First results from NA482 experiment at CERN SPS

1
First results from NA48/2 experiment at CERN SPS

• Evgueni Goudzovski
• (JINR, Dubna)
• on behalf of NA48/2 collaboration
• Cambridge, CERN, Chicago, Dubna, Edinburgh,
  Ferrara, Firenze, Mainz, Northwestern, Perugia,
  Pisa, Saclay, Siegen, Torino, Vienna

Quarks-2004 Pushkinskie Gory, May 24-30, 2004
2
Introduction

• NA48/2 experiment
• Data taking 2003-2004
• Charged kaon beams (K and K-)
• World largest multipurpose sample
• of charged kaon decays

• Main goals
• Measurement of CP-violating charge
• asymmetry of K? pp-p and
• K? p0p0p decays accuracy 210-4
• Measurement of ?-? scattering parameter
• a00 in Ke4 decays accuracy lt0.01
• Study of rare charged kaon decays
• tests of Chiral Perturbation Theory

3
NA48/2 beamline
Simultaneous K and K- beams (vertical
cross-section)
PK (602) GeV/c
114m decay volume
Beams coincide within lt1mm
Evgueni Goudzovski / Quarks-2004
4
KAon BEam Spectrometer (KABES)

• Measurement of PK
• Reconstruct of K3p with a lost pion
• Resolve Ke4 reconstruction ambiguity.
• 3 stations
• achromat UP beam
• achromat DOWN beam
• both beams.
• 1mm strips
• Max rate 2MHz/strip
• sX 50 µm drift
• sY 80 µm strips
• st 0.7 ns
• ?p/plt1.

Evgueni Goudzovski / Quarks-2004
5
NA48 detector

• Main detector components
• Magnetic spectrometer (4 DCHs)
• ?p/p 0.5 0.009p GeV/c
• Liquid Krypton EM calorimeter (LKr)
• ?E/E 3.2/vE 9/E 0.42
• Hadron calorimeter, photon vetos,
• muon veto counters

Beam pipe
Evgueni Goudzovski / Quarks-2004
6
NA48/2 trigger

• All 3-track events
• K?3p charged mode
• Events with p0 ??ee- Dalitz decay
• K?ee-e?, K?ee-µ?, K?µµ-e?, K?µµ- µ ?
• 1-track events with (PK-Pp)2gtmp02
• high-mass trigger
• K?p0p0p neutral mode
• K?pp0?, K?p??
• Suppress K?pp0
• Downscaled (D100) triggers
• Many 1-track channels K?e?, e??, p0e?, etc

Evgueni Goudzovski / Quarks-2004
7
NA48/2 data taking 2003-2004

• Data taking period 2003 50 days.
• Data presented here last month of data taking
  2003 stable running conditions.
• All 2003 data almost twice as much
• The part of data not presented may have different
  systematic effects for asymmetry measurement.
• Run 2004 started 60 days scheduled.

Evgueni Goudzovski / Quarks-2004
8
Charge asymmetry measurement (charged mode)
Evgueni Goudzovski / Quarks-2004
9
K3? matrix element

• Matrix element

• Kinematic variables

Lorentz-invariants U (s3-s0)/m?2 V
(s2-s1)/m?2 si (PK-P?i)2, i1,2,3 s0
(s1s2s3)/3.
M(u,v)2 1 gu hu2 kv2

• Direct CP-violation

Laboratory frame U 2mK(mK/3-Eodd)/m?2 V
2mK(E1-E2)/m?2.
Ag (g-g-)/(gg-)?0
Evgueni Goudzovski / Quarks-2004
10
Ag theory vs. experiment

• Theoretical predictions
• I.Scimemi, Kaon Mini Workshop, CERN
  5/05/2004
• Ag 10-5 perfectly compatible with
  SM
• 310-5ltAglt510-5 compatible with SM, but in bad
• agreement with
  e/e
• Aggt510-5 New Physics
• E.Shabalin, Kaon Mini Workshop, CERN
  5/05/2004
• Ag310-5 within SM.
• Experimental results
• BNL (1970) Ag(-7.05.3)10-3
• FNAL, HyperCP (2000) Ag(2.21.53.7)10-3
• NA48/2 goal
• dAglt210-4 still dominated by statistical error.

Evgueni Goudzovski / Quarks-2004
11
K3p Dalitz plot
720 mln K events
V
even pion in beam pipe
odd pion in beam pipe
U
Evgueni Goudzovski / Quarks-2004
12
Principle of measurement
M(u,v)2 1 gu hu2 kv2

• Project into U axis
• Neglect quadratic slopes h and k
• Remember Agltlt1

• K?pp-p g -0.21540.0035
• K?p0p0p g 0.6520.031
• h, k ltlt g

If acceptance is equal for K and K-, R(u)
N(1gu)/(1g-u) N(12gAgu)

Ag can be extracted from a linear fit to the
ratio R(U)
Evgueni Goudzovski / Quarks-2004
13
Acceptance cancellation

• Symmetrezation of acceptance analyzing magnet
  polarity (B) is switched on daily basis
• Symmetrization of beam properties achromat
  polarity (A) is switched on weekly basis.
• Data taking from August 6 to September 7, 2003

B
B-
B
B
B-
Supersample 1
B-
Achromat -
Week 1
12 day-samples
B
B
B
B-
B-
B-
Achromat
Week 2
B
B
B
B-
B-
Supersample 2
B-
Achromat -
Week 3
12 day-samles
B
B
B
B-
B-
B-
Achromat
Week 4
B
B-
Supersample 3
Achromat -
Week 5
4 day-samples
B
B-
Achromat
Evgueni Goudzovski / Quarks-2004
14
Acceptance cancellation (2)

• Physical asymmetries
• AS ? slope of ratio U(BK)/U(B-K-)
• AJ ? slope of ratio U(B-K)/U(BK-)
• Apparatus-induced asymmetries
• A ? slope of ratio U(BK)/U(B-K)
• A- ? slope of ratio U(BK-)/U(B-K-)

Top view of the setup
Saleve
Z axis (beam direction)
Jura
X axis

• ASJ (ASAJ)/2
• physics asymmetry
• A (AA-)/2 (AS-AJ)/2
• asymmetry induced by
  experimental setup
• many of the effects
  observed in A cancel in ASJ

Evgueni Goudzovski / Quarks-2004
15
The role of time-instabilities

• Measurement of AS, AJ, A, and A- implies
  combining data from different day-samples
• Therefore measurement is sensitive to
• time-instabilities of beams and setup.
• The following instabilities were found
  significant and have been corrected for
• Time-dependence of spectrometer magnetic field
• Time-dependence of spectrometer alignment
• Time-dependence of beam geometry.

Evgueni Goudzovski / Quarks-2004
16
Spectrometer calibration

• P P0(1ß)(1qb?P0)
• P0 measured momentum
• P corrected momentum
• q track charge
• b magnetic field sign.
• Time-dependent corrections
• ß for magnetic field integral
• ? for spectrometer misalignment

Time-dependence of alignment
MSaleveMJura, keV/c2
Max. DCH4 shift of 70 ?m
Day-sample
Evgueni Goudzovski / Quarks-2004
17
Sensitivity to beam geometry

• Why is the measurement sensitive to beam
  geometry?
• Acceptance is defined by beam pipe!

• Time-dependent cuts following beams movement are
  introduced to symmetrize beam geometry.

V
Beam pipe
Virtual tube SRpi-ltRKgtgtR0 at DCH1,
DCH4 ltRKgt f (kaon sign, time, momentum)
U
Evgueni Goudzovski / Quarks-2004
18
Alternative to virtual tubenot yet included
into the analysis

• Use kaon momentum measured by the beam
  spectrometer (KABES)
• Need only two detected pions to reconstruct event
  kinematics
• Low sensitivity to beam pipe acceptance
• Add 2-track statitics
• (60 events)
• Events in high-U region
• higher sensitivity to Ag

V
Beam pipe
U
Evgueni Goudzovski / Quarks-2004
19
Beam movement in time
DCH4
DCH1
Y, cm
K
K
Y, cm
B
B-
X, cm
X, cm
K-
K-
Y, cm
Y, cm
B
B-
X, cm
X, cm
Evgueni Goudzovski / Quarks-2004
20
Momentum-dependence of beam position
DCH4
AKJ, A-KS
Y, cm
55 GeV/c
65 GeV/c
55 GeV/c
Split by analyzing magnet field
X, cm
AKJ, A-KS
Y, cm
55 GeV/c
65 GeV/c
65 GeV/c
55 GeV/c
X, cm
Evgueni Goudzovski / Quarks-2004
21
Statistics in the charged mode
Selected events per supersample (in millions)
K
Events
M3p, GeV/c2
Events
K-
M3p, GeV/c2
Evgueni Goudzovski / Quarks-2004
22
Results on asymmetry

• In the following slides asymmetries as functions
  of kaon momentum and time are presented
• Physics asymmetries ASJ are presented with
  offsets no result yet to be presented.
• Apparatus-induced asymmetries no offsets,
  indications of systematics.
• Note that ASJ 2gAg -0.43Ag

Measured value
Notation in theory papers
Evgueni Goudzovski / Quarks-2004
23
Asymmetry vs. kaon momentum
?ASJ1.1710-4 ?Ag2.710-4 ?2 6/11
Units of 10-4
OFFSET!
OFFSET!
?A(1.021.17)10-4 ?2 9/11
Evgueni Goudzovski / Quarks-2004
24
Fixed achromat orientations
Achromat UP
Achromat DOWN
Units of 10-4
OFFSET!
OFFSET!
Systematics cancellation
Evgueni Goudzovski / Quarks-2004
25
Time-stability of asymmetry
Units of 10-4
OFFSET!
?213.5/13
?25.6/13
OFFSET!
Evgueni Goudzovski / Quarks-2004
26
Statistics in the neutral mode
Selected events per supersample (in millions)

• Analysis not as advanced as in the charged
  mode
• Kinematics is reconstructed
• using the information from
• the Liquid Krypton
• Calorimeter
• Sources of systematics are
• different from the
• charged mode
• Expected statistical error
• ?Ag 510-4.

Evgueni Goudzovski / Quarks-2004
27
Rare decays of charged kaons
Evgueni Goudzovski / Quarks-2004
28
Measurement of K?pp-e? (Ke4)

• Physics interest
• Low energy pp scattering predicted from Chiral
  Perturbation Theory first principles.
• Predicted in ChPT a000.2200.005
• Colangelo, Gasser, Leutwyler, hep-ph/0103088
• pp scattering length a00 can be determined from
  form-factors of Ke4 decays.
• Previous measurements
• Geneva-Saclay (1977) 30,000 events
• Brookhaven E865 (2001) 400,000 events
• E865 measurement a00 0.2160.013
• 2 experiments, errors gt theoretical uncertainties.

Evgueni Goudzovski / Quarks-2004
29
Measurement of K?pp-e? (Ke4)

• More than 300,000 events in one month of data
  taking.
• Background mainly from K?3p and K?pp0Dalitz
  with p? e mis-identification.
• Use neural network LKr information for e/p
  separation.
• Total background 0.6
• determined from wrong-sign ppe- events.
• Expect gt500k events from 2003 data.

Background
Evgueni Goudzovski / Quarks-2004
30
Measurement of K?pp-e? (Ke4)

• Cabibbo-Maksymowicz variables background in red

Evgueni Goudzovski / Quarks-2004
31
Measurement of K?pee- and K?pµµ-

• Physics interest
• Flavor-changing neutral currents, suppressed by
• GIM mechanism
• Amplitude dominated by one-photon diagram can
• be extracted from form-factor measurements
• Potentially interesting channels for
  CP-violation
• Previous measurements
• K?pee- BNL E865 found 10,500 events
• BR, form-factor measurement
• K?pµµ- Several experiments, in total 800
  events
• BR measurements

Evgueni Goudzovski / Quarks-2004
32
Measurement of K?pee-

• More than 2,600 candidates
• for meegt140 MeV/c2
• in one month of data taking.
• Small background (1-2).
• Expect data sample comparable
• to World best sample
• in 2003-2004 data.

Form-factor
Evgueni Goudzovski / Quarks-2004
33
Measurement of K?pµµ-
Events

• More than 1,000 events
• in one month of data taking
• (current World sample 800 evt!)

Evgueni Goudzovski / Quarks-2004
34
Measurement of K?p??

• Physics interest
• O(p2) ChPT amplitude vanishes
• O(p4) amplutude computed (up to an unknown
  parameter)
• Fit of M?? distribution check O(p6) ChPT
  expansion.
• Experimental problem suppression of K?pp0 at
  trigger level. Two trigger conditions for
  K?p??
• Downscaled neutral trigger downscaled (D80)
• High-mass trigger no downscaling, but no low
  M??
• Low-mass trigger below p0 mass implemented in
  2004.

Evgueni Goudzovski / Quarks-2004
35
Measurement of K?p??
Compare to 31 events of E787!
Evgueni Goudzovski / Quarks-2004
36
Conclusions

• K?pp-p
• gt1.1109 decays in one month of run 2003
• Major sources of systematics identified
• Statistical error of Ag is 2.710-4
• Uncertainty dominated by statistical error!
• K?p0p0p
• 4107 decays in one month of run 2003
• Expected statistical error of Ag is 510-4.
• Large samples of charged kaon rare decays allow
  high precision measurements.
• Run 2004 started 60 days of data taking.

Evgueni Goudzovski / Quarks-2004