Status of MiniBooNE Short
Baseline Neutrino Oscillation
Experiment
Jonathan Link
Columbia University
International Conference on Flavor Physics
October 4, 2005
Neutrino Phenomenology
If neutrinos have mass then they may oscillate between flavors.
n
ν e 0   U ei ν i
i 1
→
Schrödinger's Eq.
n
ν e t    U ei e
i ( p  mi2 / 2 E ) t
i 1
νi
With three neutrinos the mixing is governed by the MNS matrix
which relates the mass eigenstates (n1, n2 and n3) to the flavor
eigenstates.
 ν e   U e1 U e2 U e3  ν1 
  
 
 ν μ    U μ1 U μ2 U μ3  ν 2 
 ν   U U U  ν 
 τ   τ1 τ2 τ3  3 
For oscillations involving just two neutrino mass eigenstates the
oscillation probability simplifies to
P( ν e  νμ,τ )  sin 2 2θ13 sin 2 (1.27m132 L / Eν )
October 4, 2005
Jonathan Link, Columbia
ICFP05
Neutrino Oscillation Data
Unconfirmed
observation by LSND.
Seen by Super-K and confirmed by
many including K2K. (q23)
First observed by Ray Davis. Nailed
down by Super-K, SNO and
KamLAND. Presumed to be
dominated by mixing between states
1and 2 (or q12)
October 4, 2005
Jonathan Link, Columbia
ICFP05
The LSND Experiment
LSND took data from 1993-98
The full dataset represents nearly
49,000 Coulombs of protons on target.
p+ m+ nm
e+ nmne
ne p e+ n
n+H→D+γ
Energy range
of 20 to 55
MeV
LSND’s Signature
L/E of about
1m/MeV
Scintillation
Čerenkov
October 4, 2005
Baseline of 30
meters
2.2 MeV neutron capture
Jonathan Link, Columbia
ICFP05
The LSND Signal
They looked for an excess of ne events in a nm beam
They found 87.9 ± 22.4 ± 6.0
events over expectation.
With an oscillation probability
of (0.264 ± 0.067 ± 0.045)%.
Decay in flight analysis (nmne) oscillation
probability of (0.10 ± 0.16 ± 0.04) %
October 4, 2005
Jonathan Link, Columbia
ICFP05
Why is this Result Interesting?
LEP found that there are only 3
light neutrinos that interact weakly.
Three neutrinos allow only 2
independent m2 scales.
ν3
m22
ν2
ν1
m12
m32=m12+ m22
But there are experimental results
at 3 different m2 scales!?!
October 4, 2005
Jonathan Link, Columbia
ICFP05
What Does it Mean?
First, One or more of the experiments may be wrong
LSND being the leading candidate, has to be checked
→ MiniBooNE
Otherwise, add one or more sterile neutrinos…
Giving you more independent m2 scales
Best fits to the data require at least two sterile
mass
The Usual 3 ν Model
October 4, 2005
Jonathan Link, Columbia
ICFP05
A Conclusive Experiment is Needed
• With High Significance
– At least 3s over the entire LSND region
(including systematic and statistical uncertainties)
– Able to demonstrate energy dependence for oscillation
• Low and Different Systematics (Change the signature)
– Change the beam to higher energy
– Optimize detector for new signature
• High Statistics
– Significantly more events than LSND
The Mini Booster Neutrino Experiment, MiniBooNE, was formed.
The collaboration consists of about 60 scientists from 14
institutions.
October 4, 2005
Jonathan Link, Columbia
ICFP05
The MiniBooNE Neutrino Beam
nmne?
Start with an intense 8 GeV proton beam from the Booster.
In the Be target primarily pions are produced, but also some kaons.
Charged pions decay almost exclusively as p±m±nm.
K±p0e±ne, KLp±ene and m±e±ne contribute ne’s to background.
A toroidal field horn focuses the charged particles on the detector.
Initially positive particles will be focused selecting n.
The horn current can be reversed to select n.
Increases neutrino intensity by a factor of 5.
The horn is followed by a decay region.
The decay region is followed by an absorber and 450 m of dirt, beyond which
only the neutrino component of the beam survives.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Neutrino Flux at the Detector
The L/E is designed to be a good match to LSND at ~1 m/MeV.
Poscillation  sin 2 2θ sin 2 (1.27m2 L / E)
From beam simulations, the
expected intrinsic ne flux is
small compared to the nm
flux.
But the intrinsic ne flux is
comparable in size to an
LSND-like signal.
October 4, 2005
Jonathan Link, Columbia
ICFP05
The MiniBooNE Detector
12 meter diameter sphere
Filled with 950,000 liters of
pure mineral oil — 20+ meter
attenuation length
Light tight inner region with
1280 photomultiplier tubes
Outer veto region with 240
PMTs.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Background
Approximate number of events
and Background expected in
MiniBooNE
Signal
October 4, 2005
nm Charged Current, Quasi-elastic
500,000 events
Intrinsic νe (from K&μ decay) :
236 events
π0 mis-ID:
294 events
(Neutral Current Interaction)
Other νμ mis-ID:
140 events
LSND-like nmne signal:
300 events
Jonathan Link, Columbia
ICFP05
Particle Identification: m, e, and p0
Neutrino interactions in oil produce:
• Prompt Čerenkov light in a cone centered on the track.
• Delayed scintillation light distributed isotropically.
Čerenkov to scintillation ratio ~ 4 to 1
Particle ID is based on ring fuzziness, track length, ratio of prompt/late light.
Fuzzy rings distinguish electrons from muons.
p0 look like 2 electrons (usually)
Short
Exiting
October 4, 2005
Jonathan Link, Columbia
ICFP05
Beam Events
Simple trigger: takes 20 μs
about each beam spill.
1.6 μs
The spill is 1.6 μs wide.
Cutting on less than 6 veto
hits removes all primary
cosmic rays.
Requiring 200 tank hits
removes all μ decay, or
“Michel”, electrons leaving
only primary beam events.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Progress on Date Taking
Data Taking begain in September 2002.
We need at least 5×1020 protons on target (pot)…
We want 1×1021 pot
Minos Start-up
We’ve got 6.3×1020 so far
October 4, 2005
Jonathan Link, Columbia
ICFP05
Progress on the Analysis
The minimum data required for the analysis is in the bag, but no
oscillation analysis before the end of the year.
We still need…
Improved data on meson production (π and K) in proton beam
We have preliminary result from HARP on p-Be π thin target
production (Agrees well with Brookhaven E910)
We’ve had a first look at thick target data
(at most a small effect)
We are still waiting for HARP charged K cross sections.
Neutral kaons have to come from somewhere else (E910
analysis underway)
October 4, 2005
Jonathan Link, Columbia
ICFP05
Progress on the Analysis (continued)
Work continues to improve the optical model…
Without a near detector we have no choice but to demand
outstanding agreement between data and monte carlo.
We have many ex situ measurements of the oil properties.
This work has converged significantly in the last few
months, but it is not done yet.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Projected Coverage
Here we see the projected limits
for a null result with 1×1021 pot
Even with 7×1021 pot we should
still cover the entire 90% CL
region from LSND at 3σ
This sensitivity depends on
understanding the background
from intrinsic νe to 10% and the
background from mis-ID π0 to
5%
October 4, 2005
Jonathan Link, Columbia
ICFP05
Sensitivity to a Signal
Signal
Mis-ID
Intrinsic νe
Δm2 = 1 ev2
Δm2 = 0.4 ev2
The LSND question will soon be resolved
October 4, 2005
Jonathan Link, Columbia
ICFP05
Cross Section Physics
Great opportunity to contribute to neutrino interaction
cross sections in the MiniBooNE Energy Range.
Region with
MiniBooNE
Coverage
October 4, 2005
Even less data on anti-neutrino
cross sections!
Jonathan Link, Columbia
ICFP05
Charged Current π+ Cross Section
Nice triple coincident
signature
Absolute cross section is
calculated relative to the
theoretical charged current
quasi-elastic (ν+C → μ++C)
cross section.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Conclusions and Outlook
• We have over 61020 protons on target in n mode.
• With this data we will confirm or rule out the full high m2
oscillation range of LSND (CP conserving). Results Soon.
• If no signal is seen in n mode, n running is needed to investigate
possible 3+2 CP violating modles.
• We are working towards several interesting cross section results
– Charged current π+
– Neutral Current π0
• Anti-neutrino run approved for 2006. Initially to study cross
sections but can be extended to do oscillations.
• Possible upgrade to BooNE, a two detector experiment to carefully
measure m2 and look for nm disappearance.
October 4, 2005
Jonathan Link, Columbia
ICFP05
The BooNE Collaboration
October 4, 2005
Jonathan Link, Columbia
ICFP05
Inside the MiniBooNE Detector
PMTs at the bottom of the
detector just before sealing
up the inner region.
View of the Veto Region as
the first oil is added to the
detector.
October 4, 2005
Jonathan Link, Columbia
ICFP05
Beam Survey Experiments
Experiments E910 at Brookhaven and HARP at CERN are
studying K and p production with medium energy proton
beams on beryllium.
HARP took data using our
target with 8 GeV protons.
These data sets are still
being analyzed.
The results will be the
primary input to our
neutrino flux simulations.
The HARP Experiment at CERN
October 4, 2005
Jonathan Link, Columbia
ICFP05
Other Related Data
Several other experiments have looked
for oscillations in this region.
Allowed Region from Joint
Karmen and LSND fit
The most restrictive limits come from
the Karmen Experiment.
October 4, 2005
Jonathan Link, Columbia
From
Church, Eitel, Mills, & Steidl
hep-ex/0203023
ICFP05
Calibration Systems
Laser flasks provide PMT charge and
timing calibration and a means to monitor
the oil attenuation length in situ.
scintillator
cube
October 4, 2005
Muon tracker above detector and 7 optically
isolated scintillator cubes in the detector
provide cross checks for energy estimation
and reconstruction algorithms.
Jonathan Link, Columbia
ICFP05
Other Calibration Studies
Cosmic Muons
t = 2.12 ± 0.05 ms
With 8% m capture
on carbon, expected
m lifetime in oil is
2.13 ms
e Energy from m Decay
p0 Mass
Can be used to check energy
calibration at the relevant
energy scale.
October 4, 2005
Jonathan Link, Columbia
>200 hits in Tank
<6 hits in Veto
>10 p.e. in each ring
ICFP05
CCπ+
October 4, 2005
Jonathan Link, Columbia
ICFP05
CCπ+
CCπ+
October 4, 2005
Jonathan Link, Columbia
ICFP05
The Little Muon Counter (LMC)
• Detects muons at an angle of 7° from the beam center.
• At this angle all muons are from kaon decays.
• Gives us an important data point on kaons in our own
beam line.
The LMC drift pipe during construction
October 4, 2005
Jonathan Link, Columbia
ICFP05