Measurement of B (  (nS)   +   ) at CLEO István Dankó Rensselaer Polytechnic Institute representing the CLEO Collaboration 1 st Meeting of the APS topical Group on Hadronic Physics Fermilab, Oct 24-26, 2004 CLEOCESR hep-ex/ submitted to PRL

Motivation  Heavy bb-resonances  test Lattice QCD and other non- perturbative model predictions  establish the accuracy of these calculations.  Leptonic (  ee ) and total decay widths (  ) of Y(nS) are not well established.  ee : from integrated resonant hadron cross section  : too narrow to measure directly  B  is crucial to get  ! Y(nS) Res.  ee (keV) B  (%)  (keV) Y(1S) 1.32    1.8 Y(2S)    7 Y(3S) 0.48    3.5 PDG ‘04

 Verify lepton universality by comparing the decay rate to τ + τ –  new physics*?! *M. A. Sanchez, Mod. Phys. Lett. A 17, 2265 (2003), hep-ph/ Motivation (continue)  B  measures the relative strength of Y   *  l + l  to Y  ggg/γgg.  Also important to determine transition rates among the bb states since these are often measured in exclusive modes: Y(nS)   /  Y(mS) (  e + e  /  +   ). b b 2g γ b b 3g b b l+l+ l–l– γ*γ*   ++ ++

Analysis strategy  uncertainty due to luminosity systematics cancels out  large background from non-resonant processes (continuum)  Measure the decay rate to μ + μ – (Γ μμ ) relative to the decay rate to hadrons (Γ had =Γ – Γ ee – Γ μμ – Γ ττ ): then B  (assuming lepton universality, Γ ee = Γ μμ = Γ ττ ) is: and continuum subtraction using off-resonance samples:

CLEO detector  Data collected with the CLEO III detector at the Cornell Electron-positron Storage Ring (CESR) in Si Vertex Detector: 4-layer double sided Drift Chamber: 47 layers 93% of 4   p /p = p=5.3 GeV CsI Calorimeter (CC) 93% of 4   E /E = E=100 MeV Muon Counters (MUON) 85% of 4 p > 1 GeV

Data sample L (on), pb -1 L (off), pb -1 Scale (On/Off) Y(1S) Y(2S) Y(3S)  on-resonance sample : on the peak of each resonance within 2-3 MeV.  off resonance sample : MeV below each peak. On-res./Off-res. scale factor:

Selection of  +  - events Number of extra showers in CC < 2 suppress cascade decays Y(nS)  Y(mS)  0  0 /  Exactly 2 back-to-back tracks with net charge = 0,  cos  < 0.80 and 0.7 < p/E b < 1.15 Cosmic ray rejection: require tracks to come from interaction point, Bhabha/hadron rejection: using CsI Calorimeter (CC) and MUON info Efficiency for: Y   +   : 65% e + e    +   : 45%  +  MUON CC DR

Cosmic-ray suppression Remaining cosmic background:  % Separation between the tracks: < 2mm (x-y) < 5 cm (along z) Two-dimensional plot of the average distance from IP can be used to estimate cosmic background. m m Preliminary

Background from cascade decays on-res – off-res data MC (no cascade) On the Y(2S) and Y(3S), significant background due to cascade decays. The remaining background is estimated to be Y(2S): (2.9  1.5)% Y(3S): (2.2  0.7)% N(extra showers) M(  +   )/Ecm ++ N(extra showers) < 2  0000 ++ ++ Y(2S)  Y(1S)  0  0 /   Y(2S) Preliminary

Selected muon pairs Y(1S) On-res. data Off-res. data (scaled) On-res – off-res data Y(2S) Y(3S) Preliminary M( μ + μ – ) / Ecm

Data - MC comparison non-resonance data MC P/E beam resonance data MC (with FSR) MC (no FSR) resonance data MC (with FSR) MC (no FSR) GeV non-resonance data MC P(  + )/E(beam) E(matched shower) Preliminary

Rejecting e + e – → e + e – / μ + μ – /  events, beam-gas, beam-wall interactions: Selecting Y → hadrons # tracks > 2 E(visible)/E c.m. > 0.2 # tracks > 4E(CC)/E c.m. > 0.15 E(CC)/E c.m. < 0.75E(shwr1)/E beam < 0.75 Event vertex position: suppress beam-gas, beam-wall and cosmic background estimate the remaining beam-gas events. Efficiency for Y decays to hadrons: 96-98%. Continuum subtraction removes essentially all the remaining non-resonant background from two-photon fusion,,. Efficiency for Y(nS)   +   is ~26% (effective contribution is ~0.4  0.7%)

Interference Interference between resonance decay and continuum production of the same final state distorts the resonance shape. Interference effect is different for μ + μ – and hadrons (only qq interferes) hence the measured relative decay rate depends on E cm. Convolute the interference corrected BW shape with a Gaussian energy spread and a radiative tail to estimate the effect of interference. Y→μ + μ – Y→hadrons 1S: -1.6% 2S: -3.9% 3S: -1.8% Fractional correction to B μμ : MeV

Statistical and systematic uncertainties Y(1S)Y(2S)Y(3S) ε(had)1.6%1.3%1.4% N(had)0.2%0.3%0.4% ε(μμ)1.8% N(μμ)0.1%1.6%0.9% Scale(on/off)0.8%2.3%3.1% Interference1% Frac. systematic2.7%3.7%4.1%  Statistical uncertainties: subtraction of the scaled off-resonance data increases the stat. uncertainty!  Systematic uncertainties:  efficiency: detector modeling, trigger, MC statistics  N(events): background subtraction (cosmic, cascade, ττ)  Scale factor: 0.5% variation  Interference: variation in parameters and energy Y(1S)Y(2S)Y(3S) Fractional statistical uncertainty <1%1.5%3.0% Preliminary

B (Y(nS)   +   ) Y(1S)Y(2S)Y(3S) N(  )    2.7 ε(  )0.652  N(had)    0.01 ε(had)    B  (%) 2.49  0.02   0.03   0.07  0.10 Preliminary

Summary  CLEO has measured B  for Y(1S), Y(2S), Y(3S): 1S: (2.49  0.02  0.07)% PDG: (2.48  0.06)% 2S: (2.03  0.03  0.08)% (1.31  0.21)% 3S: (2.39  0.07  0.10)% (1.81  0.17)%  Br(1S) is consistent with PDG, but Br(2S) and Br(3S) is much larger.  Total decay width using  ee  had /  from PDG.  (1S) = (52.8  1.8) keV PDG: (52.5  1.8) keV  (2S) = (29.0  1.6) keV (44  7) keV  (3S) = (20.3  2.1) keV (26.3  3.5) keV  Results submitted to PRL (hep-ex/ ) Preliminary