Measurement Of D~0-D~0Mixing And CP Violation In D~0â†'K_s~0Ï€~+Ï€~-Decays

We report a measurement of D0-D0mixing parameters （x, y） and a search for indirect CP violation （|q/p|, arg（q/p）） through a time-dependent amplitude analysis of D0â†'KS0Ï€+Ï€-decays. The results are based on921fb-1of da-ta accumulated with the Belle detector at the KEKB asymmetric-energy e+e-collider.We reconstruct the D0mesons through the decay chain D*+â†'D0Ï€s+and D0â†'KS0+Ï€-, where Ï€s is referred to as the slow pion. The charge of Ï€s is used to tag the flavor of the D meson. The KS0candidates are reconstructed in the Ï€+Ï€-final state. The D0candidates are reconstructed by combining each KS0candidate with two oppositely charged pion candidates. A D*+candidate is reconstructed by combining the D0candidate with a low-momentum charged track （theÏ€s+candidate）. To suppress the combinatorial background and BB events, we require the D*+momentum in the center-of-mass （CM） frame to be greater than2.5GeV/c and3.1GeV/c for the data taken at the CM energy of T（45） and r（5S） mass, respectively. The proper decay time of the D0candidate is calculated by projecting the vector joining the production and decay vertices,（?）, onto the D0momentum vector pD in the lab frame. The uncertainty of the proper decay time （σt） is evaluated from the corresponding vertex fit covariance matrices.We determine the signal yield from a two-dimensional fit to the distribution of the D0invariant mass M=MKS0Ï€+Ï€-and kinetic energy released in the D*decay Q=（MKS0Ï€+Ï€-Ï€s—MKS0Ï€+Ï€--mÏ€s）·c2. The backgrounds are classified into two types:random irs background and combinatorial background. We find1231731±1633（stat.） signal candidates with a purity of95.5%in the signal region defined as|M-mDo|<15MeV/c2,5.75MeV<Q<5.95MeV. Mixing parameters are extracted from an unbinned maximum likelihood fit to m+2, m₂and the decay time t for the events selected in the signal region. The reconstruction efficiency over the DP plane is described by a cubic polynomial of m+2and m₂determined from a large MC sample of signal events.The Dalitz amplitudes Af and Af are expressed as a sum of quasi-two-body amplitudes. For the P-and D-wave decays, we include12intermediate reso-nances described by relativistic Breit-Wigner parameterizations with mass de-pendent widths, Blatt-Weisskopf penetration factors as form factors and Zemach tensors for the angular dependence. For the ππ S-wave dynamics, we adopt the K-matrix formalism with P-vector approximation. For the Ks0Ï€S-wave, we follow the same description as in.We validate the fitting procedure with fully simulated MC experiments. The fitter returns the mixing parameters consistent with the inputs for signal samples with and without background events included.In our final fit to the data sample, we set the free parameters to be （x, y）, the D0lifetime（?）, the parameters of the proper decay time resolution function, and the amplitude model parameters. We extract the mixing parameters assuming no CPV, where the errors are statistical, experimental systematic, and systematic due to the amplitude model, respectively. The statistical correlation coefficient between x and y is0.012. The significance of D0-D0mixing is estimated to be2.5standard deviations relative to the no-mixing point （x=0, y=0）. We also determine the D0mean lifetimer=（410.3±0.6） fs, in agreement with the world average.We also search for CPV in decays. The CPV parameters|q/p|, arg（q/p） are included in the PDF. The values for the mixing parameters from this fit are essentially identical to the ones from the CP-conserved fit. The resulting CPV parameters are which are the most accurate values of the|q/p|and arg（q/p） parameters in a single experiment. The values are consistent with no CPV.