ECAL analysis (very preliminary results)

1
ECAL analysis (very preliminary results)

• L. Morin et J-Y. Hostachy
• LPSC-Grenoble (France)
• M. Krim
• University of Casablanca (Morocco)

2
A first step inside the CERN data

• Focus on e- inside ECal at 0
• Version 04.02 of calice_reco

Run list Run list Run list Run list
310046 -20 GeV 300384 -45 GeV
310047 -15 GeV 300589 -6 GeV
310048 -15 GeV 300590 -10 GeV
310050 -12 GeV 300591 -15 GeV
310051 -10GeV 300670 -6 GeV
310052 -12 GeV 300671 -8 GeV
310053 -15 GeV 300672 -10 GeV
310054 -10 GeV 300674 -15 GeV
310055 -12 GeV 300675 -18 GeV
300377 -45 GeV 300676 -20 GeV
300378 -30 GeV
3
Electron selection (i.e. rejection of other
particles)
Sigma X,Y,Z root mean square in X,Y,Z of the
shower
Run 300762 e- 10 GeV
4
Electron selection (i.e. rejection of other
particles)

• Very clean separation (few e- lost)
• Still a tail at low energy

Run 300762 e- 10 GeV
5
Evolution of the shower shape as a function of
the energy

• ??(?x2 ?y2)1/2 has practically no variation
  with E
• ?z increases as log(E)
• ? Good criterion to identify electrons

Uncertain point
6
Comparison with the Cherenkov
Run 300762 e- 10 GeV
7
Mapping of the e- reconstructed energy

• Tails at low energy due to the guard rings
• New cuts in order to reject the events affected
  by the inefficiency of the guard rings

Run 300762 e- 10 GeV
8
Checking cuts on the e- barycentre distributions
Run 300762 e- 10 GeV
9
Energy spectra
In blue all events In brown electrons events
(cuts on the shower shape) In red electrons
events in the central surface of the wafer In
blue hatched style other electron events
affected by the guard ring surfaces
Run 300762 e- 10 GeV
10
Energy resolution Gaussian fit (-1.5 ?,3 ?)
10 GeV electrons
With a 18.7
Run 300762 e- 10 GeV
11
Energy resolution
12
Energy resolution (comparison with previous data)
data fit previous fit
13
Linearity
14
Questions

• Two effects have to be quadratically subtracted
  from our results
• The noise contribution, about 4.7 10-4 /
  E (GeV) ? (i.e. 5.5 ADC counts)
• The momentum beam spread ? No idea

15
Summary

• The electrons can be identified from the shower
  shape
• Because the beam is aligned to the Ecal prototype
  structure, the reconstructed energy peak is
  asymmetric (tail at low energy due to the
  guard-rings)To reject the guard ring effect
  leads to optimistic results.
• The sampling term, without subtracting the beam
  contribution is about 18

16
Spare slide
Barycenter distribution before cuts
17
Spare slide
Barycentre distribution after the cuts on the
shower shape