annecy 29 septembre 2006 alain blondel future accelerator neutrino beams for europe 1. where do we...
Post on 15-Jan-2016
215 views
TRANSCRIPT
Annecy 29 septembre 2006 Alain Blondel
Future accelerator neutrino beams for Europe
1. where do we stand?
2. possible options
2.1 CERN LHC injector upgrade and CNGS+
2.2 SPL Superbeam 2.3 Beta-beam baseline and new ideas 2.4 Neutrino factory how do they compare3. outlook
NB: status as of NUFACT06+ ISS
Annecy 29 septembre 2006 Alain Blondel
There are today THREE compelling and firmly established observational facts that the Standard Model fails to account for:
-- neutrino masses -- the existence of dark matter -- the baryon asymmetry of the universe
The fact that neutrino have masses and mix is established by neutrino oscillations
The neutrino masses offer a chance to explain the baryon asymmetry in the most natural way via
*** LEPTOGENESIS ***
by a combination of -- fermion number violation (authorized by neutrino masses and GUT)-- three families of neutrinos ==> leptonic CP violation (authorized by the mixing of three families with large mixing angles)
Annecy 29 septembre 2006 Alain Blondels u m m e r c o n f . s u m m a r y B S M A la i n B lo n d e l
C ita t io n s o f F u k u gita & Y a n a gid a , P L B 1 7 4 (1 9 8 6 ) 4 5o r “ le p to ge n e sis” in t i t le (SP IR E S d a ta b a se )
0
2 0
4 0
6 0
8 0
1 0 0
1 2 0
1 4 0
1 6 0
1 9 7 0 1 9 7 3 1 9 7 6 1 9 7 9 1 9 8 2 1 9 8 5 1 9 8 8 1 9 9 1 1 9 9 4 1 9 9 7 2 0 0 0 2 0 0 3
(K a y s e r )
E W B a r y o ge ne s is (f r om C K M C P vio la t io n) h a s h it d iffi c ult ie s w h e n H iggs l im it e x c e e d e d ~4 0 G e V (1 9 9 3 ) … t h e f a vo r it e t h e or y i s no w L e pt o ge ne s is .
Annecy 29 septembre 2006 Alain Blondel
1. We know that there are three families of active, light neutrinos (LEP)2. Solar neutrino oscillations are established (Homestake+Gallium+Kam+SK+SNO+KamLAND)3. Atmospheric () oscillations are established (IMB+Kam+SK+Macro+Sudan+K2K)
4. At that frequency, electron neutrino oscillations are small (CHOOZ) This allows a consistent picture with 3-family oscillations ~300 m12
2~8 10-5eV2 ~450 m23 2~ 2.5 10-3eV2 ~ 100 with several unknown parameters mass hierarchy
Where are we?
*)to set the scale: CP violation in quarks was discovered in 1964 and there is still an important program (K0pi0, B-factories, Neutron EDM, LHCb, BTeV….) to go on for >>10 years…i.e. a total of >50 yrs.
and we have not discovered leptonic CP yet!
5. LSND ? ( miniBooNe) This result is not consistent with three families of neutrinos oscillating, and is not supported (nor is it completely contradicted) by other experiments.
If confirmed, this would be even more exciting See Barger et al PRD 63 033002
Where do we go? leptonic CP & T violations=> an exciting experimental program for at least 25 years *)
Annecy 29 septembre 2006 Alain Blondel
The neutrino mixing matrix: 3 angles and a phase
Unknown or poorly known even after approved program:13 , phase , sign of m13
OR?
m223= 2 10-3eV2
m212= 8 10-5 eV2
23(atmospheric) = 450 , 12(solar) = 320 , 13(Chooz) < 130
2
m212= 8 10-5 eV2
m223= 2 10-3eV2
Annecy 29 septembre 2006 Alain Blondel
Kayser -- EPS05
Accelerator neutrinos are CENTRAL to the future program.
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
An ambitious neutrino programme is a distinct, exciting possibility,
but we must be well prepared to have a good proposal in time for the big decision period in 2010 (Funding window: 2011-2020)
Avenues that have been identified as promising:
a) Off axis Superbeam (T2HK, NOvA, NOvA2, CNGS+?) b) SuperBeam + Beta-Beam + Megaton detector (SB+BB+MD)c) Neutrino Factory (NuFact) + magnetic detector (100kton)
The physics abilities of the neutrino factory seem superior in particular for flux & normalisationbut….. « what is the realistic time scale? »
(Hardware) cost estimate of a neutrino factory ~1B€ + detectors. This needs to be verifed and ascertained on a localized scenario (CERN,
RAL…) and accounting. The cost of a (BB+SB+MD) is not very different, though perhaps lower
Cost/physics performance/feasibility comparison needed
Annecy 29 septembre 2006 Alain Blondel
= ACP sinsolar term…
sinsin (m212 L/4E) sin
… need large values of sin m212 (LMA) but *not* large sin2
… need APPEARANCE … P(ee) is time reversal symmetric (reactor s do not work)
… can be large (30%) for suppressed channel (one small angle vs two large)
at wavelength at which ‘solar’ = ‘atmospheric’ and for e ,
… asymmetry is opposite for e and e
P(e) - P(e)
P(e) + P(e)
CP violation
P(e) = ¦A¦2+¦S¦2 + 2 A S sin
P(e) = ¦A¦2+¦S¦2 - 2 A S sin
An interference phenomenon:
Annecy 29 septembre 2006 Alain Blondel
3-family oscillations:
I There will be ↔ e and ↔ e
oscillation at L atm
P (↔ e )max =~ ½ sin 22 13 +… (small)
II There will be CP or T violation
CP: P (↔ e) ≠ P (↔ e
T : P (↔ e) ≠ P (e ↔
III we do not know if the neutrino which contains more e is the lightest one (natural?) or not.
Oscillation maximum 1.27 m2 L / E =/2
Atmospheric m 2= 2.5 10-3 eV 2 L = 500 km @ 1 GeVSolar m2 = 7 10-5 eV2 L = 18000km @ 1 GeV
Oscillations of 250 MeV neutrinos;
P (↔ e)
Annecy 29 septembre 2006 Alain Blondel
Mezzetto
L= /2.54 E/m
l = /2.54 E/m
Three family oscillations look at e oscillation
Annecy 29 septembre 2006 Alain Blondel
T asymmetry for sin = 1
0.10 0.30 10 30 90
!asymmetry is
a few % and requires
excellent flux normalization
(neutrino fact., beta beam or
off axis beam withnot-too-near
near detector)
NOTES:1. sensitivity is more or lessindependent of 13 down to
max. asymmetry point
2. This is at first maximum!Sensitivity at low valuesof 13 is better for shortbaselines, sensitivity atlarge values of 13 isbetter for longer baselines(2d max or 3d max.)
3.sign of asymmetry changes with max. number.
error
Asymmetry
Annecy 29 septembre 2006 Alain Blondel
Neutrino Beams in Europe
-- potential LHC injector upgrade
-- CNGS and upgrade considerations
-- Superbeams from proton driver SPL -- other
-- beta-beams
-- neutrino factory
-- towards FP7 design studies
Annecy 29 septembre 2006 Alain Blondel
Upgrade of the proton accelerator complex at CERNProtons Accelerators for the Future (PAF) WG
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)Present chain:
weak link in Linac 2and in the PS
(old!)
Annecy 29 septembre 2006 Alain Blondel
Priority is given to LHC and efforts should be made to incorporate the demands of the High intensity neutrino programmethe cheapest way to LHC luminosity consolidation is to -- implement the LINAC 4 and replace the CERN PS
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
Step I: replace linac 2 by Linac 4
increase injection rate no major improvement
for neutrinos~2011
Annecy 29 septembre 2006 Alain Blondel
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
Step II:
new PS2 (5-50 GeV)PS remains in operationfor injectionat 5 GeV in PS2
possible increase of SPS intensity --> CNGS+
~2015
Priority is given to LHC and efforts should be made to incorporate the demands of the High intensity neutrino programmethe cheapest way to LHC luminosity consolidation is to -- implement the LINAC 4 and replace the CERN PS
Annecy 29 septembre 2006 Alain Blondel
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
PSB SPL’RCPSB
SPSSPS+
Linac4
SPL
PS
LHC / SLHC DLHC
Out
put
ener
gy
160 MeV
1.4 GeV~ 5 GeV
26 GeV40 – 60 GeV
450 GeV1 TeV
7 TeV~ 14 TeV
Linac250 MeV
SPL: Superconducting Proton Linac (~ 5 GeV)
SPL’: RCPSB injector(0.16 to 0.4-1 GeV)
RCPSB: Rapid Cycling PSB(0.4-1 to ~ 5 GeV)
PS2: High Energy PS(~ 5 to 50 GeV – 0.3 Hz)
PS2+: Superconducting PS(~ 5 to 50 GeV – 0.3 Hz)
SPS+: Superconducting SPS(50 to1000 GeV)
SLHC: “Superluminosity” LHC(up to 1035 cm-2s-1)
DLHC: “Double energy” LHC(1 to ~14 TeV)
Proton flux / Beam power
PS2 (PS2+)
Step III:
New SPL (or RCS) to ~5 GeV
inject directly in PS2Multi-MW oportunity
@~5 GeVno date yet (i.e. a few more years)
Annecy 29 septembre 2006 Alain Blondel
Intensity increase to CNGS?
can one launch an off axis programme similar to T2K and NUMI-off-axis?
-- present neutrino beam optimized for High energy (tau appearance) ==> factor >~10 less flux at off axis energy than T2K-- no near detector!
A.Rubbia et al, A. Ball et al, have proposed a low energy version of CNGS with different target and more compact optics,run off axis (E
~800 MeV for C2GT, 1.5-2 GeV GeV for Larg
A. Ball et al(C2GT) CERN-PH-EP-2006-002A. Rubbia, P. Sala JHEP 0209 (2002) 004[arXiv:hep-ph/0207084].A. Meregaglia and A. Rubbia hep-ph/0609106
Annecy 29 septembre 2006 Alain Blondel
target and horn
1.5 Mton of water in the Golf of Taranto for 25 1019 pot = 5yrs
C2GT off axis2d maximum
detectormodule
--> sensitivity (90%) to sin213= 0.0076
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
hep-ph/0609106 Imagine: 100 kton Larg detectorat 0.750 off-axis 850 km (1st max) --> search
or 1.50 off-axis 1050 km 2d maxCP violation and matter effect
or sharing 1st and 2d maximum
assume all of 50 GeV 200 kW PS2 accelerated to 400 GeV==> CNGS+ = 30 1019 pot/year
<-- sensitivity sin 2213 ~ 10-3
(2026)
5years 5 years
Annecy 29 septembre 2006 Alain Blondel
thanks to and running, sensitivity to and matter effects
example (90%)for‘known hierarchy’(assume that hierarchy is given by comparison with another expt)
Annecy 29 septembre 2006 Alain Blondel
e physics at CNGS+?
Basic issues to solve:
1. no near detector --> no knowledge of absolute cross sections(at osc. max there are no to normalize…) difficult to measure absolute rates of e and to compare vs or different energies for CP or matter effect
2. modifications of CNGS beam line are necessary. possible? perhaps easier to build new dk tunnel -- with adequate length and near detector. then why keep the same direction?
3. can SPS really handle 4x more protons?
4. 100 kton Larg or 1Mton water are large investments -- may be they deserve better!
Annecy 29 septembre 2006 Alain Blondel
LINAC4--> PS2: an opportunity for MultiMW physics
Eventually the PS should be phased out completely: need for a machine that bridges 1.4 (booster) to 5 GeV, or better 0.16 Linac4 to 5 GeV (PS2)
Superconducting Proton Liac or Rapid Cycling Synchrotronboth fast cycling (O(10-50 Hz). potentially a high power machine serving -- LHC -- neutrinos -- nuclear physics (Eurisol)
for neutrino physics:conventional p decay superbeam proton driver for neutrino factory
Annecy 29 septembre 2006 Alain Blondel
Super-beams: SPL-Frejus
TRE
CERN SPLLSM-Fréjus
Near detector
130km
Annecy 29 septembre 2006 Alain Blondel
SPL (2.2 GeV) superbeam 20m decay tunnelsingle open horn, L Hg target
Low energy --> low Kaon ratebetter controlled e
contamination
Annecy 29 septembre 2006 Alain Blondel
2 years run to 440 ktonFrejusEMeVsmall cross-sectionslimited sensitivity( sin 2213 ~ 210-3)
near detector design?
Main technical issues-- 50 Hz horn operation-- handling of 4 MW in target and environment.
Annecy 29 septembre 2006 Alain Blondel
CERN: -beam baseline scenario
PS
Decay
RingISOL target & Ion source
SPL
Cyclotrons, linac or FFAG
Decay ring
B = 5 T
Lss = 2500 m
SPSECR
Rapid cycling synchrotron
Nuclear Physics
,
Same detectors as Superbeam !
target!
Stacking!
neutrinos of Emax=~600MeV
eFNe e189
1810
eLiHe e63
62
Annecy 29 septembre 2006 Alain Blondel
3.5 GeV SPL
beam
-- low proton energy: no Kaons e background is low--region below pion threshold (low bkg from pions)
but:low event rate and uncertainties on cross-sections
Annecy 29 septembre 2006 Alain Blondel
Combination of beta beam with super beam
combines CP and T violation tests
e (+) (T) e (+)
(CP)
e (-) (T) e (-)
Annecy 29 septembre 2006 Alain Blondel
Eurisol baseline Study
CERN site (use PS and SPS as are) -- could benefit from PS2Max. ion in CERN SPS is 450 GeV Z/Mion
= 150 for 6He, = 250 for 18Ne ==> EeV
2.9*1018 /yr anti-e from 6He Or 1.1*1018 /yr e from 18Ne (1017 with avail. tech.)
race track (one baseline) or triangle (2 base lines) so far study CERN--> Fréjus (130km)
longer baseline ~ 2-300km would be optimal + moderate cost: ion sources, 450 GeV equiv. storage ring (O(0.5M€))+ no need for 4MW target
Emax
=2. Q0. ion
Annecy 29 septembre 2006 Alain Blondel
J-E. Campagne et al. hep/ph0603172
combine SPL(3.5 GeV) + B==> improves sensitivity by T violation!
10 year exposure
issues:
-- 18Ne flux?-- low energy --> cross-section accuracy? (assume 2%) -- energy reconstruction OK-- near detector concept?
sensitivity sin2213 ~2-5 10-4
3 sensitivity to sin2213
Annecy 29 septembre 2006 Alain Blondel
Better beta beams:
main weakness of He/He beta-beam is low energy (450 GeV proton equiv. storage ring produces 600 MeV neutrinos)
Solution 1: Higher (Hernandez et al) Use SPS+ (1 TeV) or tevatron ==> reach expensive!
Solution 2: use higher Q isotopes (C.Rubbia)
8B --> 8Be e+ e
or 8Li --> 8Be e-
anti- e
Annecy 29 septembre 2006 Alain Blondel
A possible solution to the ion production shortage:
Direct production in a small storage ring, filled [Gas + RF cavity] for ionization cooling
For 8B or 8Li production, strip-inject 6Li / 7Li beam, collide with gas jet (D2 or 3He)
reaction products are ejected and collected
goal: >~ 1021 ions per year
Annecy 29 septembre 2006 Alain Blondel
Advantages of 8B5+ (e Q=18MeV ) or 8Li3+ (anti-e Q=16MeV)vs 18Ne, 6He (Q~=3 MeV)
The storage ring rigidity is considerably lower for a given E
==> for ~1 GeV end point beam for 8B5+ : 45 GeV proton equiv. storage ringfor 8Li3+: 75 GeV proton equiv. storage ring
Two ways to see it: 1. Beta-beams to Fréjus (Emax =600 MeV) could be accelerated with PS2 into a 50 GeV proton-equivalent storage ring (save €)2. Beta beams of both polarities up to end-point energy of ~6 GeV can be produced with the CERN SPS (up to 2000km baseline)
A new flurry of opportunities
Annecy 29 septembre 2006 Alain Blondel
EC: A monochromatic neutrino beam
Decay T1/2 BR EC/ ECI B(GT) EGR GR QEC E E
148Dy 148Tb* 3.1 m 1 0.96 0.96 0.46 620 2682 2062
150Dy 150Tb* 7.2 m 0.64 1 1 0.32 397 1794 1397
152Tm2- 152ET* 8.0 s 1 0.45 0.50 0.48 4300 520 8700 4400 520
150Ho2- 150Dy* 72 s 1 0.77 0.56 0.25 4400 400 7400 3000 400
Electron Capture: N+e- N’+e Burget et al
Annecy 29 septembre 2006 Alain Blondel
-- Neutrino Factory -- CERN layout --
e+ e
_
interacts
giving
oscillates e
interacts giving
WRONG SIGN MUON
Golden Channel
1016p/s
1.2 1014 s =1.2 1021 yr
3 1020 eyr
3 1020 yr
0.9 1021 yr
target!cooling!
acceleration!
also (unique) eSilver channel
Annecy 29 septembre 2006 Alain Blondel
NB: This works just as well
INO ~7000 km (Magic distance)
Annecy 29 septembre 2006 Alain Blondel
Neutrino fluxes + -> e+ e
/ e ratio reversed by switching
e spectra are different No high energy tail.
Very well known flux (10-3)
-- E& calibration from muon spin precession
-- angular divergence: small effect if < 0.2/
-- absolute flux measured from muon current or by e -> e in near expt.
-- in triangle ring, muon polarization precesses and averages out (preferred, -> calib of energy, energy spread)
Similar comments apply to beta beam, except spin 0 Energy and energy spread have to be obtained from the properties of the storage ring (Trajectories, RF volts and frequency, etc…)
polarization controls e flux:
+ -X> e in forward direction
Annecy 29 septembre 2006 Alain Blondel
Magnetized Iron calorimeter(baseline detector, Cervera, Nelson)
B = 1 T = 15 m, L = 25 m
t(iron) =4cm, t(sc)=1cm
Fiducial mass = 100 kT
Charge discrimination down to 1 GeV
Baseline
3500 Km
732 Km 108
4 x 106
2 x 108
7.5 x 106
3.4 x 105
3 x 105
CC e CC signal (sin2 13=0.01)
Event rates for 1020 muon decays (<~1 year)
(J-PARC I SK = 40)
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
A Strawman Concept for a Nufact Iron Tracker Detector (Jeff Nelson)
15m diameter polygon4 piece laminateCan be thin if planes
interconnectede.g. down to 1cm
Idea from 1st NOVA Proposal
60kA-turn central coil0.5m x 0.5mAverage field of 1.5TExtrapolation of MINOS
Triangular liquid scintillator cellsStructure based on NOvA using
MINERvA-like shapes4cm x 6cm cells (starting point)3mm thick PVC wallsLooped WLS fibers & APDs
A sample would look like1 cm Fe0.7 cm PVC3.3 cm LS2/3rds Fe; ρ ≈ 2
Based on 175M$ for 90kt
6 cm
4 cm
NB: in fact it is not best to go to iron:scint=1:4 cm
Dp/P is better by 20% for iron:scint= 4:2 cm
--> probably more mass/better cost possible
Annecy 29 septembre 2006 Alain Blondel
This ‘conceptual detector’ was used for the sensitivity studies. Performance from MINOS proposal
CUTS optimized for low 13 limits with cut-off at muon energy of 5 GeV
resulting signal efficiency:
13.1 m
14.5 m B B
Monolith
Iron (4cm) + active (1cm) Iron (4cm) + active (1cm) NUFACT detector studied so far:
30 mat 3000 km, 1st max is at 6 GeV2d max is at 2 GeV
Annecy 29 septembre 2006 Alain Blondel
at 3000 km, 1st max is at 6 GeV2d max is at 2 GeV
Annecy 29 septembre 2006 Alain Blondel
New analysis (Cervera) OLD: P> 5 GeV
NEW: L > Lhad + 75cm
(shown for three different purity levels down to << 10-4 )
old analysis
new analysis
Annecy 29 septembre 2006 Alain BlondelLocation of INOLocation of INO
Annecy 29 septembre 2006 Alain Blondel
INO Detector ConceptINO Detector Concept
Annecy 29 septembre 2006 Alain Blondel
-- Emulsion detector:
this is a typical NUFACT detector for the
silver channel eEGeV
experience from OPERA silver channel has opposite sign dependence than golden--> solving ambiguities, test unitarity.
By necessity less precisethan golden channel. Here 5kton ECC combined with+ 40 kton LMDonly tau--> muon decay channel
Annecy 29 septembre 2006 Alain Blondel
supermodule
8 m
Target Trackers
Pb/Em. target
ECC emulsion analysis:
Vertex, decay kink e/ ID, multiple scattering, kinematics
Extract selected brick
Pb/Em. brick
8 cm Pb 1 mm
Basic “cell”
Emulsion
trigger and locate the neutrino interactions muon identification and momentum/charge measurement
Electronic detectors:
Brick finding, muon ID, charge and p
Link to muon ID,Candidate event
Spectrometer
p/p < 20%
Annecy 29 septembre 2006 Alain Blondel
LARGE MAGNETIC VOLUME
Observing the platinum channel
or the silver channel for more decay channels
requires a dedicated Low Z and very fine grained detector
immersed in a large magnetic volume
Annecy 29 septembre 2006 Alain Blondel 4
Magnet
Steel
Magnetic Cavern
Multiple Solenoids - Conceptual Layouts
15 m x 15 m x 15m modules; B = 0.5T
Magnetic Tunnel
Annecy 29 septembre 2006 Alain Blondel
3rd International Scoping Study Meeting Rutherford Appleton Laboratory
14
1.2 Totally Active Scintillation Detector 1.2 Totally Active Scintillation Detector
turns
10 solenoids next to each other. Horizontal field perpendicular to beamEach: 750 turns, 4500 amps, 0.2 Tesla. 42 MJoules . 5Meuros.Total: 420 MJoules (CMS: 2700 MJoules)Coil: Aluminium (Alain: LN2 cooled).
Problem: Periodic coil material every 15m:Increase length of solenoid along beam?
How thick?
Possible magnet schemes for TASD Camilleri, Bross
Annecy 29 septembre 2006 Alain Blondel 3
Options
Technologies Room temperature Cu or Al conductor –Well, maybe
Power dissipation is high and field is relatively low– But I will consider it
High Tcsuperconductor –NO*
At this point in time for the same Ampere-Turns: 200X more expensive than convention SC
*However, development progress in recent years has been rapid so the situation could change in the near (5 yr) future.
Conventional SCLots of experience, but this size is new.Technically –certainly doableBUT WHAT IS THE COST?
Annecy 29 septembre 2006 Alain Blondel
8
Cost Extrapolations for Baseline NF Detector ONE Magnet
Cost via stored energy Stored energy 340 MJ From Green and Lorant
C(M$) 0.5(340)0.66224M$
Cost via Magnetic Volume From Green and Lorant
C(M$) 0.4(.5 X 3400)0.63545M$
Reference Point –CMS Solenoid C(M$) 0.5(2700)0.66293M$ (Stored energy) C(M$) 0.4(4 X 370)0.63541M$ (Magnetic volume)
Most Optimistic Extrapolation Use stored energy and conclude formula overestimatesby factor
of 1.7 (93/54) based on CMS case Then NF magnet extrapolated cost –14M$
Most Pessimistic Extrapolation Use magnetic volume and conclude formula underestimatesby a
factor of 1.3 (54/41) based on CMS case Then NF magnet extrapolated cost –60M$
X 10 Magnets=140-600M$conventional SC magnet:
Annecy 29 septembre 2006 Alain Blondel
Also considered was a conventional Al magnet. Cost = ~50 M$, but power cost is about 10-30 M$ a year.
Clearly a more detailed estimate would be highly worthwhile to understand what is the best way to build a large magnet on the walls of a large underground cavern.
SC Magnets of this size can certainly be built, but better cost estimates will only come after some real engineering analysis
6-12 month effort
It could be very expensive and unpractical therefore this is not considered to be a baseline nevertheless it seems very worthwhile to study.
Annecy 29 septembre 2006 Alain Blondel 15
Conclusions II
At this time it appears that a large volume air-core magnetized detector for a neutrino factory is not feasible from cost considerations, but is certainly technically feasible
R&D aimed at the mechanical engineering issues is required to see if the costs can be reduced.
Developments in high TcSC could change this picture Reduction in cost of the high Tcconductor itself Possibility for non-vacuum insulated vessels (Icarus
example) for SC operating at 77K
There is a long way to go to make this viable
Annecy 29 septembre 2006 Alain Blondel
x ~ a few X0=14cm….
B > 0.5 T
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
Magnetized ECC structure
We have focused on the “target + spectrometer” optimization
Electronic det:e//µ separator
&“Time stamp”
Rohacell® plateemulsion films35 stainless steel plates
spectrometer
target
shower absorber4.5 cm, 2 X0
Annecy 29 septembre 2006 Alain Blondel
µ end electron momentum resolution: 3 gaps (3cm thick) and 0.5 T
For the electron only hits associated to the primary electrons u sed in the parabolic fit (Kalman not used)Given the non negligible energy loss in the target, the electron energy is taken downstream for the comparison of true against reconstructed
µ electron
FIRST CONVINCING DEMONSTRATION THAT THE PLATINUM CHANNEL COULD BE USED!
Annecy 29 septembre 2006 Alain Blondel
A revealing comparison:
A detailed comparison of the capability of observing CP violation was performed by P. Huber (+M. Mezzetto and AB) on the following grounds
-- GLOBES was used.
-- T2HK from LOI: 1000kt , 4MW beam power, 6 years anti-neutrinos, 2 years neutrinos. systematic errors on background and signal: 5%.
-- The beta-beam 5.8 1018 He dk/year 2.2 1018 Ne dk/year (5 +5yrs) The Superbeam from 3.5 GeV SPL and 4 MW.Same 500kton detectorSystematic errors on signal efficiency (or cross-sections) and bkgs are 2% or 5%.
--NUFACT 3.1 1020 + and 3.1 1020 + per year for 10 years 100 kton iron-scintillator at 3000km and 30 kton at 7000km (e.g. INO). The matter density errors of the two baselines (uncorrelated): 2 to 5% The systematics are 0.1% on the signal and 20% on the background, uncorrelated.
all correlations, ambiguities, etc… taken into account
Annecy 29 septembre 2006 Alain Blondel
[00-900 ] [900-1800 ]
Annecy 29 septembre 2006 Alain Blondel
[2700-3600 ] [1800-2700 ]
NB: 3sigma = 60 means that +-1 sigma = +-3.50
Annecy 29 septembre 2006 Alain Blondel
What do we learn?
1. Both (BB+SB+MD) and NUFACT outperform e.g. T2HK on most cases.
2. combination of BB+SB is really powerful.
3. for sin22below 0.01 NUFACT as such outperforms anyone
4. for large values of systematic errors dominate. Matter effects for NUFACT, cross-sections for low energy beams.This is because we are at first maximum or above, CP asymmetry is small!
matter effect for NUFACT
Annecy 29 septembre 2006 Alain Blondel
Consequences - 1
3. for sin22below 0.01 NUFACT as such outperforms anyone
by 2010-12 we will know if this is the case:
Annecy 29 septembre 2006 Alain Blondel
Consequences -2
4. for large values of systematic errors dominate.
Matter effects for NUFACT, cross-sections for low energy beams.
This is because we are at first maximum or above, CP asymmetry is small!
for NUFACT:
work on understanding systematic errors on matter effect
try to reach second maximum by lowering the muon detection threshold
try to achieve wrong sign electron detection
for superbeam/betabeam:
must have a near detector concept that demonstrates ability to measure
detection and efficiency with high precision
going to second maximum requires a different baseline X 3
cf: project in Corea for T2HK baseline, or NUMI 2d max (farther off-axis)
not easy to achieve for both Superbeam and Beta-beam
Annecy 29 septembre 2006 Alain Blondel
for NUFACT:
work on systematic errors on matter effect
A preliminary study was made by
E. Kozlovskaya, J. Peltoniemi, J. Sarkamo,
The density distribution in the Earth along the CERN-Pyhäsalmi baseline and its effect on neutrino oscillations. CUPP-07/2003
the uncertainties on matter effects are at the level of a few%
J. Peltoniemi
Annecy 29 septembre 2006 Alain Blondel
Annecy 29 septembre 2006 Alain Blondel
Errors in density
1.5%1.8%Oceanic 9000 km
1.7%2.0%Continental 9000 km
1.7%2.6%Oceanic 2500 km
2.9%4.7%Continental 2500 km
“best”“a priori”location length
Errors are ~2 sigma(errors not really Gaussian)
Recommendations
Avoid:
• Alps
• central Europe
• thick crust (e.g. Fenoscandia)
• Europe to Japan
Recommendations
Best profiles:
• Western Europe to Eastern US
• Atlantic Islands (Canaries, Maderia, Azores) to Portugal, western Spain, NW France, southern Ireland, western England
Such a study, in collaboration with geophysicistswill be needed for candidate LBL sites
ISS-3 at RAL Warner
Annecy 29 septembre 2006 Alain Blondel
-- Near detectors and flux instrumentation
-- flux and cross-section determinations -- other neutrino physics
a completely new, yet essential aspect of superbeam, beta-beam and neutrino factory
NO PERFORMANCE EVALUATION SHOULD BE TAKEN SERIOUSLY UNTIL THE NEAR DETECTOR CONCEPTS HAVE BEEN LAYED DOWN!
Soler, Sanchez tomorrow
Annecy 29 septembre 2006 Alain Blondel
near detector constraints for CP violation
= ACP sin2 solar term…
sinsin (m212 L/4E) sin sin P(e) - P(e)
P(e) + P(e)
Near detector gives e diff. cross-section*detection-eff *flux and ibid for bkg
BUT: need to know and diff. cross-section* detection-eff
with small (relative) systematic errors.
knowledge of cross-sections (relative to each-other) required knowledge of flux!
interchange role of e and for superbeam
ex. beta-beam or nufact:
Annecy 29 septembre 2006 Alain Blondel
Bsig
need to know this:
sig
e
sig
sig
e
sig
/
/
experimental signal= signal cross-section X efficiency of selection + Background
and of course the fluxes… but the product flux*sig is measured in the near detector
this is not a totally trivial quantity as there is somethig particular in each of these cross-sections:
for instance the effects of muon mass as well as nuclear effects are different for neutrinos and anti-neutrinos
while e.g. pion threshold is different for muon and electron neutrinos
Annecy 29 septembre 2006 Alain Blondel
3.5 GeV SPL
beam
-- low proton energy: no Kaons e background is low--region below pion threshold (low bkg from pions)
but:low event rate and uncertainties on cross-sections
Annecy 29 septembre 2006 Alain Blondel
Uncertainties in the double ratio (Sobczyk at RAL meeting)
1. problem comes from compound of Fermi motion and binding energy with the muon mass effect.
)(
)(,
)(
)(
ee
RR
the double ratio calculation is very insensitive to variations of parameters … but
Annecy 29 septembre 2006 Alain Blondel
)(
)()(
)(
e
eDR
at 250 MeV (first maximum in Frejus expt) prediction varies from 0.88 to 0.94according to nuclear model used. (= +- 0.03?)
Hope to improve results with e.g. monochromatic k-capture beam
Annecy 29 septembre 2006 Alain BlondelAlain Blondel
NUFACT Accelerator baseline
40Beam durationb) (s)
2 ± 1Bunch length, rms (ns)
3,5a)No. of bunch trains
50Repetition rate (Hz)
4Beam power (MW)
10 ± 5Energy (GeV)
Valueproton driver parameter
a)Values ranging f rom 1–5 possibly acceptable.b)Maximum spill duration for liquid-metal target.
1021/ yr0.1/
decays per straightangular divergence
100ns trains of both signs of muons separated by
20GeV40GeV
stored muon energy upgradable to
yescooling
phase rotation and bunching
L Hg+ 20T solenoid
Target+capture
ValueNuFact Parameter
RF
Goals achieved!
NEUTRINO FACTORY -- paradoxically quite mature option. ISS (International Scoping Study) revisited accelerator and detector options in 2005-2006.
Annecy 29 septembre 2006 Alain BlondelAlain Blondel
NUFACT detectors baseline
large magnetic volume for Larg, TASD, MECC(Platinum)
beyond the baseline to be investigated
~3 GeV (actual analysis to be used)
eff ective muon selection threshold
+- 2%matter density uncertainty
MI ND (30 kton) detector 2
largest MI ND1)(100 kton) 4cmFe/1cm scint res. 1cm+ ECC 5- 10 kton (Silver)
detector 1
7500 +- 500 kmdistance detector 2
3000+- 1000 kmdistance detector 1
1) Magnetized I ron Neutrino Detector
MI ND
new selection
old selection
LAr
GAr
B? 0.11 T
LHe
HTc coil?HTc coil? hep-ph/0510131Frascati, 2005
Annecy 29 septembre 2006 Alain Blondel
Overall comparisons from ISS (nearly final plots)
sign m2
CP phase
NuFACT does it all…(+ univ. test etc…)but when can it do it and at what cost?
Annecy 29 septembre 2006 Alain BlondelAlain Blondel
EP2010:
« pursue an internationally coordinated, staged program in neutrino physics »
CERN-SG: Studies of the scientific case for future neutrino facilities and the R&D into associated technologies are required tobe in a position to define the optimal neutrino programme
based on the information available in around 2012;Council will play an active role in promoting a coordinated Europeanparticipation in a global neutrino programme.
Towards a high-intensity neutrino programme
Annecy 29 septembre 2006 Alain Blondel
Conclusions
CERN priority to LHC makes it unlikely to raise a new neutrino programme until at least 2016. However opportunities are open by the upgrades of the LHC acclerator complex
-- upgrade of CNGS … tempting and politically attractive. but is it feasible? worth it given the time scales?
-- SPL would offer a powerful low energy beam-- beta-beam offers extremely clean e beam new ideas to improve flux/energy/cost…. -- baseline detector for sub-GeV neutrinos is WaterCherenkov-- in few GeV range, Larg, TASD etc… competitive -- near detector and monitoring systems should not be forgotten
Annecy 29 septembre 2006 Alain Blondel
-- neutrino factory still the ultimate contender, especially if 13
is very small. Requires magnetic detectors
-- design studies of Superbeam/betabeams/ NuFact and of the associated detector systems will be necessary for a choicearound 2010/2012; organization ongoing.
Conclusions (ctd)