MEASUREMENTS FOR THE FLAVORSPECIFIC CP VIOLATING ASMMETRY IN

MEASUREMENTS FOR THE FLAVOR-SPECIFIC CP VIOLATING ASMMETRY IN DECAYS Implications for Te. V-scale workshop, July 2012 Z. Xing ON BEHALF OF THE LHCb COLLABORATION 1

The LHCb detector Dipole magnet ↑↓ Muon ID 10 – 300 mrad p p tracking 16/7/2012 Z. Xing Implications for Te. V-scale workshop 2

A little bit of theory I n. Mass eigenstates n. Time evolution of Flavor Eigenstates 16/7/2012 Z. Xing Implications for Te. V-scale workshop 3

A little bit of Theory II n Observable quantities are differences in masses and decay widths. In addition, we have the quantity n We can access as by measuring asymmetries in flavor specific final states, for example semileptonic decays. Standard Model predictions 16/7/2012 Z. Xing Implications for Te. V-scale workshop A. Lenz ar. Xiv: 1205. 1444 4

What we measure We need to account for detection asymmetries 0. 2% in our case Production asymmetry ~1% ε(t) Untagged semileptonic asymmetry MC 16/7/2012 Z. Xing Implications for Te. V-scale workshop 5

Key elements of the analysis q q n Bs production asymmetry does not affect the measurement: fast oscillations suppress this effect by 0. 2% out of the Θ (1%) initial asymmetry P. L. B 713 (2012) 186 Prompt Ds have negligible production asymmetry ar. Xiv: 1205. 0897 v 2 (-0. 33± 0. 22± 0. 10) % and represent a small fraction of the signal Backgrounds are small and have negligible asymmetries We have MAGNET UP and MAGNET DOWN data samples of almost equal size, which allows to average out residual charge asymmetries in detection efficiency (arithmetic average) Note: Small means ~ 1% 16/7/2012 Z. Xing Implications for Te. V-scale workshop 6

Analysis method PHILOSOPHY: DATA DRIVEN ANALYSIS ALL CORRECTIONS ARE DERIVED FROM DATA, WITH TWO INDEPENDENT METHODS, WHENEVER POSSIBLE q q Determination of the signal yields with 2011 full data set: 447 pb-1 collected with magnet polarity UP and 595 pb-1 collected with magnet polarity DOWN mass cut provides almost equal kaon momentum spectra. Detailed analysis of background sources, mostly based on data. Efficiency ratio derived from calibration samples, can be expressed as 16/7/2012 Z. Xing Implications for Te. V-scale workshop 7

Four calibration samples n P. L. B 713 (2012) 186 Partially reconstructed ar. Xiv: 1205. 0897 v 2 demonstrated that tracking efficiency ratio ε(π+)/ε(π-) does not depend upon particle momentum and p. T. Since π and μ have opposite charges in , the tracking asymmetries cancel out. 16/7/2012 Z. Xing Implications for Te. V-scale workshop 8

Continued n n n Kinematically Selected in samples triggered by hadronic B decays not including J/ψ in the final state (next slide for details) Muon Selected where a detached J/ψ is found by combining one track (probe muon) with an opposite sign track that is well identified as a muon (tag muon) B→D+μ-X with D+→K-π+π+ for software trigger checks 16/7/2012 Z. Xing Implications for Te. V-scale workshop 9

Kinematically selected Kinematically Selected Muon Selected q selected in events triggered by b-hadron not decaying into a J/ψ, using only kinematic selections q. We divide calibration and signal samples into 50 Muon p-px-py bins to reduce sensitivity to kinematic biases 16/7/2012 Z. Xing Implications for Te. V-scale workshop 10

Efficiency ratios n The KS J/ψ is used to determine ε(μ+)/ε(μ-) for the muon identification algorithm and muon dependent triggers Muon identification Magnet up data Efficiency ratio for μ hardware trigger Logarithmic scale 16/7/2012 Z. Xing Implications for Te. V-scale workshop 11

The signal yields q Signal yields are extracted through invariant mass fits. q Default method includes PDFs for Ds and D+ signals, and 2 nd order polynomials for combinatorial background. q Alternative PDFs used both for signal and background for sys. checks. Magnet UP Magnet Down 40, 945± 285 55, 755± 278 39, 849± 239 56, 447± 294 Dsμ candidates – magnet DOWN data D+ D- Logarithmic scale 16/7/2012 Z. Xing Implications for Te. V-scale workshop 12

Misidentified muon background q This background is assessed with B→DsπX and Ds. KX samples q K & π probabilities to be identified as μ derived from D*+→D 0π+, D 0→K-π+ calibration sample q For each muon momentum bin this background <1% of signal q Asymmetry in K±(π±)→μ± fake rates ~1%, so product is ~10 -4 K and π probabilities to be identified as μ 16/7/2012 Z. Xing Implications for Te. V-scale workshop 13

Background from B→Ds. DX & B→Ds. KμνX D (D 0, D+, Λc etc. ) q Background Dsμ combinations produced by b�ccs- where Ds originates from W�cs - charmed hadron decays semileptonically. and the q Reconstruction efficiency is highly suppressed due to detached and softer muons q Taking into account 33% suppression factor due to B 0 mixing, we get an overall background fraction of (3. 5± 0. 9)% q Assuming 1. 5% b production asymmetry, we assign 0. 05% uncertainty to this source q Background from B→Ds. KμνX is (3. 3± 0. 9)% and production asymmetry tend to cancel with previous contribution➯ we do NOT assign additional systematic uncertainty. 16/7/2012 Z. Xing Implications for Te. V-scale workshop 14

The measured asymmetry Magnet Up Preliminary Ameas=(0. 10± 0. 41± 0. 15)% Magnet Down Preliminary Ameas=(-0. 34± 0. 35± 0. 13)% 16/7/2012 Z. Xing Implications for Te. V-scale workshop Average Preliminary Ameas=(-0. 12± 0. 27± 0. 10)% 15

Systematic uncertainties (prelim. ) Source σ(Ameas)(%) Signal modeling in Ds mass fit 0. 06 Background from other b hadrons 0. 05 Kinematics difference between π and μ 0. 06 Momentum difference between same sign and opposite sign kaons 0. 02 Varying run conditions between field-up and field-down 0. 01 Muon corrections 0. 05 Muon misidentification 0. 01 Muon related software trigger biases 0. 05 Statistical uncertainty on the efficiency ratios 0. 10 Total 0. 16 n. Tracking efficiency asymmetry coupled to the small kinematics difference between π+ and μ- gives a 0. 06% effect from the non-perfect cancellation 16/7/2012 Z. Xing Implications for Te. V-scale workshop 16

Results [lhcb-conf-2012 -022] =2 Ameas ➪we obtain the preliminary result ➪Most precise measurement of ➪In agreement with Standard Model’s prediction (Lenz, ar. Xiv: 1205. 1444) 16/7/2012 Z. Xing Implications for Te. V-scale workshop 17

LHCb, B-factories and D 0 [Phys. rev. d 84(2011)052007] D 0 dimuon result: a) is consistent with SM➯ (compatible at 1. 8 σ with our result) b) is consistent with SM➯ (compatible with ϒ(4 S) at 1. 6 σ) c) ➯ (compatible with ϒ(4 S) and LHCb results at 1. 4 σ ) 16/7/2012 Z. Xing Implications for Te. V-scale workshop 18

Conclusions q LHCb reports the preliminary result q This result is consistent with the SM prediction of ~ 0 Not the end of the story: more Ds decay modes will be added soon and additional ~ 2. 2 fb-1 expected by the end of the year q 16/7/2012 Z. Xing Implications for Te. V-scale workshop 19

Some other recent CPV results in LHCb q CPV in multibody charmless B decays 16/7/2012 Z. Xing Implications for Te. V-scale workshop 20

Continued 16/7/2012 Z. Xing Implications for Te. V-scale workshop 21

THE END SO FAR 16/7/2012 Z. Xing Implications for Te. V-scale workshop 22

Systematic checks n. We have studied the sensitivity of the raw asymmetry (not corrected for efficiency ratios) as a function of several variables (time, muon pt, event multiplicity etc. ) and we found no evidence of systematic biases. 16/7/2012 Z. Xing Implications for Te. V-scale workshop 23

Ds decay time n. Ds decay time dependence checked in terms of two exclusive software triggers Software trigger I 16/7/2012 Z. Xing Implications for Te. V-scale workshop Software trigger II Combined 24

Prompt Ds background Small prompt Ds fraction (1. 4± 0. 01)% & Ds production asymmetry (0. 33± 0. 22± 0. 10)% ar. Xiv: 1205. 0897 [hep-ex] it can be neglected 16/7/2012 Z. Xing Implications for Te. V-scale workshop 25

Misidentified muon background Hadrons identified as K Full μ momentum range [6, 100] Ge. V 16/7/2012 Z. Xing Implications for Te. V-scale workshop 26