MiniBooNE

1
MiniBooNE
Vth Rencontres du Vietnam 2004 David
Schmitz Columbia University
Outline

• MiniBooNE Motivation
• LSND Signal
• Interpreting the LSND Signal
• MiniBooNE Overview
• Experimental Setup
• Neutrino Events in the Detector
• The Oscillation Search
• Studying MiniBooNE Hadron Production at HARP
• The HARP Data Set
• HARP Analysis

2
MiniBooNE Motivation The LSND Result

• The Liquid Scintillator Neutrino Detector was
  the first accelerator based neutrino oscillation
  experiment to see a signal.
• LSND saw a 3.8s excess (above expected
  background) of ne in a nm beam.

combined analysis allowed region

• The KARMEN experiment was a similar experiment
  that saw no signal neutrinos. KARMEN had less
  statistics and a slightly different experimental
  L/E.
• A combined analysis of LSND and KARMEN leaves a
  substantial allowed region.

3
MiniBooNE Motivation Interpreting the LSND
Signal

• What to make of 3 independent Dm2 values?
• solar exp. (Super-K, K, SNO, KamLAND, )
  Dm2 10-5 eV2
• atmospheric exp. (Super-K, K, )
  Dm2 10-3 eV2
• accelerator exp. (LSND)
  Dm2 1 eV2

4
MiniBooNE Motivation Interpreting the LSND
Signal

• What to make of 3 independent Dm2 values?
• solar exp. (Super-K, K, SNO, KamLAND, )
  Dm2 10-5 eV2
• atmospheric exp. (Super-K, K, )
  Dm2 10-3 eV2
• accelerator exp. (LSND)
  Dm2 1 eV2

• One of the experimental results is incorrect.
  Must verify each Dm2.
• atmospheric and solar results are well
  confirmed.
• accelerator and reactor based exp. in the atmo.
  and solar ranges (K2K, MINOS, KamLAND)
• LSND requires confirmation.

5
MiniBooNE Motivation Interpreting the LSND
Signal

• What to make of 3 independent Dm2 values?
• solar exp. (Super-K, K, SNO, KamLAND, )
  Dm2 10-5 eV2
• atmospheric exp. (Super-K, K, )
  Dm2 10-3 eV2
• accelerator exp. (LSND)
  Dm2 1 eV2

• One of the experimental results is incorrect.
  Must verify each Dm2.
• atmospheric and solar results are well
  confirmed.
• accelerator and reactor based exp. in the atmo.
  and solar ranges (K2K, MINOS, KamLAND)
• LSND requires confirmation.

• Addition of 1 or more Sterile neutrinos to the
  3 neutrino standard model.
• LSND could be explained by oscillations to
  sterile neutrinos.

6
MiniBooNE Motivation Interpreting the LSND
Signal

• What to make of 3 independent Dm2 values?
• solar exp. (Super-K, K, SNO, KamLAND, )
  Dm2 10-5 eV2
• atmospheric exp. (Super-K, K, )
  Dm2 10-3 eV2
• accelerator exp. (LSND)
  Dm2 1 eV2

• One of the experimental results is incorrect.
  Must verify each Dm2.
• atmospheric and solar results are well
  confirmed.
• accelerator and reactor based exp. in the atmo.
  and solar ranges (K2K, MINOS, KamLAND)
• LSND requires confirmation.

?

• Addition of 1 or more Sterile neutrinos to the
  3 neutrino standard model.
• LSND could be explained by oscillations to
  sterile neutrinos.

• Other possibilities
• CPT violation
• CP violation sterile neutrinos
• others

7
MiniBooNE Motivation Interpreting the LSND
Signal

• What to make of 3 independent Dm2 values?
• solar exp. (Super-K, K, SNO, KamLAND, )
  Dm2 10-5 eV2
• atmospheric exp. (Super-K, K, )
  Dm2 10-3 eV2
• accelerator exp. (LSND)
  Dm2 1 eV2

• One of the experimental results is incorrect.
  Must verify each Dm2.
• atmospheric and solar results are well
  confirmed.
• accelerator and reactor based exp. in the atmo.
  and solar ranges (K2K, MINOS, KamLAND)
• LSND requires confirmation.

• Addition of 1 or more Sterile neutrinos to the
  3 neutrino standard model.
• LSND could be explained by oscillations to
  sterile neutrinos.

The LSND signal must be confirmed or ruled out to
know how to proceed in the neutrino sector.

• Other possibilities
• CPT violation
• CP violation sterile neutrinos
• others

8
MiniBooNE Overview Experimental Setup
Decay region
25 m
50 m
450 m

• MiniBooNE receives 8.9 GeV/c protons from the
  Fermilab Booster.
• Protons are focused onto a 1.7 interaction
  length beryllium target producing various
  secondaries (ps, ps, Ks).
• Secondaries are focused via a magnetic focusing
  horn surrounding the target. The horn receives
  170 kA pulses at up to 10 Hz.

9
MiniBooNE Overview Experimental Setup
Decay region
25 m
50 m
450 m

• Secondary mesons (ps, Ks) decay in the 50m
  decay region to produce the MiniBooNE neutrino
  beam.
• A removable 25m absorber can be inserted. A
  great advantage for studying backgrounds.
• The horn is capable of running with the polarity
  reversedanti-neutrino mode.

( )
( )
10
MiniBooNE Overview Experimental Setup
Decay region
25 m
50 m
450 m

• Neutrinos are detected 500 m away in a 12 m
  diameter Cerenkov detector.
• 950,000 liters of mineral oil
• 1280 photomultiplier tubes
• 240 optically isolated veto tubes

11
MiniBooNE Overview Neutrinos in the Detector

• We look for remnants of n CC events in the
  detector producing a ring of prompt Cerenkov
  light and a small amount of delayed scintillation
  light.

• NC p0 events are characterized by the double
  rings produced by p0 g g. These events can
  look like electron events when the photons
  overlap or the decay is asymmetric.

12
MiniBooNE Overview More About CCQE Events

• Reconstruct the lepton angle with respect to the
  beam direction.
• Measure visible energy from Cerenkov light and
  small amount of scintillation light.
• 10 En resolution at 1GeV with no background

13
MiniBooNE Overview More About CCQE Events

• Reconstruct the lepton angle with respect to the
  beam direction.
• Measure visible energy from Cerenkov light and
  small amount of scintillation light.
• 10 En resolution at 1GeV with no background

nm CCQE Event Reconstruction
PRELIMINARY
PRELIMINARY
PRELIMINARY
14
MiniBooNE Overview nm ne Oscillation
Sensitivity

• Recall that the MiniBooNE ne appearance analysis
  is a blind analysis.
• ne CCQE events suffer from larger backgrounds
  than nm events.
• Use measurements both internal and external to
  constrain background rates.

15
MiniBooNE Overview nm ne Oscillation
Sensitivity

• Recall that the MiniBooNE ne appearance analysis
  is a blind analysis.
• ne CCQE events suffer from larger backgrounds
  than nm events.
• Use measurements both internal and external to
  constrain background rates.

• With 1x1021 protons on target
• Average 5 uncertainty on background rates.

16
MiniBooNE Overview nm ne Oscillation
Sensitivity

• Recall that the MiniBooNE ne appearance analysis
  is a blind analysis.
• ne CCQE events suffer from larger backgrounds
  than nm events.
• Use measurements both internal and external to
  constrain background rates.

• With 1x1021 protons on target
• Average 5 uncertainty on background rates.

17
MiniBooNE Overview nm ne Oscillation Signal
18
MiniBooNE Beam Hadron Production at HARP
MiniBooNE has cooperated with the HARP experiment
(PS-214) at CERN to measure hadron production
from the MiniBooNE beryllium target.

• The first goal is to measure p production cross
  sections for Be at pproton 8.9 GeV/c.
• Additional measurements include
• p- production (important for n running)
• K production (important for intrinsic ne
  backgrounds)

19
MiniBooNE Beam Beryllium Target

• The MB target is 71 cm long and 1 cm in
  diameter
• Cooling fins (also Be)
• Comprised of seven 10 cm slugs

20
HARP Cross Section Measurement
pion purity
migration matrix
acceptance
pion yield
tracking efficiency
pion efficiency
21
HARP Cross Section Measurement
pion purity
migration matrix
acceptance
pion yield
tracking efficiency
pion efficiency

• Acceptance is determined using the MC (compare
  to MB requirements)

22
HARP Cross Section Measurement
pion purity
migration matrix
acceptance
pion yield
tracking efficiency
pion efficiency

• Acceptance is determined using the MC (compare
  to MB requirements)
• Tracking Efficiency and Migration (no time to
  discuss today).

23
HARP Cross Section Measurement
pion purity
migration matrix
acceptance
pion yield
tracking efficiency
pion efficiency

• Acceptance is determined using the MC (compare
  to MB requirements)
• Tracking Efficiency and Migration (no time to
  discuss today).
• Raw Particle Yields and Efficiency and Purity of
  the selection.

24
MiniBooNE Beam Relevant Phase Space
Momentum distribution peaks at 1.5 GeV/c and
trails off at 6 GeV/c. Angular
distribution of pions is mostly below 200 mrad.
Acceptance in P for qylt50 mrad
qxlt200 mrad Acceptance in qx for
qylt50 mrad P gt 1 GeV
Momentum and Angular distribution of pions
decaying to a neutrino that passes through the MB
detector.
Acceptance of HARP forward detector
25
HARP Detector Overlapping PID Detectors
0 1 2 3 4 5 6
7 8 9 10
P (GeV)
CAL
p/p
TOF
CERENKOV
p/k
TOF ?
CERENKOV
p/e
TOF
CERENKOV
CALORIMETER
CERENKOV
26
HARP Detector Overlapping PID Detectors
0 1 2 3 4 5 6
7 8 9 10
P (GeV)
CAL
p/p
TOF
CERENKOV
p/k
TOF ?
CERENKOV
p/e
TOF
CERENKOV
CALORIMETER
CERENKOV
27
HARP Detector Overlapping PID Detectors
0 1 2 3 4 5 6
7 8 9 10
P (GeV)
CAL
p/p
TOF
CERENKOV
p/k
TOF ?
CERENKOV
p/e
TOF
CERENKOV
CALORIMETER
CERENKOV
Bayes Theorem
28
HARP Detector Overlapping PID Detectors
0 1 2 3 4 5 6
7 8 9 10
P (GeV)
CAL
p/p
TOF
CERENKOV
p/k
TOF ?
CERENKOV
p/e
TOF
CERENKOV
CALORIMETER
CERENKOV
momentum distribution
calorimeter
tof
cerenkov
29
Pion ID Beam Particles

• Use no target runs to determine correction
  factor for PID. Beam detector ID is considered
  true ID.
• PID Input (for 1st iteration) is found from
  crude cuts on detector data. But method is quite
  insensitive to starting input.
• Need MC to determine efficiency and purity for
  continuous p, q

PRELIMINARY
PRELIMINARY
PRELIMINARY
30
Pion ID Beryllium 5 Target

• Run iterative PID algorithm on Be 5 target data
  to extract raw pion yields.
• PID efficiency and purity determined using no
  target data (MC).
• Tracking efficiency determined using both data
  and MC.
• Acceptance determined from the MC.

PRELIMINARY
PRELIMINARY
31
Next Steps

• Continue to improve particle probability
  functions for the three detectors using data and
  MC.
• Implement tracking, PID, and acceptance
  corrections to raw particle yields.
• Move towards normalized pion cross section
  measurement.

Next Next Steps

• Study pion absorption and reinteraction effects
  in the thick target by using data from three
  different target lengths.
• How well can we do p/K separation?
• Finally, generate neutrino fluxes for MiniBooNE
  using measurements from HARP.