Feb 18, 2005Kowalewski - FNAL seminar1 Results on Semileptonic B Decays from BaBar Bob Kowalewski University of Victoria and the Laboratori Nazionali di Frascati Currently at Università degli Studi di Roma

Feb 18, 2005 Kowalewski - FNAL seminar 2 Outline Context – B decays and the CKM sector Theoretical framework for inclusive B decays Inclusive b  cℓ ν decays and |V cb | Inclusive b  uℓ ν decays and |V ub | B  D * ℓ ν : FF and |V cb | B  π ℓ ν and |V ub | If I have time….

Feb 18, 2005 Kowalewski - FNAL seminar 3 B decays – a window on the quark sector The only 3 rd generation quark we can study in detail Access 3 of the 4 CKM parameters CKM matrix Cabibbo angle B d B d and B s B s oscillations, Radiative decays B lifetime, SL decay ~1 CP Asymmetries (phase)

Feb 18, 2005 Kowalewski - FNAL seminar 4 CKM picture  Unitarity Triangle   RtRt RuRu 

Feb 18, 2005 Kowalewski - FNAL seminar 5 Current status in ρ-η space Measurements are consistent with SM CP asymmetries from B factories dominate the determination of η Improved precision needed on |V ub | and other angles (α,γ) Also need information on B s oscillations! |V ub |/|V cb | band (±2σ) corresponds to σ=12%

Feb 18, 2005 Kowalewski - FNAL seminar 6 Y(4S) experiments e + e - → Y(4S) → B + B - or B 0 B 0 ; roughly 50% each B nearly at rest (βγ ~ 0.06) in 4S frame; decays overlap B energy = ½ c.m. energy; valuable constraint, since σ E ~5 MeV for e + e - beams on peak off peak (q=u,d,s,c) 2m B qq BB

Feb 18, 2005 Kowalewski - FNAL seminar 7 Our research tools PEP-2 performance has been marvelous Design was 30 fb -1 /year… BaBar L max (10 33 /cm 2 /s) 9.2 best day (pb -1 ) 681 total (fb -1 ) 244 design

Feb 18, 2005 Kowalewski - FNAL seminar 8 Our research tools PEP-2 performance has been marvelous; almost as good as KEK-B… Design was 30 fb -1 /year… BaBar Belle L max (10 33 /cm 2 /s) best day (pb -1 ) total (fb -1 ) design

Feb 18, 2005 Kowalewski - FNAL seminar 9 PEP-II luminaries Jonathan Dorfan Pier Oddone

Feb 18, 2005 Kowalewski - FNAL seminar 10 Our research tools BaBar has recorded >96% of delivered luminosity Trigger efficiency for BB events ~ 100% DIRC DCH IFR SVT CsI (Tl) e - (9 GeV) e + (3.1 GeV) BaBar 5-layer SVT tracker 40-layer Drift CHamber dE/dx Novel RICH based on total internal reflection (DIRC) CsI(Tl) crystal calorimeter (e ±, γ ) RPC and LST chambers in flux return for muon ID Angular coverage ~ 92% of 4 π in cm frame (challenge for ν reco) Boost: β 4S = 0.56 in lab

Feb 18, 2005 Kowalewski - FNAL seminar 11 Why we study SL decays |V ub | and |V cb | are crucial in testing CKM unitarity and SM mechanism for CP violation Presence of a single hadronic current allows control of theoretical uncertainties  precise determinations of |V ub | and |V cb | SL decays also provide a laboratory for probing our ability to calculate hadronic quantities in the presence of a mix of scales in QCD

Feb 18, 2005 Kowalewski - FNAL seminar 12 Inclusive Decays – the Big Picture

Feb 18, 2005 Kowalewski - FNAL seminar 13 Understanding inclusive SL decays The Operator Product Expansion provides a systematic method of separating perturbative from non-perturbative scales OPE + Heavy Quark symmetry [1]  HQE Heavy Quark Expansion now calculated to α s 2, m B -3 Essentially all we need to know for b  cℓ ν Coefficients of operators calculated perturbatively (EW and QCD); non-perturbative physics enters through matrix elements of operators [1] soft gluons cannot probe heavy quark properties

Feb 18, 2005 Kowalewski - FNAL seminar 14 Semileptonic B Decays & HQE contains b- and c-quark masses, m b and m c   2 related to kinetic energy of b-quark  G 2 related to chromomagnetic operator: B / B * mass splitting Darwin term ( ρ D 3 ) and spin-orbit interaction ( ρ LS 3 ) enter at 1/m b 3  = scale which separates effects from long- and short-distance dynamics r = m c /m b ; z 0 (r), d(r): phase space factors; A EW = EW corrections; A pert = pert. corrections (  s j,  s k  0 ) No 1/m b term! Several HQE schemes exist; op’s and coeff’s are scheme dependent free-quark rate

Feb 18, 2005 Kowalewski - FNAL seminar 15 Inclusive b  cℓ ν Measure electron momentum spectrum and mass of hadronic system in SL decay Determine moments to allow comparison with parton- level calculations (duality assumed) Calculations exist for the following: Fit for HQE parameters and |V cb | Mass of hadronic system Lepton energy spectrum

Feb 18, 2005 Kowalewski - FNAL seminar 16 Electron Energy Spectrum B A B AR data, 47.4 fb -1 at  (4S) fb -1 off-peak Select events with an electron having p * >1.4 GeV; study spectrum of 2 nd electron for p * > 0.5 GeV as f n of charge Unlike-sign events dominated by B  X c ev Like-sign events from D  Xev, B 0 mixing As done by ARGUS, CLEO… B A B AR PR D69: Unlike-sign Like-sign B A B AR

Feb 18, 2005 Kowalewski - FNAL seminar 17 Electron Energy Spectrum Determine E e spectrum Subtract B  X u e υ Correct for efficiency Correct for the detector material (Bremsstrahlung) Move from  (4S) to B rest frame Correct for the final state radiation using PHOTOS Calculate 0 th -3 rd E e moments for E 0 = 0.6 … 1.5 GeV B A B AR B A B AR PR D69: All but ~few % can be measured E e (GeV)

Feb 18, 2005 Kowalewski - FNAL seminar 18 Select events with a fully-reconstructed B Use ~1000 decay chains B  D[(n π )(mK)] - Flavor and momentum of “recoil” B known Find a lepton with E > E 0 in the recoil-B Lepton charge consistent with the B flavor m miss consistent with a neutrino All remaining particles belong to X c Improve m X with a kinematic fit (require m B2 =m B1 and m miss =0) Gives  (m X )= 350 MeV XcXc m ES [GeV/c 2 ] Events / 1.8 MeV/c 2 BABARBABAR Hadron Mass Moments B A B AR PR D69: Fully reconstructed B  hadrons lepton v L = 81 fb -1

Feb 18, 2005 Kowalewski - FNAL seminar 19 Hadron Mass Moments Unmeasured particles  measured m X < true m X Calibrate using simulation Depends (weakly) on decay multiplicity and m miss Validate in MC after applying correction Validate on data using partially reconstructed D *±  D 0   ±, tagged by the soft   ± and lepton Calculate 1 st -4 th mass moments with E 0 = 0.9 … 1.6 GeV B A B AR B A B AR PR D69: Validation:

Feb 18, 2005 Kowalewski - FNAL seminar 20 Inputs to OPE Fit m X moments E ℓ moments B A B AR B A B AR PRL 93: Error bars are stat. & syst. with comparable sizes Error bars (sys+stat) are smaller than point size

Feb 18, 2005 Kowalewski - FNAL seminar 21 Fit Parameters Calculation by Gambino & Uraltsev (hep-ph/ , ) Kinetic mass scheme to E ℓ moments m X moments 8 parameters to determine 8 moments available with several E 0 Sufficient degrees of freedom to determine all parameters without external inputs Fit quality tells us how well OPE works kinetic chromomagnetic Darwin spin-orbit B A B AR PRL 93:011803

Feb 18, 2005 Kowalewski - FNAL seminar 22 Fit Results m X moments E ℓ moments ● = used, ○ = unused in the nominal fit Red line: OPE fit Yellow band: theory errors B A B AR   2 /ndf = 20/15 B A B AR PRL 93:011803

Feb 18, 2005 Kowalewski - FNAL seminar 23 Fit Consistency OPE describes B A B AR data very well   2 /ndf = 20/15 Separate fits of E e and m X moments agree B A B AR B A B AR PRL 93:011803

Feb 18, 2005 Kowalewski - FNAL seminar 24 Fit Results Impressive agreement between data and theory ≈ identical results obtained in another renorm. scheme: Bauer, Ligeti, Luke, Manohar, Trott in hep-ph/ kinetic mass scheme with μ=1 GeV Fitted values consistent with external knowledge   2 /ndf = 20/15 Uncalculated corrections to  B A B AR PRL 93: precision on m b = 1.5% precision on |V cb | = 2%

Feb 18, 2005 Kowalewski - FNAL seminar 25 Inclusive |V cb | status BaBar result compared with previous measurements: Agrees with value coming from exclusive B  D * ℓ ν decays (from HFAG):

Feb 18, 2005 Kowalewski - FNAL seminar 26 Inclusive b  u Limiting factor in CKM precision tests; known much less well than |V cb | CKM suppressed – therefore harder to measure Challenge for experiment and theory Overview of theory uncertainties Overview of measurements Prospects

Feb 18, 2005 Kowalewski - FNAL seminar 27 Starting point: HQE Just like b  cℓ ν …, and with similar accuracy …until limited expt’l acceptance is considered Poor convergence of OPE in region where b  cℓ ν decays are kinematically forbidden Non-perturbative Shape Function must be used to calculate partial rates  = scale which separates effects from long- and short-distance dynamics A EW = EW corrections; A pert = pert. corrections (  s j,  s k  0 )

Feb 18, 2005 Kowalewski - FNAL seminar 28 Shape Function – what is it? light-cone momentum distribution of b quark: F(k + ) Property of a B meson; universal...but new “sub- leading” SFs arise at each order in 1/m b Consequences: changes effective m b, smears spectra Rough features (mean Λ, r.m.s. λ 1 ) are known Detailed shape, and especially the low tail, are not constrained

Feb 18, 2005 Kowalewski - FNAL seminar 29 Measuring the Shape Function SF can’t be calculated: measure it! Photon energy distribution in b  s γ probes b quark directly; same is in b  uℓ ν to O(1/m b ) It’s a hard measurement… SF moments are related to HQE parameters: Bosch-Lange-Neubert-Paz, hep-ph/ ; Benson-Bigi-Uraltsev, hep-ph/ Further constrains SF models NEW! Belle 3 models tried 1 st and 2 nd moment of SF determined

Feb 18, 2005 Kowalewski - FNAL seminar 30 Theory input for |V ub | At present, all |V ub | measurements based on inclusive SL decays use fully differential SL rate calculated to O(α S, m b -2 ) (DeFazio and Neubert, JHEP 06:017 (1999)) Input required includes values for the mean and r.m.s. of the Shape Function. In what follows we use as input the parameters determined by a fit (hep-ex/ ) to the Belle b  s γ spectrum: Λ = 0.66 GeV, λ 1 = GeV 2 + associated covariance; δ Λ ~ δm b ≈ 80 MeV

Feb 18, 2005 Kowalewski - FNAL seminar 31 Experimental approaches q 2 (GeV^2) E e (GeV)m X (GeV) Must suppress b  cℓ ν ; rate is 50x higher than signal Kinematic variables: E e, q 2 = (p e +p ν ) 2 and m X u E e easy, q 2 more difficult, m X u hardest b  c allowed

Feb 18, 2005 Kowalewski - FNAL seminar 32 Electron Endpoint 80 fb -1 on  (4S) resonance Select e ± in 2.0 < E ℓ < 2.6 GeV Larger signal acceptance  smaller theoretical error Accurate model of background is crucial (S/N ~ 1/14) Data taken below the  4S) resonance determine the light-flavor background Fit E e spectrum with b → uℓv, B → Dℓv, B → D * ℓv, B → D ** ℓv … Entries >2.2 GeV put in 1 bin to reduce signal model dependence Fully correct spectrum for efficiency, resolution, radiation, etc. Data (eff. corrected) MC Data (continuum sub) MC for BB background B A B AR hep-ex/ PRELIMINARY

Feb 18, 2005 Kowalewski - FNAL seminar 33 Electron Endpoint Fully corrected partial BF in Y(4S) frame: Translate  B into |V ub |: Significant decrease in theory uncertainty at lower E e (SF sensitivity is substantially reduced) E e (GeV)  B (10 -4 ) |V ub | (10 -3 ) B A B AR 2.0 – ± 0.29 stat ± 0.53 sys 4.40 ± 0.13 stat ± 0.25 sys ± 0.38 theo CLEO 2.2 – ± 0.15 exp ± 0.35 sys 4.69 ± 0.15 stat ± 0.40 sys ± 0.52 theo Belle2.3 – ± 0.11 exp ± 0.10 sys 4.46 ± 0.20 stat ± 0.22 sys ± 0.59 theo B A B AR hep-ex/ CLEO PRL 88: BELLE-CONF-0325

Feb 18, 2005 Kowalewski - FNAL seminar 34 E e vs. q 2 Use p ν = p miss in addition to p e  Calculate q 2 Given E e and q 2, maximum possible m X 2 is (in B frame) +small modifications for Y(4S) frame Resolution on p ν modest; some b  cℓ ν decays smeared to higher q 2 Acceptance ~25% less than cut on E e alone, but S/N=1/2 is much better B A B AR hep-ex/ q 2 (GeV^2) E e (GeV) b  cℓ ν allowed Accepted region

Feb 18, 2005 Kowalewski - FNAL seminar 35 E e vs. q 2 80 fb -1 on Y(4S) resonance Subtract off-peak data B → D (*) ev control samples used to validate ν reco Subtract BB background normalized by sideband Unfolded partial BF: Total b  uℓ ν BF: Translate to |V ub | B A B AR hep-ex/ normalize bkg extract signal ν resolution on B  D (*) ℓ ν PRELIMINARY

Feb 18, 2005 Kowalewski - FNAL seminar 36 b  uℓ ν in tagged events Same recoil technique as for b → cℓv  m X  moments Find a lepton (p ℓ > 1GeV) in recoil B Lepton charge consistent with the B flavor m miss consistent with a neutrino All left-over particles belong to X Improve m X with a kinematic fit Calculate q 2 of lepton-neutrino Calculate anything you want! Fully reconstructed B  hadrons lepton v X B A B AR hep-ex/

Feb 18, 2005 Kowalewski - FNAL seminar 37 Charm suppression cuts Suppress b → cℓv by vetoing D (*) decays D decays usually produce at least one kaon  Reject events with K ± and K S B 0 → D *+ (→ D 0   + )ℓ − v has peculiar kinematics D *+ momentum can be estimated from   + alone Calculate for all   +  Reject events consistent with m ν 2 = 0 Vetoed events are depleted in b → uℓv Use them to validate simulation of background distributions B A B AR hep-ex/

Feb 18, 2005 Kowalewski - FNAL seminar 38 Fitting m X B A B AR data, 80 fb -1 on resonance Simple fit in m X shows clear b → uℓv signal (S/N ~ 2/1 for m x < 1.55 GeV) Inclusive BF measured to be B A B AR hep-ex/ B A B AR

Feb 18, 2005 Kowalewski - FNAL seminar 39 m X vs. q 2 2-D fit to measure  B in {m X 8} Different theory uncertainty than pure m x cut Good resolution allows clean extraction of  B Signal event fraction into the “ box ” taken from theory B A B AR hep-ex/ PRELIMINARY Bauer et al. hep-ph/ DeFazio-Neubert, JHEP 06:017 (1999)

Feb 18, 2005 Kowalewski - FNAL seminar 40 Unfolding the m x distribution The shape of the m x distribution in b  uℓ ν decays carries information about SF parameters (b  uW akin to b  s γ ) Moments can also be used to test OPE, as in b  cℓ ν Unfolding accounts for resolution, uncertainties in theory input… Measured moments: M 2 ’ = 2 nd central moment M 0 (GeV)Value M1M ± GeV M2’M2’ ± GeV 2 M1M ± GeV M2’M2’ ± GeV 2

Feb 18, 2005 Kowalewski - FNAL seminar 41 BaBar Inclusive |V ub | Results Technique |V ub | × 10 3 E ℓ > 2.0 GeV 4.40 ± 0.13 stat ± 0.25 sys ± 0.38 theo E ℓ vs. q ± 0.23 stat ± 0.42 sys ± 0.32 theo m X vs. q ± 0.41 stat ± 0.40 sys ± 0.23 theo m X < 1.55 GeV 5.22 ± 0.30 stat ± 0.31 sys ± 0.43 theo average4.80 ± 0.47, χ 2 = 3.5/3 B A B AR hep-ex/ B A B AR hep-ex/ B A B AR hep-ex/ S/N increasing acceptance increasing statistics decreasing Statistical correlation between the m X and m X -q 2 results is 72%. Others correlations negligible Each measurement has a somewhat different sensitivity to SF Average accounts for correlated systematic/theory errors

Feb 18, 2005 Kowalewski - FNAL seminar 42 m X vs. q 2 Inclusive |V ub | in Perspective |V ub | is measured to ±9%! E ℓ endpoint m X fit E ℓ vs. q 2 Some results have been re-adjusted by the Heavy Flavor Averaging Group

Feb 18, 2005 Kowalewski - FNAL seminar 43 What to look for in future Improved precision of |V ub | requires re-evaluation of theoretical uncertainties Precision on SF parameters Λ and λ 1 will improve, due to better b  s γ spectra and the use of OPE parameters determined in b  cℓ ν transitions Power (1/m b ) corrections differ between b → uℓv and b → s  Quantitative estimates have appeared in literature Weak annihilation may have a large (20%?) effect at high q 2 Difference between B 0 and B + needs to be measured Expect |V ub | averages to approach 5% uncertainty soon

Feb 18, 2005 Kowalewski - FNAL seminar 44 Exclusive B  X c ℓ ν decays Measure |V cb | using a completely independent theoretical framework from that of inclusive decays Test HQET Reduce background uncertainty for other measurements, notably |V ub |

Feb 18, 2005 Kowalewski - FNAL seminar 45 B  D * ℓ ν BF and |V cb | Heavy Quark Effective Theory gives us a framework for understanding B  X c ℓ ν transitions: FF for decay depends only on q 2 : universal Isgur-Wise f n FF unknown, but normalization is unity in heavy quark limit (m b =m c =∞) at “zero-recoil” point: Can be used with both B  D * ℓ ν and B  Dℓ ν decays bc e ν beforeafter Light d.o.f. unchanged!

Feb 18, 2005 Kowalewski - FNAL seminar 46 What we measure B  D * ℓv decay rate is given by F(1) = 1 in the heavy-quark limit; Lattice calculation gives Form of F(w) unknown Parameterized with  2 (slope at w = 1) and ratios R 1 and R 2 of FFs that are ~ independent of w Use R 1 and R 2 determined by CLEO, PRL 76 (1996) 3898 Measure d  /dw to fit F(1)|V cb | and  2 phase space form factor D * boost in the B rest frame Hashimoto et al, PRD 66 (2002)

Feb 18, 2005 Kowalewski - FNAL seminar 47 B  D*ℓv Sample B A B AR data, 80 fb -1 on  (4S) Find events with a D *+ and a lepton with 1.2 < p ℓ < 2.4 GeV/c Background Fake D * : use D * – D mass difference True D * but not B  D * ℓv: use variable sensitive to what accompanies D * ℓ: D*l D**l Uncorrelated leptons Continuum Fake D* B A B AR electron sample B A B AR hep-ex/

Feb 18, 2005 Kowalewski - FNAL seminar 48 Determination of F(1)|V cb | Correct for efficiency  w distribution Slow pion (from D * decays) efficiency depends on w Fitting dN/dw gives FF ratios also under study: from ICHEP, being refined B A B AR hep-ex/ R 1 = ± ± R 2 = ± ± PRELIMINARY

Feb 18, 2005 Kowalewski - FNAL seminar 49 Determination of |V cb | B A B AR result compares well with existing measurements  Results adjusted to use common inputs Using F(1) = 0.91 ± 0.04, the world average is Agrees with the inclusive measurement

Feb 18, 2005 Kowalewski - FNAL seminar 50 Exclusive B  X u ℓ ν New unquenched Lattice calculations of B  π ℓ ν FF now available New measurement techniques give much better S/N by reconstructing some/all of the non-signal B Expect improvements soon It will be interesting to compare inclusive and exclusive determinations of |V ub |

Feb 18, 2005 Kowalewski - FNAL seminar 51 Conclusion |V ub |/|V cb | band (±2σ) corresponding to σ=5% ρ BaBar measurements on semileptonic decays are improving our knowledge of |V cb | and |V ub |: |V cb | from inclusive decays: ±2%, HQE validated! |V ub | from inclusive decays: error <10%, can still improve Expect significant progress this year on Inclusive b  uℓ ν decays Exclusive B  X u ℓ ν decays

Feb 18, 2005 Kowalewski - FNAL seminar 52 Backup Slides

Feb 18, 2005 Kowalewski - FNAL seminar 53 Performance snapshot Tracking: [σ(p t )/p t ] 2 = (.0013 p t ) Ecal: σ E = 3% at 1 GeV, 4.5% at 0.1 GeV PID: control samples used to evaluate ε and fake rate K- π separation in DIRC measured using D 0  K π decays σ ΔE ~15MeV