minos layout - harvard universityusers.physics.harvard.edu/~feldman/lecture_3.pdf · gary feldman...

Gary Feldman SLAC Summer Institute 14 - 25 August 2000 89

MINOS Layout


MINOS Far Detector

l 8m octagonal tracking calorimeter

l 486 layers of 1 in iron plates

l 4.1 cm-wide scintillator stripswith WLS fiber readout, readout from both ends

l 8 fibers summed on eachPMT pixel

l 25,800 m2 (6.4 acres) ofactive detector planes

l Toroidal magnetic field<B> = 1.3 T

l Total mass 5.4 kTHalf of MINOSfar detector


MINOS Near Detector

l 280 Òsquashed octagonÓ plates

l Same plate thickness,scintillator thickness andwidth as far detector

l Target/calorimetersection: 120 planesl 4/5 partial area

instrumented

l 1/5 full area instrumented

l Muon spectrometersection: 160 planesl 4/5 uninstrumented

l 1/5 full area instrumented

980 tons


Near Far Comments

Rates(8 µs spill)

~ 3 MHz ~ 30 Hz • Simulations ⇒ not a problem• Low intensity runs

Electronics Deadtime-less 19 nsdigitizations

Sampleand hold

• Calibrate in test beam

Readout Singleended w/reflector

Doubleended

• Similar light levels• Calibrate in test beam

Multiplexing None 8-fold • Simulations ⇒ not a problem

MINOSNear-Far Detector Differences


MINOS Energy Options

Different beamenergies correspond to different horncurrents and positions

Will start with low E beam for best sensitivity to match SK results


MINOS Near-Far EnergyDifferences: Hadronic Hose

l Want near and far energy spectra to be identical.

l This is impossible without focusing in the decaypipe:

l The solution is the hadronic hose Ñ a wirecarrying 1 kA down the center of the beam pipe:


MINOS: Advantages of theHadronic Hose

¥ Better near-far agreement ¥ More events


MINOS Physics Goals

l Verify dominant νµ → ντ oscillationsl τ appearance is not necessary.

l νµ CC disappearance with no NC disappearance andno νe CC appearance ⇒ νµ → ντ oscillations. There is noother possibility.

l Precise measurement of dominant ∆m2 andsin2(2θ).

l Search for subdominant νµ → νe and νµ → νs

oscillations.

l Study unconventional explanations: neutrinodecay, extra dimensions, etc.


MINOS Physics Tools

l νµ CC spectruml Information from both rates and shape. The latter is

independent of the near / far normalization.

l NC / CC ratiol Independent of the near / far normalization .

l νe CC appearancel Use topological criteria: fraction of energy in first few

radiation lengths, shower asymmetry, etc.


MINOS CC Spectra

Low-energy beam2 year run

Charge current spectra for sin2(2θ) = 1

Open = no oscillationShaded = oscillation


MINOS Sensitivity

CC E spectrum1σ contours2 year run


MINOS ντ vs. νs Discrimination

Sterile neutrinodiscriminationby NC/CC ratio


MINOS νµ → νe Sensitivity

90% c.l. νµ → νe sensitivity Preliminary graph Ñ improved calculations being done.

Electron identificationby topological criteria

N.B. Graph is incorrect (orat least misleading) fortwo reasons Ñ see nextslide


l Need to use 3-neutrino mixingl Assume that MINOS is only sensitive to the

mass scale. Then

and MINOSloses a factor of 2 insensitivity tocompared to CHOOZ

MINOS: νµ → νe Subtleties

P m L Ee( ) sin ( )sin ( )sin ( . / ).ν ν θ θµ → = 223

213

2 22 1 27∆ ∆ ∆m m13

2232≈

θ π23 4≈ /

θ13

m3

m2m1

} solar ∆m2

atmospheric ∆m2


MINOS: νµ → νe Subtleties

l Matter effects are not negligiblel Matter effects in the earth increase (or decrease) νµ → νe

oscillations by about 25% and increase (or decrease) theoscillation length slightly.

l If MINOS can see νµ → νe oscillations, then it canmeasure the sign of ∆m2 by measuringoscillations, where the sign of the matter effect reverses.

ν νµ → e

m3

m2m1

positive ∆m2

m2m1

m3

negative ∆m2


CERN Neutrino Beamto Gran Sasso (CNGS)


CNGS / NuMI Comparison

NuMI CNGS

p Energy (GeV) 120 400

pot / yr (1019) 27 4.5

νµ CC events / kT / yr(no oscillations)

3200(High E)

2450

Baseline (km) 730 730

Turn on 2003 2005

Near detector(s) Yes No*

*Bad choice


CNGS: Why No Near Detectors?

l CERN says that the purpose of CNGS is to doappearance experiments, particularly ντappearance, and for these experiments neardetectors are not necessary.

l The argument is weak:l Granting the argument for ντ appearance, observing

direct ντ appearance in 2005 will not be interesting. Therates are low (at ∆m2 = 3 10-3 eV2, 22 events / kT-yr, beforeefficiency corrections), so precise measurements cannotbe done.

l Two detectors are always better than one, since theyallow for a good control of systematic errors and providediscovery potential.


CNGS: OPERA Proposal

l OPERA is a proposal for direct detection of ντusing the ECC (emulsion cloud chamber)technique:


l Emulsions are in bricks. A brick with an event isidentified by electronic detectors and removed formeasurement.

OPERA Detector


OPERA Detector

Active mass 2 kT

τ detection efficiency 8.7%

⇒ 18 events in 5 years for ∆m2

= 3.2 10-3 eV2


ICANOE Proposal

l ICANOE is a marriage of the ICARUS and NOEproposals

l 4 modules, each with a 1.4 kT LAr TPC and a 0.8 kTsolid calorimeter ⇒ 9 kT total


ICANOE: Liquid Argon Detector

l Huge LAr TPCs have bubble-chamber like imagingcapabilities, excellent fully-active calorimetry, andexcellent particle identification from dE/dx andimaging.

A real quasi-elastic eventfrom a 50 liter testchamber in theCERN ν beam


ICANOE: Solid Detector

l The solid detectors consist of 5 mm magnetizediron plates sandwiched with scintillating fibers anddrift tubes.

l They serves as tail-catchers and crosschecks forthe LAr and a magnetic spectrometer.


ICANOE: ντ Appearance

l ICANOE will detect ντappearance by observinglow-energy excesselectrons from theτ → eνν decay.

l In 4 years, at ∆m2 = 3 10-3

eV2, there will be about 90such events in theICANOE LAr (100%detection efficiency).

l Analyses can also bedone with kinematic cutsfor low νe background.


ICANOE: νµ → νe Oscillations

l ICANOEÕs superb electron identificationcapabilities allows for a sensitive measurement ofθ13.

l However, it will be limited by systematic errorsfrom the lack of a near detector and, probably, thewrong beam tune.


The Future:A Neutrino Factory?

l A simple idea: Store muons in a ring with longstraight sections and observe neutrinos from themuon decays.

l Design can be race-track,triangle, or bow tie.Straight sections need topoint down to serve twoexperiments ⇒ arcs needto point up.


Neutrino Factory Basics

l Like the KARMEN experiment: if µ+ stored, then

l Normal CC give only µ+ and e-.

l oscillations give e+.

l oscillations give µ-.

l Need magnetic detection to see wrong-signleptons and very massive detectors for rate.

l Electron charge difficult ⇒ emphasis on wrong-sign muon detection.

µ ν νµ+ +→ e e .

ν νµ → e

ν νµe →


Neutrino Factory Advantages

l Beam background free.

l Well-known fluxes from monochromatic parents.

l Narrow band beam.l An advantage?

l Flux increases asEµ

3:l E2 for divergence.

l E for the crosssection.

l Is highest possible Ealways best?


Neutrino Factory Parametersand Physics Goals

l Parameters under discussion:l 20 ≤ Eµ ≤ 50 GeV

l Lower bound from pµ > 4 GeV for detection

l Higher Eµ is better

l 1019 to 1021 decays per year

l Baseline about 3000 kml Reason to be discussed

l Physics goals:l Precise measurement of θ13

l Determination of the sign of ∆m232

l Measurement of CP violation in lepton sector (δ)


Measurement of θ13 andSign of ∆m13

2

l If θ13 >> sin(2θ12) ∆m122 L /E (true for SMA, low, and vacuum

solar solutions), the situation is simple:

enhanced or reduced by matter effects, as discussed forMINOS.

P m L Ee( ) sin ( )sin ( )sin ( . / ),ν ν θ θµ → = 223

213

21322 1 27∆

0

1

2

3

4

5

0 1000 2000 3000 4000 5000 6000

Matter Effect Enhancement Factorat First Oscillation Maximum

Baseline (km)

∆m232 = 3 10-3 eV2

ρ = 2.8 g/cc

0

2

4

6

8

10

0 1000 2000 3000 4000 5000 6000

Energy of FirstOscillation Maximum

Mo

st S

ensi

tive

En

erg

y (G

eV)

Baseline (km)

ρ = 2.8 g/cc

∆m232 = 3 10-3 eV2



2

sin2(2θ13) reachfor 1019 decays(from Barger et al.,hep-ph/0003184 v2)



2

l Sensitivity to sin2θ13 at 90% c.l. from Cervera, et al., (hep-ph/0002108)

730 km disfavoredby backgrounds

7300 km disfavoredby statistics


Measurement of θ13,

the Sign of ∆m132, and δ

l However, if LMA solar oscillation scenario is correct, thingsare much more complicated (and interesting).

l Sensitivity to the Òatmospheric term,Ó the Òsolar term,Ó andan interference term between them, all modified by mattereffects. (The interference term contains the CP violation.)

l To simplify as much as possible, take sin2(2θ23) = sin2(2θ12) =1, and following Cervera, et al., expand to second order inthe small parameters θ13 and (∆m12

2 L/E), ignoring mattereffects:

P m L E m L E

m L E m L E m L E

e( ) sin ( . / ) . ( / )

. ( / )sin( . / )cos( . / )

ν ν θ

θ δµ → = +

+ ± −

2 1 27 0 81

1 27 1 27 1 27

132 2

132

122 2

13 122

132

132

∆ ∆

∆ ∆ ∆




l For a short baseline (NuMI or CNGS), each term has thesame L / E dependence and we measure

with a small modification due to matter effects.l If ∆m12

2 is known from solar measurements, then θ13 can berecovered with a small uncertainty due to the cos(δ) term.

l However, the cos(δ) term cannot be separated from θ13.

l The CP-violating sin(δ) term, which reverses with polarity, islower order and has been dropped.

P m m m m

m L E

e( ) . ( / ) ( / )cos( )

( . / )

ν ν θ θ δµ → = + +[ ]×

2 0 5

1 27

132

122

132 2

13 122

132

132 2

∆ ∆ ∆ ∆

∆




l At longer baselines, we want to use the different L / Edependence to untangle the different terms, including the CPviolating term:

l This requires a fairly large L / E: E ~ 6 GeV for L = 3000 km.

l The temptation is to increase L further. This does not work,at least for measuring CP violation, for a somewhat subtlereason.

P m L E m L E

m L E m L E m L E

e( ) sin ( . / ) . ( / )

. ( / )sin( . / )cos( . / )

ν ν θ

θ δµ → = +

+ ± −

2 1 27 0 81

1 27 1 27 1 27

132 2

132

122 2

13 122

132

132

∆ ∆

∆ ∆ ∆


Measuring CP Violation

l CP violation only occurs for three or more neutrino flavors.Therefore, it has to be proportional to every MNS matrixelement and every oscillatory term. That is why it occursonly in the interference term.

l For this reason, unless the solar solution is LMA, measuringCP violation is hopeless.

l Using the same LMA approximation, the CP odd term in theabsence of matter effects is

P

m LE

m LE

m LE

CP e− → = ±

×

( ) sin( )

sin.

sin.

sin.

ν ν θ δµ 2

1 27 1 27 1 27

13

122

132

232∆ ∆ ∆



l The problem with long (~8000 km) baselines arises due tothe matter effect for ∆m12

2.

l The effective ∆m2 added by the earth is 2AE, where

for earth density corresponding to L = 8000 km.

l In the presence of matter

l For reasonable energies, E ³ 4 GeV, 2AE is at least an orderof magnitude larger than ∆m12

2. Thus,

A G nF e= ≅ −2 1 5 10 4. eV /GeV2

sin

.

| |sin

. | |1 27

2

1 27 2122

122

122

122∆ ∆

∆∆m L

Em

AE m

AE m LE

→− ±

− ±

| | .− ± →2 2122AE m AE∆



l Therefore, the CP odd term contains a pure matteroscillation,

which goes to zero at 8000 km, independent of allparameters.

l Calculation by Cervera et al.

∆m232 = 2.8 10-3 eV2

∆m122 = 1.0 10-4 eV2

θ23 = 45o; θ12 = 22.5o

θ13 = 8o; δ = 90o

P

mAE

ALCP e− → ∝

( ) sin( . ),ν νµ∆ 12

2

22 54



l Fits by Cervera et al. for 1021 decays at 50 GeV with a 40 kTdetector.

l ∆m232 = 2.8 10-3 eV2; ∆m12

2 = 1.0 10-4 eV2

θ23 = 45o

θ12 = 22.5o

θ13 = 8o

δ = 54o


Where to Build aNeutrino Factory

l For a ~3000 km baseline, the choices are quitelimited.

minos layout - harvard universityusers.physics.harvard.edu/~feldman/lecture_3.pdf · gary feldman...

Documents