B Flavor Identification Using Opposite Side Muons

Denys Usynin, Matthew Jones, Usynin, Joseph Kroll
University of Pennsylvania

Barry Wicklund
Argonne National Lab

Introduction

The soft muon tag is one of the first methods for b-flavor identification that has been investigated in detail at CDF in Run-II. This analysis measures its performance in the lepton+SVT data samples using techniques described in the sample composition analysis.

The l+SVT trigger provides high statistics samples of semileptonic b and c decays with additional backgrounds from prompt tracks that produce fake lepton triggers. The bottom component can be isolated by requiring that 2 < M(l,SVT) < 4 GeV and by subtracting the prompt background using the signed impact parameter of the SVT track. The resulting sample is equivalent to a pure b sample in which the opposite side muon tag can be studied in detail.

Soft Muon Identification

Muons are identified using any combination of the CMU,CMUP,CMP,CMX or IMU muon subsystems. Loose matching requirements are made to associate tracks with stubs in these subdetectors. Muons are classified by which subdetectors were associated with the muon. The following list defines the muon types:

Muon type	Definition
CMU	Satisfies CMU matching cuts but not fiducial in CMP
CMP	Matched to stub in CMP but none in CMU
CMUP	Matched to stubs in both CMP and CMU
CMX	Matched to CMX. CMU/IMU stub required if fiducial in CMU/IMU.
IMU	Matched to stub in IMU.

Low quality muons, such as those that are fiducial in CMP, but did not have an associated stub, have dilution consistent with zero and are not used for tagging.

In this analysis, IMU muons are matched to COT tracks which restricts the available acceptance. Some additional improvements should eventually result from the use of silicon standalone tracks matched to IMU stubs.

Supporting documentation

More text can be found in CDF note 6483.

Blessed Results

After correcting for mixing and sequantial decays on trigger side the epsilon-D^2, expressed in percent and averaged over e+SVT and mu+SVT is:
eps D^2 = 0.660 +- 0.093 %
Comparison with Run-I (essentially from note 3808):
eps D^2 = 0.584 +- 0.082 %

Dilution, efficiency and epsilon-D^2 for different muon types measured in e+SVT and mu+SVT samples separately:

Table_2.gif
Table_2.tex

Dilution, efficiency and epsilon-D^2 for each muon subsystem average over e+SVT and mu+SVT samples:

Table_3.gif
Table_3.tex

Dilution as a function of pT-rel of the tag muon with respect to the opposite side jet. The data points with pT-rel < 0 correspond to tracks not associated with a jet.

d_vs_ptrel.eps
d_vs_ptrel.gif

This is the weighted average of the e+SVT and mu+SVT samples.

d_vs_ptrel_all.eps
d_vs_ptrel_all.gif

Parameterization of dilution as a function of pT-rel:

Table_1.gif
Table_1.tex