Sampling Detectors for n Detection and Identification

Report
Sampling Detectors for ne
Detection and Identification
Interest de jour: what is sin22q13 
• oscillations nm -> ne
• ‘superbeams’
• ‘Current’ generation of experiments
• How can we do better
• Sampling detectors for ne detection
Adam Para, Fermilab
NuFact02
Imperial College
Different baselines: where the
oscillation peaks are ?
L(km)/n
1
2
3
300
0.73
GeV
0.24
GeV
0.15
GeV
750
1.82
GeV
0.60
GeV
0.36
GeV
1500
3.64
GeV
1.21
GeV
0.73
GeV
Flux/rates drop
En < 1 GeV (KEK/JHF to
SuperK, CERN to Frejus
0.3 < En < 3 GeV (NuMI)
0.5< En < 6 GeV
(CERN to Taranto, BNL to ?)
Neutrino Cross Sections
N+lepton
N+l+p
Many particles
What will MINOS do?
Two functionally identical
neutrino detectors
"
Det. 1
Det. 2
ne Interactions in MINOS?
NC interactions:
NC, Eobs = 3 GeV

Energy distributed
over ‘large’ volume
ne CC interactions (low y) :
•Electromagnetic shower:
•Short
•Narrow
•Most of the energy in a
narrow cluster
Detector Granularity:
•Longitudinal: 1.5X0
energy
•Transverse: ~RM
ne CC, Etot = 3 GeV
Needle in a Haystack ? NC
Background
n spectrum
ne (|Ue3|2 = 0.001)
NC (visible
energy), no
rejection
ne background
Spectrum mismatch:
These neutrinos
contribute to
background, but no
signal
MINOS Limits on nm to ne
Oscillations
10 kton-yr exposure,
Dm2=0.003 eV2, |Ue3|2=0.01:
Signal (e = 25%) - 8.5 ev
ne background
- 5.6 ev
Other (NC,CC,nt) – 34.1 ev
M. Diwan,M. Mesier, B. Viren, L. Wai, NuMI-L-714
90% CL: | Ue3|2< 0.01
Limit comparable to a far
superior detector (ICARUS)
in CNGS beam
Sample of ne candidates defined using topological cuts
Receipe for a Better Experiment




More neutrinos in a signal region
Less background
Better detector (improved efficiency, improved rejection
against background)
Bigger detector
Lucky coincidences:
• distance to Soudan = 735 km, Dm2=0.025-0.035 eV2
1.27Dm2 L p

E
2
 E
2.54Dm2 L
p
 1.6  2.2 GeV
• Below the tau threshold! (BR(t->e)=17%)
Two body decay kinematics
At this angle, 15 mrad,
energy of produced
neutrinos is 1.5-2 GeV
for all pion energies 
very intense, narrow
band beam
‘On axis’: En=0.43Ep
pL   ( p* cosq *   E* )
pT  p* sin q *
Off-axis ‘magic’ ( D.Beavis at al.
BNL Proposal E-889)
1-3 GeV intense beams with well
defined energy in a cone around
the nominal beam direction
2
 2  A
Flux  
2 2 
1


q  4pz 2

NC/ ne /p0 detectors
CHARM II (nme scattering)
Challenges:
 Identify electrons
 Small cross section, large
background from NC
interactions
Solution:
•Low Z, fine grained
calorimeter
Detector(s) Challenge

Surface (or light overburden)



High rate of cosmic m’s
Cosmic-induced neutrons
But:



Duty cycle 0.5x10-5
Known direction
Observed energy > 1 GeV
Principal focus: electron neutrinos identification
• Good sampling (in terms of radiation/Moliere length)
Large mass:
• maximize mass/radiation length
• cheap
A possible detector: an example
Cheap low z absorber:
recycled plastic pellets
Cheapest detector: glass
RPC (?)
Constructing the detector ‘wall’


Containment issue: need very large detector
Engineering/assembly/practical issues
On the Importance of the Energy
Resolution
Cut around the expected
signal region too improve
signal/background ratio
M. Messier, Harvard U.
Energy resolution vis-à-vis
oscillation pattern


First oscillation
minimum: energy
resolution/beam
spectrum ~ 20%
well matched to the
width of the
structure
Second maximum:
20% beam width
broader than the
oscillation minimum,
need energy
resolution <10%.
Tails??
Energy Resolution of Digital
Sampling Calorimeter

Digital sampling
calorimeter:






1/3 X0 longitudinal
3 cm transverse
Energy = Cx(# of hits)
DE ~ 15% @ 2 GeV
DE ~ 10% 4-10 GeV
~15% non-linearity @ 8
GeV, no significant nongaussian tails
Improve energy resolution?
Total Absorption Calorimeter: HPWF
Energy resolution limited by fluctuations
of the undetected energy: nuclear binding
energy, neutrinos and not by sampling
fluctuations
‘Crude’ sampling calorimeter (CITFR), 10
cm steel, better energy resolution than
total absorption one (HPWF)
Neutrino energy, Quasi-elastics ?
m
nm + n → m + p
(Em , pm)
n
En 
mN Em  mm2 2
mN  Em  pm cos q m
En(reconstruct)
p
m events
s=80MeV
En(reconstruct) – En (True) (MeV)
~ 2 GeV: CC ne / NC interactions
~ 2 GeV: nm CC interaction
~ 7 GeV: CC ne / NC interactions
CC ne vs NC events: example

Electron candidate:






Long track
‘showering’ I.e. multiple
hits in a road around the
track
Large fraction of the
event energy
‘Small’ angle w.r.t. beam
NC background sample
reduced to 0.3% of the
final electron neutrino
sample (for 100%
oscillation probability)
35% efficiency for
detection/identification
of electron neutrinos
Detector questions/issues







What is the optimal absorber material (mostly an
engineering/cost question, if DX0 kept constant)
What longitudinal sampling (DX0)?
What is the desired density of the detector?
(containment/engineering/transverse segmentation)
Containment issues: fiducial volume vs total volume,
engineering issues: what is the practical detector size?
What is the detector technology (engineering/cost issue if
transverse segmentation kept constant)
What is the optimal transverse segmentation (e/p0,
saturation,…)
Can a detector cope with cosmic ray background? What is
the necessary timing resolution?

similar documents