Beautyfull Physics with more than 2 Muons

1
Beauty(full) Physics with more than 2 Muons

• Andreas Morsch
• Dimuon Workshop
• Berg-en-dal, South Africa
• June 7, 2004

2
Introduction

• The ALICE Dimuon-Spectrometer identifies and
  measures muons in 2.5 lt h lt 4 and with pT gt 1
  GeV.
• Further particle identification relies on
• Measurement of the resonance peaks in the
  unlike-sign di-muon mass-spectrum.
• The low-mass high-pT signature of b-chain
  decays
• Different pT spectra of muons from semileptonic
  charm (soft) and beauty (hard) decays
• Differences due to mb gt mc and harder
  fragmentation of b.
• Muon multiplicity is another signature which can
  be exploited.

3
Fishbone ...
m
m-
m
b
c
t
n
n
4
... and if you have more appetite

• mm-?r(w, f,h) ?QQ?m (light mesons)
• mm?cc ?gg?cc?mm (double gluon splitting)
• mm?bb ?gg?bb?mm
• m?c ?W?b?cm?m
• m?cm?b ?Z0?b?cm?m
• m?QQ ?gZ0?mm (Drell Yan initial
  state bremsstrahlungsgluon splitting)
• m?QQ ?gqq?g(Z0) ?mm
• m?cm?bm ?tt?bm?cm?m
• m?cm?b ?W?t?bm?cm?m

These channels have small cross-sections /
acceptance !
5
Main Motivation

• Correlated background in the di-muon channel
  consists of contributions from charm and beauty
  semi-leptonic decays
• For low-pmT (gt 1 GeV) equal contribution from
  both sources.
• J/y signal has contributions from direct
  quarkonium production and B?J/y X (15 at low
  pT).
• Correlated 3(4)-muons, however, can only origin
  from b. The 3rd muon tag could allow to measure
  the b-fraction of the J/y signal and select a
  clean beauty sample.

6
Main multi-Muon contributions
m D X
m/- X
m- X
J/y X
m/- X
m- m-
7
B semi-leptonic decays
Maximum of 4 muons with product of branching
ratios (0.1)nm
8
Recall Dimuons in pp
pmT gt 1 GeV
S/B 13 S/v(SB) 1400 b-fraction fb 0.15
Correlated muons
Correlated muons from b
Uncorrelated muons
9
Tri Muon analysis

• b event tagged by presence of 3rd muon
• pmT gt 1 GeV
• Uncorrelated background is obtained by
  multiplying the dimuon spectrum with the
  single-muon/event probability.

10
Caveat

• Signal decreases
• By factor of 1/10 due to extra branching ratios
• By factor of 1/5 due to acceptance
• In A-A, for large A, background is dominated by
  uncorrelated muons and un-correlated 3(4)-muon
  events are produced copiously.

11
So does it work ?

• In pp, pA certainly
• In central Ar-Ar it is already very difficult
• Pb-Pb is excluded
• However, if the current J/y suppression scenario
  observed by NA50 can be extrapolated to LHC, the
  most interesting physics might be in peripheral
  to central ArAr collisions with pp and pA for
  comparison.

12
Tri- muons in pp
b-chain
S/B 3 S/v(SB) 80
13
Tri-Muon topology
fake J/y s
good J/y s
DF
m
J/y
mJ/y
DF
m
14
Tri-Muon topology
J/y from B
Dimuons from continuum in the J/y mass region
15
What about 4m ?

• Tag b by presence of two additional muons
• S/B 3
• S/v(SB) 30

16
To what pT range are we sensitive ?
90
17
J/y m measures B-hadrons almost down to pT 0
18
How does this evolve with A ?

• Two contributions to dimuon continuum
• Correlated NbbNcc
• Uncorrelated Nuc
• For tri-muon tag there is one additional
  parameter single muon/event probability pm (of
  course directly related to Nuc ).
• Uncorrelated pm (NbbNccNuc)
• A-scaling
• Signal and correlated muon Aa
• Uncorrelated muons A2a
• pm Aa
• a 1.2
• Hence
• S Aa
• B A2a(NbbNccAaNuc) A3a for large A
• S/B A-2a for large A
• S/v(SB) vNevents A-a/2
• Nevents by factor 5000 smaller in Pb-Pb as
  compared to pp due to smaller event rate and
  running time (1/10).

19
Tri-muons in AA, pT gt 1 GeV

• central (10) Ar-Ar
• S/B 0.05
• S/v(SB) 20
• central (10) Pb-Pb
• S/B 0.003
• S/v(SB) 1.5

20
Conclusions

• Possibility to measure the b-contribution to the
  dimuon continuum and J/y signal at low-pT by
  using a tri-muon tag.
• Uncorrelated background scaling like A3 limit
  this possibility to pp, pA and maybe ArAr
  collisions.
• Event topology analysis not jet fully exploited.
• Clean b-samples for beauty studies.
• pp
• Azimuthal correlations in pp sensitive to NLO/LO
  processes (as).
• AA
• Momentum and angular correlations are sensitive
  to parton energy loss.