Evidence for D0-D0bar Mixing

### Primary authors: Paul Karchin, Nagesh Kulkarni, and Mark Mattson

This web page summarizes the time dependent measurement of D0-D0bar mixing, using approximately 1.5 fb-1 of CDF II data.

This analysis is a continuation of the work done for the time independent measurement of D0 &rarr K+ &pi - Wrong Sign DCS Decays using 0.35 fb-1, where the fits were done with the assumption of no-mixing. The time independent DCS analysis was published in PRD RC 74, 031109 2006

Summary of the charm mixing analysis and results can be found in this public note (PDF).  "Right Sign" refers to the decay chain D* &rarr &pi+ D0, D0 &rarr K- &pi+, which results from the Cabibbo Favored decay of D0.
"Wrong Sign" refers to the decay chain D* &rarr &pi+ D0, D0 &rarr K+ &pi-, which results from Doubly Cabibbo Suppressed decays and charm mixing.

Plots:

• Ratio WS/RS D* versus proper decay time
EPS, PDF
• (Bayesian) probability intervals for the mixing parameters (y', x'2)
EPS, PDF
• impact parameter for RS D* (to distinguish D* coming from B decays)
EPS, PDF
• Right-sign KPi mass plot (to illustrate D0 signal)
EPS, PDF
• Right-sign mass difference plot (to illustrate RS D*)
EPS, PDF
• Wrong-sign mass diff plot (to illustrate WS D*)
EPS, PDF
The results are:
• Best fit for mixing parameters (uncertainties are combined stat. and syst.)
• Rd = (3.04 +/- 0.55) times 10-3
• y' = (8.5 +/- 7.6) times 10-3
• x'2 = (-0.12 +/- 0.35) times 10-3
• Fit chi2 = 19.2 for 17 dof
• Best fit assuming no mixing
• Rd = (4.15 +/- 0.10) times 10-3
• y' = x'2 = 0
• Fit chi2 = 36.8
• No-mixing probability (Bayesian statistics)
0.015% (equivalent to 3.8 standard deviations)
• No-mixing probability (Frequentist p-value)
0.013% (equivalent to 3.8 standard deviations)

Support plots: