November 2001CP Violation at Belle Observation of CP Violation at Belle HEP Seminar Brookhaven National Lab Daniel Marlow Princeton University November 8, 2001

November 2001CP Violation at Belle Talk Outline Introduction/Review (with apologies to the experts) The KEK-B Collider The Belle Detector Indirect CP Violation Measurement/Analysis Conclusions

November 2001CP Violation at Belle Physics Motivation Direct CP violation is perhaps the simplest case. Consider the CP mirror processes: The CP asymmetry is defined as The decay amplitudes are Note that the weak phase changes sign.

November 2001CP Violation at Belle Direct CP Violation Note that Yielding We need some sort of interference, two amplitudes, for example

November 2001CP Violation at Belle Direct CP Violation Despite its conceptual and experimental simplicity, there are two problems with direct CP violation: Cases where there are two comparable amplitudes that are large are (probably) rare. The strong phases are poorly understood, making it difficult to extract the weak (KM) phases that are of greatest interest. One is thus led in the direction of asymmetric e + e - B factories aimed at the study of indirect CP violation.

November 2001CP Violation at Belle Indirect CP Violation In this approach we provide the interference using state mixing, i.e.,

November 2001CP Violation at Belle Indirect CP Violation Thus for a decay where is a CP eigenstate, we have two “indistinguishable” decay paths Working through the algebra, yields a time- dependent CP asymmetry Where and are the weak phases for the mixing and decay diagrams, respectively and

November 2001CP Violation at Belle The Kobayashi-Maskawa Mixing Scheme Where the second matrix is the Wolfenstein parameterization. Quark mixing is described via The “d b” unitarity relation yields

November 2001CP Violation at Belle CP Phases in the Gold-Plated Mode Decay Mixing The KM phase comes from the mixing.

November 2001CP Violation at Belle The Unitarity Triangle The gold-plated mode determines the angle which is also called.

November 2001CP Violation at Belle Constraints

November 2001CP Violation at Belle Existing Constraints Höcker et al. hepex/ Eur.Phys.J. C21 (2001) sin2  results are average prior to July They are not included in the fit.

November 2001CP Violation at Belle The Measurement tag CP Need to: Measure momenta ID leptons & K’s Measure vertices

November 2001CP Violation at Belle B0B0 B0B0 The Measurement The time-dependent asymmetry appears mainly as a mean shift in the distribution between events tagged as decays and events tagged as decays.

November 2001CP Violation at Belle The KEKB Asymmetric e + e - Collider KEKB is similar to PEP- II in many ways, although there is an important difference in the way in which the beams are brought into collision.

November 2001CP Violation at Belle KEK B Asymmetric Collider Two separate rings e + (LER) : 3.5 GeV e  (HER) : 8.0 GeV E CM : GeV at  (4S) Luminosity target: cm -2 s -1 achieved: 5.2 x cm -2 s -1 ±11 mrad crossing angle Small beam sizes:  y  3  m;  x  100  m asymmetric e + e  collider

November 2001CP Violation at Belle Beam Crossing Schemes By colliding the beams at an angle, the KEK-B design achieves a simplified interaction region. Moreover, every RF bucket (2 ns spacing) can be filled to achieve maximum luminosity. However, this approach risks destabilizing couplings between transverse and longitudinal modes of the machines. A crab-crossing cavity is under development in case this proves to be a problem.

November 2001CP Violation at Belle Machine Performance Reported here Summer ’01 shutdown Integrated/day Total Integrated

November 2001CP Violation at Belle Typical~good day.

November 2001CP Violation at Belle Another Typical Day This is probably a bit worse than usual.

November 2001CP Violation at Belle KEKB/PEP II Luminosity Bakeoff KEKBPEP-II Peak Shift92.8 pb pb hour277 pb pb -1 7 day1661 pb pb -1 Integrated37.6 fb fb -1 As of ~November 5, 2001 Comparisons not quite apples to apples.

November 2001CP Violation at Belle KEKB Future Prospects Chief limitation recently has been the photo- electron instability in the LER. This has been reduced, although not fully mitigated by installing a solenoidal winding around much of the beam pipe (last ~500 m this summer). Past history has shown that the situation slowly improves, in a manner consistent with a reduction in quantum efficiency of the beam-pipe wall brought that is reminiscent of “scrubbing.” If that is the case, we can expect additional modest gains in luminosity, but it is hard to know.

November 2001CP Violation at Belle

November 2001CP Violation at Belle An international collaboration involving about 10 Countries, 50 Institutes, & 250 People The BELLE Collaboration

November 2001CP Violation at Belle Detector Performance  Silicon Vertex Detector   ~  55  m for 90 o  Central Drift Chamber   p /p ~ 1GeV/c    (dE/dx)  ~  7   K  id up to p lab =3.5 GeV/c  TOF (  95 ps)  Aerogel (n = 1.01 ~ 1.03)  , e  with CsI crystals   E /E  ~ 1GeV  KL and   with KLM (RPCs)    : effic. > 90% ; ~2% fakes

November 2001CP Violation at Belle Mass Reconstruction Belle

November 2001CP Violation at Belle CsI EM Calorimeter Performance

November 2001CP Violation at Belle Vertex Detector  ~  55  m for 90 o Occupancy at present luminosity ~4% Upgraded SVD coming: 4 layers Radiation hard Level 1 trigger Summer ’02 installation Excl. mode B lifetime

November 2001CP Violation at Belle Belle: ToF & Aerogel The ToF has a resolution of  =100 ps, which provides  /K separation up to about 1.2 GeV/c, at which point the ACC array kicks in. The index of refraction of the counters varies as a function of angle, reflecting the boosted CM.

November 2001CP Violation at Belle Belle: Aerogel Barrel Module

November 2001CP Violation at Belle Particle ID (dE/dx, ToF, & Aerogel)

November 2001CP Violation at Belle PID Bake Off BaBar Belle Looks like BaBar does better.

November 2001CP Violation at Belle Muon Identification Muons are important for the same reason. They are identified by the way in which they penetrate the iron return yoke of the magnet. The gaps are instrumented with RPCs.

November 2001CP Violation at Belle Putting it all together

November 2001CP Violation at Belle Analysis Flowchart

November 2001CP Violation at Belle CP Mode Sample

November 2001CP Violation at Belle J/  Reconstruction Require one lepton to be positively identified and the other to be consistent with lepton hypothesis. For e + e -, add any photon within.05 of electron direction.

November 2001CP Violation at Belle B Reconstruction

November 2001CP Violation at Belle Other Modes M bc Total events = 76 Bkgd  9 evts (12%)

November 2001CP Violation at Belle  ’ Modes Both leptons tagged.

November 2001CP Violation at Belle Other Charmonium Modes M(l  l   )  M(l  l  ) 1st observation of inclusive B   c2 X  c2  c1

November 2001CP Violation at Belle CP Even Mode (B  J/  K L ) Reconstruction K L hits KLKL Assume (2-body) kinematics. Look for K L recoiling from J/  hits in RPCs cluster in ECL Remove positively tagged background modes: J/  K +, J/  K *, etc. Plot

November 2001CP Violation at Belle Detection Inclusive “K L ’s”

November 2001CP Violation at Belle B  J/  K L Reconstruction 61% signal purity

November 2001CP Violation at Belle CP Mode Summary

November 2001CP Violation at Belle Flavor Tagging We need to know the flavor of the spectator B 0. Track level tags High momentum leptons Medium momentum K  High-momentum   (from e.g., ) Low-momentum   (from D*’s). Need to take into account multiple tags and correlations.

November 2001CP Violation at Belle Flavor Tagging

November 2001CP Violation at Belle Flavor Tagging Data Monte Carlo

November 2001CP Violation at Belle Flavor Tagging Two measures of tagging performance:  = efficiency w = wrong-tag fraction Amplitude of mixing oscillation depends on w

November 2001CP Violation at Belle Vertexing tag CPCP Reality Cartoon The B lifetime is of the same order as the vertex resolution, so the effect is quite subtle.

November 2001CP Violation at Belle Vertexing Common track requirements –# of associated SVD hits > 2 –Use run-dependent IP For CP-side, use J/   l + l - tracks –Reject poorly fit events. For Tag-side, use tracks with : –|  z|<1.8mm, |  z |<500  m, |  r|<500  m –Iterate: discard worst track until fit is acceptable. Require |z CP - z tag |<2 mm (  10  B )

November 2001CP Violation at Belle Test of Vertex Resolution

November 2001CP Violation at Belle Fit to Lifetime

November 2001CP Violation at Belle Fitting Signal Background Response function

November 2001CP Violation at Belle Fitting

November 2001CP Violation at Belle The Data The first order effect is a mean shift between positively and negatively tagged samples.

November 2001CP Violation at Belle The Fitted Result PRL 87, (2001)

November 2001CP Violation at Belle Sources of Systematic Error Vertex algorithm  0.04 Flavor tagging  0.03 Resolution function  0.02 K L background fraction  0.02 Background shapes  0.01  m d and  B0 errors  0.01 Total  0.06

November 2001CP Violation at Belle Time-Dependent Asymmetry One can plot the excess/deficit on a bin-by-bin basis. First we start with a non-CP sample to look for false asymmetries.

November 2001CP Violation at Belle Subsample Dependence

November 2001CP Violation at Belle Subsample Dependence

November 2001CP Violation at Belle Comparison to other sin2  1 Measurements It looks like CDF had it right!

November 2001CP Violation at Belle Comparison to Other CKM Results BaBar

November 2001CP Violation at Belle Future Prospects Oide

November 2001CP Violation at Belle Conclusions CP violation in the B system has been observed at the  level. Our data are consistent with the SM. The KEKB accelerator is working very well and continues to improve. There will be lots of good physics to come.

November 2001CP Violation at Belle Additional Slides

November 2001CP Violation at Belle Two Pseudoscalar Modes

November 2001CP Violation at Belle What if | |  1 ? If |λ| is allowed to float, (i.e. a cos(Δm t) term) The CPV asymmetry is unchanged. In general we have Note: if | |  1, then the gold- plated mode isn’t really gold plated.

November 2001CP Violation at Belle K L Details Two classes: KLM+ECL and ECL only (397 and 172 entries, respectively) Fitted yield is 346 events. Background composition: 71% non-CP modes. The remainder is ψK L π 0 (13%), ψK s (10%) which are both CP=-1 and 5%(ψK, χ c1 K s, ψπ 0 ) which are CP=+1. p B * background shape is used to find bkg level.

November 2001CP Violation at Belle

November 2001CP Violation at Belle Particle ID (dE/dx, ToF, & Aerogel)