 P T = 1 GeV and 15 higher P T Can provide: K/π separation for P tot < 400 MeV p/π separation for P tot <700 MeV NOT yet used for PID Mike Strauss The University of Oklahoma 10

11 Belle s Discovery of X(3872) X J/ψ π + π No signal in γx c1 decay Mass doesn t fit easily into charm spectroscopy models Near D 0 D* 0 threshold Charmonium? A loosely bound meson state? M X = ± 0.6 (stat) ± 0.5 (sys) Mike Strauss The University of Oklahoma 11

12 X(3872) CDF and DØ have confirmed Belle s discovery of the X(3872) 730 ± 90 candidates ~12 σ effect M X = ± 0.7 (stat) ± 0.4 (sys) MeV/c 2 M = ± 3.1(stat) ± 3.0 (sys) MeV/c 2 M +M(J/ψ) = ±4.3 MeV/c 2 Mike Strauss The University of Oklahoma 12

13 Long Lifetime fraction of X(3872) Is the X charmonium, or something else? ψ(2s): X(3872): ψ(2s): 28.3±1.0(stat) ±0.7(syst)% X(3872): 16.1±4.9(stat) ±2.0(syst)% Mike Strauss The University of Oklahoma 13

14 X(3872) ψ(2s) comparison Is the X charmonium, or something else? y <1 cosθ π <0.4 dl<0.1 cosθ µ <0.4 p T >15 GeV/c Iso=1 DØ multi-parameter comparison Mike Strauss The University of Oklahoma 14

15 First Observation of B S φφ Charmless B VV decay not yet observed Dominated by penguin contributions Cuts optimized on B s J/ψ φ BR error on this decay is dominant systematic uncertainty BR(B s φφ ) = 1.4 ± 0.6(stat) ± 0.2(syst) ± 0.5(BR) 10-5 Mike Strauss The University of Oklahoma 15

16 B h ± h + Fractions h, h = π, K Motivation is to construct many charmless two body decays for CP studies Hadronic B trigger using SVT de/dx and α used for PID Signal reconstructed with vertex constrained fit Fractions measured with unbinned likelihood fit using M ππ, α, ID1, ID2. Mike Strauss The University of Oklahoma 16

17 PID using α = (1-p 1 /p 2 )q 1 Mike Strauss The University of Oklahoma 17

18 B h ± h + Fractions Mike Strauss The University of Oklahoma 18

19 B h ± h + Fractions BR BR ( B ± m π π ) ( B ± m K π ) d ( stat) 0. ( syst) d = 0.24 ± 0.06 ± 05 ( 0 + ) ( 0 + B ) d K π N Bd K π ( 0 + ) ( 0 + B K π + N B K π ) N ACP = = 0.04 ± 0.08 ± N d d 0.01 Assuming B s K + K - is 100% CP even, Γ s /Γ s = 0.12±0.06, Γ s =Γ d, ( ± m B ) d π π ( ± m B K K ) f BR d = 0.48 ± 0.12 ± f BR 07 s s ( ± m B ) s K K ( ± m B K π ) ( stat) 0. ( syst) f BR d = 0.50 ± 0.08 ± f BR 07 d d ( stat) 0. ( syst) Mike Strauss The University of Oklahoma 19

20 Search for B 0 (s,d) µ+ µ SM BR(B s0 µ + µ - ) (3.4±0.5) 10-9 ; BR(B d0 µ + µ - ) (1.5±0.9) Expected BG: 1.05 ± 0.30 BR(B s0 µ + µ ): < at 95% CL (CDF) Expected BG: 3.7 ± 1.1 events BR(B s0 µ + µ ): < at 95% CL (DØ) BR(B d0 µ + µ ): < at 95% CL Mike Strauss The University of Oklahoma 20

21 Observation of B** B Spectroscopy: B (J p = 0 ) B*(J p = 1 ) decays to Bγ (100%) M = M(B * ) M(B) = 46 MeV/c 2 The B** consists of four separate states 2 narrow states B 1 (1 + ) and B 2 * (2 + ), decay via D-wave; 2 wide states B 0 *(0 + ) and B 1 ' (1 + ), decay via S-wave; None of these individual states are well established Decay channels used: B d ** B ± π + ; B** + B d π + ; B** B*π Bπ (γ) B ± J/ψ K ± ; B d J/ψ K *0 ; B d J/ψ K 0 s Mike Strauss The University of Oklahoma 21

22 Distinct Narrow B** States 350 pb 1 Sum of 3 decay modes The first direct measurement of masses and splitting between B 2 * and B 1 M(B * ) = M(B) + 46 MeV (γ) B 1 Bπ (γ) B 2 * Bπ (γ) B 2 * Bπ M(B 1 ) = 5724 ± 4 ± 7 MeV /c 2 M(B 2* ) M(B 1 ) = 23.6 ± 7.7 ± 3.9 MeV/c 2 Mike Strauss The University of Oklahoma 22

23 Observation of B C J/ψ µ X Combined unbinned likelihood fit made to mass and lifetime 95±12±11 candidates M(B c ) = ± 0.34 GeV/c τ(b c ) = ± ps First 5σ measurement Mike Strauss The University of Oklahoma 23

24 B s D s l + X DØ RunII Preliminary, Luminosity = 250 pb B µ - φ π + X 9481±253 D s φ π ±239 D + φ π + µ - φ π + µ - φ π M(φ π + ) GeV/c 2 Mike Strauss The University of Oklahoma 24

25 Fully Reconstructed B s D* s and others D s Useful for B s Mixing CDF golden channels : B s D s π D s φπ, K*K, πππ B s D s π π π Mike Strauss The University of Oklahoma 25

26 B d Mixing In SM B d mixing is explained by box diagrams Constrains V td CKM matrix element Mixing frequency m d has been measured with high precision at e + e B factories (0.502 ± ps -1 ) m d measurement at Hadron Colliders Confirms initial state flavor tagging for later use in B s and m s measurements Mike Strauss The University of Oklahoma 26

27 B Oscillation Variables Opposite side b 1/3 Q<0 b B 0 π b B π + Same side b K (CDF) B s K + b (CDF) Mike Strauss The University of Oklahoma 27

28 Final State Ambiguities in SS Tag Mike Strauss The University of Oklahoma 28

29 B 0 Mixing with SS Tag A = (N RS N WS )/(N RS +N WS ) N RS :N(B 0 π + ) N WS :N(B 0 π ) m d = ± 0.052(stat) ± 0.030(sc) ± 0.012(syst) ps -1 Mike Strauss The University of Oklahoma 29

30 B 0 Mixing with SS Tag B µ D*X, D* D 0 π Visible Proper Decay Length: x M = L xy M B c /p T µd Preliminary Tagging purity: 55.8 ± 0.7 ± 0.8 % m d = ± 0.066(stat) ± 0.044(syst) ps -1 Mike Strauss The University of Oklahoma 30

31 B 0 Mixing with OS µ Tag Preliminary Decay Mode: B µ D*X, D* D 0 π Tagging: muon p T > 2.5 GeV/c cos φ(µ,b) < 0.5 Tagging efficiency: 4.8 ± 0.2 % Tagging purity: 73.0 ± 2.1 % Fit procedure Binned χ 2 fit m d = ± 0.055(stat) ± 0.049(syst) ps -1 Mike Strauss The University of Oklahoma 31

32 B 0 Mixing with Combined Tag Uses B l D ( * ) Lepton plus SVT trigger Combines: Same Side Pion Tagging Opposite Side Muon Tagging Opposite Side Jet Charge Tagging With and without vertex tagging Ten subsamples Tagged with SST Tagged with OST (3 samples) Tagged with SST and OST that agree (3 samples) Tagged with SST and OST that don t agree (3 sample) Tagging priority is determined Mike Strauss The University of Oklahoma 32

33 Efficiency and Dilution Channel ε(%) D(%) εd 2 (%) Only SST ± ± ± Only SMT 1.17 ± ± ± SST and SMT (agree) 1.34 ± ± ± SST and SMT (disagree) 1.22 ± ± ± Only JQT-SecVtx 2.73 ± ± ± SST and JQT-SecVtx (agree) 3.38 ± ± ± SST and JQT-SecVtx (disagree) 3.19 ± ± ± Only JQT-High P t ± ± ± SST and JQT-High P t (agree) ± ± ± SST and JQT-High P t (disagree) ± ± ± Total ± ± Preliminary m d = ± 0.037(stat) ± (sc)± 0.015(syst) ps -1 Mike Strauss The University of Oklahoma 33

34 B 0 Mixing with Combined Tag B 0 D* ± µ + X B D 0 µ + X Kππ mass - Kπ mass Dominated by B + events Mike Strauss The University of Oklahoma 34

35 B 0 Mixing with Combined Tag Soft muon tag If not tagged by muon: Opposite side jet charge Soft Pion Soft muon tag D 0 µ asymmetry Preliminary Jet charge and pion m d = ± 0.034(stat) ± 0.025(syst) ps -1 Mike Strauss The University of Oklahoma 35

36 B Hadron Lifetimes Naive quark spectator model: a 1 3 decay process common to all B hadrons. (NLO) QCD Heavy Quark Expansion predicts deviations in rough agreement with data Experimental and theoretical uncertainties are comparable Lifetime differences probe the HQE to 3 rd order in Λ QCD / m b Goal: measure the ratios accurately Mike Strauss The University of Oklahoma 36

37 B Hadron Lifetime Ratios τ(b )/τ(b 0 ) τ(b s )/τ(b 0 ) τ(λ b )/τ(b 0 ) τ(b baryon) /τ(b 0 ) From PDG lifetime ratio Mike Strauss The University of Oklahoma 37

38 Λ b Lifetime DØ DØ Preliminary New Measurement from DØ Λ b J/ψ Λ 0 DØ τ(λ b ) = ± ps τ(λ b )/τ(b d0 ) = ± CDF Preliminary from 2003: τ(λ b ) = 1.25 ± 0.26 ± 0.10 ps Mike Strauss The University of Oklahoma 38

39 B s Lifetime using B s J/ψ φ CDF DØ 250 pb -1 Preliminary Improvements since 2003: Selection minimizes stat syst 12 parameter maximum likelihood fit 240 pb -1 DØ analysis is similar to this CDF improved analysis Mike Strauss The University of Oklahoma 39

40 B s Lifetime using B s J/ψ φ CDF Preliminary DØ 250 pb -1 τ(b s ) = 1.369± ps τ(b s ) = ±0.020 ps Uses one exponential decay in the fit Mike Strauss The University of Oklahoma 40

41 B d Lifetimes Using B d J/ψ K s * 0 CDF DØ Preliminary 250 pb -1 τ(b 0 ) = 1.539±0.051±0.008 ps τ(b s )/τ(b 0 ) = 0.890± τ(b d0 ) = ±0.023 ps τ(b s )/τ(b ) = ±0.003 Mike Strauss The University of Oklahoma 41

42 B + Lifetime Using B J/ψ K + τ(b + ) = 1.662±0.033±0.008 ps τ(b + )/τ(b 0 ) = 1.080±0.042 Most systematic uncertainties cancel in the ratio Mike Strauss The University of Oklahoma 42

43 Lifetime Ratio τ (B + )/τ (B 0 ) Novel Analysis Technique using B µd c(*) X Directly measure ratio instead of individual lifetimes Split D 0 Kπ sample: D* + (with slow π + ) mainly from B 0 D 0 mainly from B + 12% B + 2% B S PV B 0 D *- µd0 µ + K + π D 0 - π - ν D* + D 0 86% B0 16% B 0 2% B S 82% B + Mike Strauss The University of Oklahoma 43

44 D 0 and D* + Candidates 109k inclusive B µ ν D 0 candidates 25k B µ ν D* candidates DØ RunII Preliminary, Luminosity = 250 pb ±750 D *- D 0 π µ + D 0 π - µ + D 0 π π ± Dominated by B + decays M(D 0 π)-m(d 0 ) (GeV/c 2 ) Dominated by B 0 decays Mike Strauss The University of Oklahoma 44

45 Lifetime Ratio τ (B + )/τ (B 0 ) Measure N(µD* + )/N(µD 0 ) in bins of VPDL In both cases fit D 0 signal to extract N Use slow pion only to distinguish B 0 from B + (not in vertexing, K- factors etc., to avoid lifetime bias) τ(b + )/τ(b 0 ) = ± 0.021(stat) ± 0.022(syst) Mike Strauss The University of Oklahoma 45

46 Lifetime Ratio τ (B + )/τ (B 0 ) Mike Strauss The University of Oklahoma 46

47 B Decay Angular Amplitudes Uses B s J/ψ φ; Uses B d J/ψ K* 0 Allows measurement of many parameters including polarization amplitudes and Γ s = 1/τ L 1/τ H B B H s L s = = p p B B s s + q q Initial particle or antiparticle B B s s = = ( B + B ) ( B B ) B B s s s s = = Mike Strauss The University of Oklahoma s s CP odd CP even ( ) H L B + B s ( ) H L B B s s s

48 Decay Modes Mike Strauss The University of Oklahoma 48

49 Transversity Angles The J/ψ rest frame KK defines (x,y) plane K + (K) defines +y direction Θ, Φ: polar & azimuthal angles of µ + Ψ: helicity angle of φ(k*) Extract polarization amplitudes: A 0 : Longitudinal A, A : Transverse Mike Strauss The University of Oklahoma 49

50 Angular Projections and fit for B s Decay Angular Distribution: 4 d P r = dρ dt 6 i= 1 A i g ρ =(Θ, Φ, Ψ) i () t f ( ρ ) i r Mike Strauss The University of Oklahoma 50

51 Decay Angular Distributions Mike Strauss The University of Oklahoma 51

52 B d Amplitudes vs BaBar/Belle Mike Strauss The University of Oklahoma 52

53 B s and B d Amplitudes DØ results coming soon For B 0 d A 0 = ± ± A = (0.473 ± ± 0.006) ei(2.86 ± 0.22 ± 0.04) A = (0.482 ± ± 0.014) ei(0.15 ± 0.15 ± 0.04) For B 0 s A 0 = ± ± A = (0.510 ± ± 0.013) e A = ± ± Mike Strauss The University of Oklahoma 53 i(1.94 ± 0.36 ± 0.03)

54 B s Mass and Lifetime Projections Unconstrained fit τ L = 1.05 ± 0.02 ps τ H = ± 0.03 ps Γ = 0.47 ± 0.01 ps Γ Γ = 0.65 ± Using SM and constrained fit: +65 m s = 125 ps 1 55 Mike Strauss The University of Oklahoma 54

55 Conclusions The Tevatron is working great, producing many B hadrons. We now have about 500 pb -1 DØ and CDF are measuring many properties of B hadrons that nicely complement those measured at B factories More exciting results are expected in the future Mike Strauss The University of Oklahoma 55

56 CDF Sensitivity Estimate SM BR(B s0 µ + µ - ) (3.4±0.5) 10-9 ; Mike Strauss The University of Oklahoma 56

57 Lambda Lifetimes Mike Strauss The University of Oklahoma 57

58 Constraining Unitarity Triangle From Colin Gay, FNAL Wine and Cheese, July 16, 2004 Mike Strauss The University of Oklahoma 58