Exploring the D* way
Ch. de la Vaissiere, 7 octobre 1998
- Explore channels other than the rare J/psi channel of the book study
- Find channels which are also be well signed and pure
- Exploit kinematical+geometrical constraints to achieve accuracy on
the vertex position of the reconstructed B and remove combinatorial background.
Sample and description of the analysis:
- 4000 Aslund B0-B0 events with one B0 decaying
in J/psi K pi.
- This sample is used to develop prototype kinematical
and geometrical fits algorithms.
- In this sample, there is only one useful B0
- We attempt to reconstruct decay chains
B0 --> D* +/- + pions
- The D* +/- are looked in the
D* +/- --&gONT COLOR="#0000FF">D* +/- --> D0
+ pi+/- mode.
- The D0 are searched in 4 modes
- K pi,
- K pi pi0,
- K pi pi0 pi0
- K pi pi pi
- Starting from D0 candidates
, we construct D* +/- then B0.
- Kinematical(mass, Ebeam)+geometrical fits are performed at each step
to purify the sample.
Results on D0
- Many candidates and important combinatorial background except for the
K pi mode.
- Geometrical and kinematical fits (including pi0s) are performed
- Plots of D0 fit
probabilities fits are given in the following figure
- After a probability cut at 1% one finds :
- 2935 accepted, 414 true D0
- 0.103 D0 per B0
- The average purity is only 14%
Results on D* +/- --> D0
+ pi+/- selection
- After a classical cut on Delta-M between the D*
and D0 masses, one finds:
- 499 accepted, 171 true D*
- 0.042 D* per B0 .
- The average D* purity is 35%
- One applies a fit with 2 kinematical constraints + 6(10) geometrical
constraints, and 4 unknowns ; 4(8) n.d.o.f.
- Plots of D*
fit probabilities fits are given in the following Figure
- After a probability cut at 1% one get
- 214 accepted, 168 true D*
- 0.042 D* per B0 .
- The average D* purity is 78%
Results on the reconstruction of B0
--> D* +/- + pions.
- We add to the D* , 1,3 charged pions +
0, 1,2 pi0s
- Raw Selection ; M(D* +/- + pions)
~ M(recoil) ~ M(B)
- 38 accepted, 27 coil) ~ M(B)
- We perform a full fit of the decay chain
- Highly constrained fit ; <n.d.o.f> = 11
- 24 candidates accepted, all fully reconstructed B0
- Accuracy on the vertex : < 60 microns> similar to golden Psi
- Plots of B0 fit probabilities
fits and accuracies are given in the following Figure
- Extrapolation of the 24 reconstructed B0
to 5M Upsilon(4S) events : 30 K B0
- Kinematical + geometrical fitting are efficient
to purify D*
and B0 .
- NB: Background may be higher in ordinary B0B0
- ASLUND may be optimistic, but it is not the end
of the road:
- In a near future addition ofIn a near future addition of
- channels with K0s
- extra D* +/-
--> D+/-+ pi0
- Reconstruct B0
- Problem of
B+/- lifetimes :
decay in D0 + pi0 or
D0 + gamma ;
- There is no equivalent to the golden chain D0
- Measuring of the B+/-
lifetimes with D* could
be more difficult.