Heavy quark production with high energy neutrino beams

1
Heavy quark production with high energy neutrino
beams

• Pasquale Migliozzi
• INFN - Napoli
• Heavy Quarks and Leptons
• 27th-31th May 2002, Vietri

2
Outline

• What do we measure and why it is interesting
  (a short theoretical introduction)
• How do we measure and the experimental challenge
  (past and present experiments)
• Experimental results and near future achievements
  on
• from charm quark to charmed hadrons
• dimuon analyses
• low energy charm production
• associated charm production in CC and NC n
  interactions
• charm production cross-section
• low multiplicity charm production
• Perspectives at a Neutrino Factory
• Conclusion

3
A short theoretical introduction
4
n DIS charm production
Quark density functions, strange sea (?)
Production from d(anti-d) quarks Cabibbo
suppressed ? large s contribution ?50 in n and
?90 in anti-n
5
pQCD NLO CC charm production
Recently a code for NLO analyses became
available S.Kretzer,D. Mason,F.Olness
hep-ph/0112191 Important tool for future NLO
analyses of neutrino charm-production data NB So
far only CCFR performed a NLO analysis
6
Why charm physics with n is interesting?

• Charm produced in nCC DIS interactions
  (single charm production)
• measure strange content of the nucleon
• potential strange/anti-strange asymmetry ?
  non-perturbative QCD effects
• crucial role in relating charged-lepton and
  neutrino F2 structure functions
• knowledge of the strange sea is important to
  search for stop at hadron colliders
• largest background gs?Wc
• R.Demina et al., Phys. ReV. D 62 (2000) 035011
• S.J.Brodsky and B.Ma, Phys. Lett. B 381 (1996)
  317

7
Why charm physics with n is interesting?

• constrain/study charm production models
• in NLO pQCD is a challenging theoretical problem
• 2 scales, LQCD and charm mass
• (J.Conrad et al. Rev.Mod.Phys. 70 (1998)
  1341-1392)
• measure charm mass and Vcd
• Charm produced in nNCCC DIS interactions
  (associated charm production)
• measure charm mass and check its universality
• constrain associate charm production models
  boson-gluon fusion, intrinsic charm parton
  distribution, npQCD effects producing unusually
  large c(x,Q2) at high x,

8
Experimental issues
9
Experimental issues

• Two important ingredients to study charm
  production are
• The neutrino beam horn focused, WANF, (CERN)
  sign-selected quadrupole-triplet, SSQT,
  (Fermilab)
• Detection techniques
• massive high density detectors (CDHS, CCFR,
  CHARMII, NuTeV, NOMAD FCAL, CHORUS Calo)
• bubble chamber filled with heavy liquid (BEBC,
  Fermilab 15-ft)
• nuclear emulsions (E531, CHORUS)

10
Massive high-density detectors

• These experiments study charm production by
  looking at dimuon events

• Pro large statistics
• Contra background from p,K decays not sensitive
  to low-neutrino energies (Enlt15GeV) not possible
  to study separately the different charmed types
  Bm is needed

11
Dimuon available statistics
An analysis of all available data on dimuon
cross-section normalised to CC interactions has
been performed in G. De Lellis, A. Marotta, P.M.
J.Phys. G 28 (2002) 713-724
NuTeV Calorimetry ?4500 ?1100
12
Emulsion experiments

• These experiments study charm production by
  looking directly at the decay topology of the
  charmed hadron with micrometric resolution
• Contra till few years ago the charm
  statistics was limited by the scanning
  power (but this is not the
  case anymore) the
  anti-n statistics is very poor
• Pro low background sensitivity to
  low En?mc thr. effect id.
  hadron species reconstruction of the charmed
  hadron kinematics (direction and
  momentum) ?fragmentation studies are
  possible

13
Review of available experimetal results
14
From charm quark to charmed hadrons

• fh cannot be calculated with pQCD
• ? experimental determination needed
• p2T cannot be calculated with pQCD
• ? experimental determination needed
• for D there are several phenomenological
  approaches, which depend on one parameter
• ? to be determined experimentally

15
Charmed fractions

• fh can only be measured in emulsions!
• Present results based on the 122 E531 events and
  a reanalysis discussed in T. Bolton hep-ex/9708014

Analysis in progress of the CHORUS data ?1000
events Results should be available in a couple of
months
16
D0 production rate A.Kayis-Topaksu et al., Phys.
Lett. B 527 (2002) 173-181
17
pT2 distribution of charmed particles
18
p2T distribution of charmed particles
One measurement available with 360 GeV pp
interactions b 1.1?0.3 (GeV/c)-2 M.Aguilar-B
enitez et al., Phys. Lett B 123 (1983) 103

• Analysis in progress of the CHORUS data ?1000
  events
• Results should be available in a couple of
  months
• New analysis of the NOMAD data in progress

19
Fragmentation functions
The z distribution can be parametrized as follows
20
Fit to z distribution

• Direct measurements
• E531, NOMAD, CHORUS-Emul in progress
• z distribution is extracted for
• charmed hadrons and fitted

• Indirect measurements
• CDHS, CCFR, CHARMII, NuTeV, CHORUS-Calo in
  progress
• ep or ec (depends on the choice) is one of the
  free parameters of the fit to the dimuon data,
  see later

21
Determination of ep and ec
All the above numbers have been obtained with a
LO analysis
At ee- exps s1/2?10 GeV eP?0.16(D)?0.27(Lc) O.Bi
ebel, P.Nason, B.R.Webber hep-ph/0109282
22
Relevant parameters of the fit to dimuon data

• Input parameters
• Charmed fractions and decay model constrained by
  other experiments
• Vcs (In the following we use 0.9960.024
  Riv.NuovoCim. 23(2000)1)
• Bm?BR(C ? m) (In the following 9.310.95 for En
  gt30 GeV)
• Output parameters
• Charm mass mc
• Element of the CKM matrix Vcd
• Fragmentation parameter e
• Two parameters for each mode (n and anti-n) that
  describe the magnitude and the shape of the s and
  anti-s PDFs
• ?2S/(UD) is the proportion of s-quarks to non
  strange quarks in the nucleon sea
• x(1-x)a is the shape of the s-quark PDF

23
Bm as measured in CHORUS
Basic idea High purity selection (special MCS
treatement in track and vertex fit)
Number of selected events 1055
90.6 selection purity
956 ? 35
Dimuon sample 88 ? 10 (stat) ? 8 (syst)
B? 9.3 ? 0.9 (stat) ? 0.9 (syst)
only direct measurement available
these correspond to less than 50 of the CHORUS
statistics
24
mc determination
25
Vcd determination
As expected LO and NLO give consistent results!
26
k determination
LO and NLO give consistent results at 1.2s
27
Dimuon analysis

• Hard to extract model independent charm
  production cross-section
• Missing n from charmed-hadron decay
• Fragmentation and cross-section model assumption
  dependence
• So, usual experimental technique
• Parameterize s-PDF, c-production model in MC
• Fit MC to dimuon distributions, extract
  c-production model and s-PDF parameters
• Results can only be used indirectly in global
  PDF fits and depend on model and functional form
  choice
• Usual criticism
• Result depends on choice of functional form
• Result depends on model assumptions

28
NuTeV approach

• Extract dimuon production cross-section
  (M. Goncharov et al., Phys. Rev. D 64 (2001)
  112006)
• Use LO production model, obtain good description
  of observed rates
• Extract LO model parameters (used also as
  consistency check)
• Use MC to correct for experimental effects and
  extract dimuon cross-section
• cross-section can be used to fit any model
• NLO fits (in progress)
• NLO fits to data and cross-section table
  (consistency check)
• NLO fits to cross-section table use different
  models, PDFs, etc

29
Low energy charm production

• Quasi-elastic charm production
• In the literature only 3 events observed in
  nuclear emulsions (E531)
• CHORUS is performing a dedicated search for QE
  charm production so far 54 events consistent
  with a QE charm topology have been observed
• With full statistics few hundred events are
  expected. Possible measurements
• differential cross-section
• absolute Lc BR (very interesting see next slide)

30
Quasi-elastic charm production
a) ?? n ? ?- ?c b) ?? n ? ?- ?c
(?c) c) ?? p ? ?- ?c(?c)

• QE charm production has a peculiar topology and
• always a Lc in the final state
• pure Lc sample can be built with a small (lt10)
• normalization error ? absolute BR determination

P. Migliozzi et al. Phys. Lett. B 462 (1999)
217-224
31
Lc BR problem

• So far only model dependent extractions are
  available
• Two different methods, relying on different
  theoretical assumption on B physics, give
  different results! (see PDG2000
  pag. 801)
• Model independent determination
• New insight on underlying b-physics

32
Associated charm production

• Charged-current interactions (gluon
  bremsstrahlung)
• In the past this search was based on the
  observation of trimuon events m-(mm-) and
  same-sign dimuons
• Large background from p and K decays
• Observed rate 60 times larger than expected from
  theoretical calculations! (K.Hagiwara Nucl.Phys.B
  173 (1980) 487)
• Currently a search for this process is in
  progress in the CHORUS emulsions
• 1 event has been already observed and confirmed
  by a kinematical analysis (paper submitted to
  Phys. Lett. B)
• A new analysis with a larger statistics is in
  progress. In the future the discrepancy between
  data and theoretical predictions should be
  clarified.

33
Associated charm production

• Neutral-current interactions (g-brem. Z-g
  fusion)
• In the past only one event observed in the E531
  emulsion
• Producton rate 1.33.1-1.1 x10-3 normalised to CC
• Indirect search performed by NuTeV
  A.Alton et al., Phys. Rev. D64 (2001) 012002
• Production rate (2.0?1.6)x10-3 normalised to CC
  at 154 GeV
• mc(1.400.83-0.36 ?0.26) GeV, in agreement with
  other measurements
• Currently a search for this process is in
  progress in the CHORUS emulsions
• A cross-section measurement should become
  available by the end of this year.

34
Perspectives at NuFact

• Very high intensity neutrino beams
• Well defined beam composition
• GOAL 1010 nmCC interactions in the target
• 108 events with charm in the final state
• 102-103 events with bottom in the final state
• A lot of interesting physics could be exploited
• References
• I.Bigi et al. BNL-67404
• M.L. Mangano, P.M. et al. hep-ph/0105155

35
Physics reach of a NuFact by studying heavy-quark
production

• Direct determination of Vcd and Vub with lt1
  accuracy
• Systematically different determination of Vcs and
  Vcb at few
• Direct and precise determination of absolute BR
  of charmed hadrons
• D0-D0 oscillations through the wrong-sign
  semi-leptonic decay channel
• Precise study of the strange-sea PDF k and a at
  NLO
• Physics beyond the Standard Model
• Precise extraction of fh and mc
• B physics nlN-gtnlbbX or nlN-gtlbcX

36
Conclusion

• Charm physics in n interactions very interesting
• Near future physics reach
• NLO analyses (NuTeV) are in progress or will
  start very soon (CHORUS-Calo, NOMAD-FCAL)
• clarification of the strange-sea current results
  (?)
• better mc determination
• Precise study of charm fragmentation (final
  statistics in CHORUS-Emul about 4000 events)
  ? pT2, e as a function
  of Ch understanding (?)
• Absolute determination of Lc BR(CHORUS-Emulsion)
• Physics reach at future NuFact
• Charm production studies with very high accuracy
• Possibility to study b-physics in neutrino
  interactions

37
(No Transcript)
38
The strange sea distribution

• No new results
• Both at LO and NLO k0.5
• At LO a is not zero a2.5?0.7, while it is at NLO
• Daa-a-0.46?0.42?0.36?0.65?0.17
• The momentum distributions of s and anti-s are
  consistent and the difference in the two
  distributions is limited to 1.9lt Dalt1.0 at 90
  C.L.
• In the near future NuTeV NLO analysis will be
  available CHORUS-Calo and NOMAD-FCAL LO analyses
  will be available, too

39
pQCD LO CC charm production
Production from d(anti-d) quarks Cabibbo
suppressed ? large s contribution ?50 in n and
?90 in anti-n

• ?x(1m2c/Q2)(1-x2M2-Q2), for kinematical effects
  of heavy charm
• z fraction of the charm momentum carried by the
  charmed hadron
• p2T transverse momentum of Ch wrt to the charm
  direction
• fh mean multiplicity of the charmed hadron
  h(D0,D,Ds,Lc) in n
• charm production
• D probability for a charm quark to fragment into
  a charmed hadron
• h with a given z and p2T

40
The neutrino beam

• Both WANF and SSQT are high energy beams En
  range 3400 GeV
• Main advantage of SSQT possibility to run n and
  anti-n beams with very low wrong-sign
  contamination (lt10-3)

41
The NuTeV detectorD.A. Harris et al. Nucl.
Instr. Meth. A 447 (2000) 377-415

• Target Calorimeter (690 tons)
• Steel/Scintillator, DE/E?0.86/E½ energy sampling
  every 10 cm Fe tracking chambers for m and vtx
  determination every 20 cm Fe
• Toroid spectrometer
• 3 Fe magnets, each contains 4 chamber stations
  Dp/p?11 MCS dominated
• always focusing primary muon
• Detector calibrated with test beam calorimeter
  to 0.43 and spectrometer to 1

42
Dimuon as seen by NuTeV
43
CHORUS detectorE.Eskut et al., Nucl. Instr.
Meth. A 401 (1997) 7-44
? -
Calorimeter
T5
h-

Muon spectrometer
Heart of the detector Nuclear emulsion target
770 kg emulsion target and scintillating fibre
tracker
Air core spectrometer
and emulsion tracker
Veto plane
44
CHORUS emulsion
800 kg active target MIP 30 ? 40 grains / 100
?m transverse resolution 0.5 ?m depth of focus
1 to 3 ?m
S. Aoki et., Nucl. Instr. Meth. A 447 (2000)
361-376
45
Automatic Scanning
Tracks reconstructed by a hardware video
processor frame to frame emulsion grains
coincidence
150x150 ?m view
microscope stroke
track
X50 magnification 3?m focal depth
tomographic image
350 ? m emulsion sheet
90 ?m plastic backing
emulsion plate
350 ? m emulsion sheet
T.Nakano, Ph.D. Thesis, Nagoya Univ., 1997
46
Netscan analysis in CHORUS

• All track segments (? lt 0.4 rad) in
• Fiducial volume 1.5 x 1.5 mm2 x 8 plates
• Offline analysis of emulsion data

47
Inclusive charm production cross-section induced
by n
CHORUS is currently analysing about 1000
evts Final statistics (gt1year from now) about
3000-4000 evts!
48
Average inclusive charm production
cross-sectionG. De Lellis, A. Marotta, P.M.
J.Phys. G 28 (2002) 713-724
By using the fh measured by E531, dimuon and
emulsion data have been combined to extract the
world average sc/sCC ratio
Parametrization useful for background calculation
in oscillation experiments
49
Dimuon cross-section
50
Dimuon cross-section
51
NuTeV Dimuon cross-section extraction

• Use hit-level MC with LO c-production
  cross-section model
• Fit MC to data obtain good description of
  dimuon observables
• Extract LO parameters (consistency check)
• Use MC to correct for experimental effects and
  extract forward
• dimuon cross-section

Flux/normalization - Beam MC tuned to
inclusive data - Normalize dimuon MC and data to
inclusive rate report cross-section ratio
(minimize sensitivity to flux)
Normalization sample
52
MC fit to dimuon sample
Fragmentation and decay models constrained from
external data. p /K decay Test Beam (shower)
Lepto-Lund (primary) s-PDF parameterized by
relative size to non-strange sea (k) and shape
(1-x)a
Low Evis sensitive to mc EvisEm1Em2Ehad
ZvisEm2/(Em2Ehad) Sensitive to fragmentation
Low-xvis n sensitive to size xvis anti-n
sensitive to shape
Crosses p/K background Stars s-sea contribution
53
Dimuon cross-section measurement

• Model independent result cross-section
• measurement for Em2gt5GeV (acceptance gt55)
• MC describes data well with all LO
• s-PDF sets used
• use MC to unfold flux, correct
• for acceptance and smearing
• (keep it lt40 in all bins)
• cross-section extracted with
• all 3 PDF no model dependence
• as a check, fit on cross-section
• tables, and compare model
• parameters to direct fits to
• data. No bias found!

54
Low multiplicity charm production

• Diffractive Ds() production
• Cross-section (normalized to CC) measurement
• NuTeV (3.2?0.6)x10-3/CC Phys.
  Rev. D61 (2000) 092001
• BEBC (2.8?1.1)x10-3/CC
• Z. Phys. C 58 (1993) 55
• Weighted Average (3.1?0.5)x10-3/CC
• The observation of Ds() production by CHORUS is
  consistent with this result (see next slide)

55
Diffractive Ds production P. Annis et al.
(CHORUS) Phys. Lett. B 435 (1998) 458-464
?? N ? Ds ?- N Ds ? ? ??
? ????
56
Associated charm production in CC interactions
57
Why is it important to measure accurately fDs?

• A better fDs determination (DfDslt10)
• discriminate among different theoretical
  calculations
• more confident about the predictions of fB and
  fBs crucial quantities for a quantitative
  understanding of B0(s) oscillations and Vtd (Vts)
  extraction from them
• A method based on antineutrino induced
  diffractive D()s production allows DfDslt5

58
The CKM matrix

• Vus
• 1
• Ke3 decay

• Vub
• 25
• b ? u l ?

• Vud
• 0.1
• nuclear beta decay

d
d

• Vcb
• 5
• Be3 decay

• Vcs
• 15
• De3 decay

• Vcd
• 7
• ? charm production

s
s

b
b

• Vtd

• Vts

• Vtb
• 30
• t ? b l ?

Review of particle physics, 98 edition
59
A method to extract fDs (G. De
Lellis,P.M.,P.Zucchelli Phys.Lett.B507(2001) 7-xx)
By using diffractive production induced by anti-n
an almost pure Ds sample can be built
(contamination lt5)
60
Leptonic Ds decays
Several theoretical predictions for fDs are
available and lie in the range 190-360 MeV!
Uncertainties larger than 30! Main source
normalisation used for Ds BR determination
61
Description of the method(G. De
Lellis,P.M.,P.Zucchelli Phys.Lett.B507(2001) 7-xx)
By selecting events with this topology
AND
By applying a simple kinematical analysis
An almost pure Ds sample can be
built (contamination lt5)
N.B. D- and Lc- events do not affect the Ds?t
channel