DAEALUS
DAEALUS
Jose Alonso, Luciano CalabrettaMIT INFN-LNS
For the DAEALUS Collaboration
High Current H2+ Cyclotrons for Neutrino Physics:
The IsoDAR and DAEALUS Projects
DAEALUSTRIUMF, Sept 13, 2012 2
Outline• Neutrino oscillations and CP violation are very hot topics!
– Background– Decay-at-rest (DAR) program
• The DAEALUS experiment– Requirements
• The IsoDAR experiment– Requirements– Siting options
• Cyclotron designs and calculations– Luciano Calabretta
• Status and Developments
DAEALUSTRIUMF, Sept 13, 2012 3
Background…
• Neutrino representations in 2 sets of eigenstates:– Flavor (ne, nm, nt) related to production reaction– Mass (n1, n2, n3)
• Coupling between determined by “mixing angles”• Example (2-state system):
Flavor states
DAEALUSTRIUMF, Sept 13, 2012 4
Background…
• 3-state system coupling is a “simple” extrapolation
Mass eigenstates
Flavoreigenstates
DAEALUSTRIUMF, Sept 13, 2012 5
Background…
• Uij are sin/cos function of mixing angles– All now measured– q13 latest to be measured
• Some elements also contain CP violation terms ei
DAEALUSTRIUMF, Sept 13, 2012 6
OscillationsPosc = sin22q sin2(1.27Dm2L/E)
P = oscillation probability sin22q = mixing strength
Dm2 = mass difference (squared) for mass eigenstates (eV2)
L = baseline length (km)E = neutrino energy (GeV)
Start with a pure ne beam
(for 2-state system)
DAEALUSTRIUMF, Sept 13, 2012 7
OscillationsPosc = sin22q sin2(1.27Dm2L/E)
P = oscillation probability sin22q = mixing strength
Dm2 = mass difference (squared) for mass eigenstates (eV2)
L = baseline length (km)E = neutrino energy (GeV)
(for 2-state system)
For Dm2 ~ 2.4 x 10-3, eV2
E ~ 1 GeV,
DL ~ 1000 km (optimized Long Baseline conditions)
DAEALUSTRIUMF, Sept 13, 2012 8
Oscillations (for 3-state system)
DAEALUSTRIUMF, Sept 13, 2012 9
Oscillations (for 3-state system)
terms depending onmass splittings
terms depending onmixing angles
CP violatingterms
Sign difference between n and n
DAEALUSTRIUMF, Sept 13, 2012 10
Oscillations (for 3-state system)
terms depending onmass splittings
terms depending onmixing angles
CP violatingterms
Sign difference between n and n
DAEALUSTRIUMF, Sept 13, 2012 11
CP ViolationOption 1: Long-baseline configuration
(~1000 km, ~1 GeV neutrinos)Changing between nm and nm (- or +)
Option 2: Decay-at-rest configuration(1-20 km, ~30 MeV neutrinos)Changing baseline length (changing D)
------NOTE: HUGE benefit from combining both measurements in same detector!
DAEALUSTRIUMF, Sept 13, 2012 12
Decay-At-Rest
DAEALUSTRIUMF, Sept 13, 2012 13
DAR Detection• Look for appearance of ne
• Inverse beta decay (IBD) in detector with lots of free protons – Water Cherenkov or Liquid Scintillator
• Unique signature: Delayed coincidence– Huge benefit in discrimination against background
~ 2 ms
DAEALUSTRIUMF, Sept 13, 2012 14
DAEALUSTRIUMF, Sept 13, 2012 15
DAEALUSTRIUMF, Sept 13, 2012 16
Six MW Target Concept
Copper
Graphite(hollow)
Chris Tschalaer, MIT 2011
r0 = 20 cmz0 = 8 mweight = 55 tons
DAEALUSTRIUMF, Sept 13, 2012 17
Accelerator Requirements• Beam on target: ~800 MeV protons• Beam power*:
– 1.5 km site: 0.8 MW average– 8 km site: 1.6 MW average– 20 km site: 4.8 MW average
• Accelerator duty factor ~ 20%– Instantaneous power requirement is x5 average pwr___________
*Based on matching data rates in proposed 200 kTon Water Cherenkov counter, and LBNE beamline from FNAL to Homestake
DAEALUSTRIUMF, Sept 13, 2012 18
Accelerator requirementsin perspective
AVERAGE power PEAK power
DAEALUSTRIUMF, Sept 13, 2012 19
Cyclotrons: Viable technology?
• PSI is current world power leader in this energy range ~ 1 MW average, 590 MeV protons
• Extrapolation to higher power?_________• Problems:
1. Capture of more ions… space-charge at injection2. Clean extraction… max loss 200 W (~10-4)
DAEALUSTRIUMF, Sept 13, 2012 20
Proposed Solution: H2+ ions
• Two protons for every ion (1 emA = 2 pmA)• Perveance of 5 emA H2
+ at 35 keV/amusame as 2 emA of 30 keV protons– Axial injection of 2 emA protons at 30 keV is
well within state of the art• Extract with stripping foil
– Clean turn separation is not needed, only high-acceptance extraction channel
DAEALUSTRIUMF, Sept 13, 2012 21
Concept (Luciano Calabretta)INFN-Catania
DAEALUSTRIUMF, Sept 13, 2012 22
Development Strategy
• With highly-complex projects:– Develop phased approach– Design phases such that good physics can be done
with each
DAEALUSTRIUMF, Sept 13, 2012 23
Development Strategy
• With highly-complex projects:– Develop phased approach– Design phases such that good physics can be done
with each• In our case… Injector cyclotron can in itself
be a “barn-burning” neutrino source– The IsoDAR Project– 10 pmA (CW) at 60 MeV/amu = 600 kW!
DAEALUSTRIUMF, Sept 13, 2012 24
IsoDAR (Isotope Decay At Rest)• Proton beam into neutron-producing target• Secondary neutrons into ~50 kg pure 7Li blanket• 8Li decay produces ne, high beta endpoint energy
– Good for discriminating background • As with DAEALUS, use IBD in liquid scintillator
– Good spatial, energy resolution
GOAL: STERILE NEUTRINO SEARCH
DAEALUSTRIUMF, Sept 13, 2012 25
Sterile Neutrino
• Anomalies in n, n behaviors• LSND, MiniBooNE• Reactor flux anomaly
• Possible explanation:• New neutrinos that do not
interact via weak force butstill can oscillate with otherneutrino states
• Most likely Dm2 ~ 1 eV2
• DL ~ few meters!
DAEALUSTRIUMF, Sept 13, 2012 26
Oscillations can be seen within detector!
DAEALUSTRIUMF, Sept 13, 2012 27
SNO+ also encouraging IsoDAR
DAEALUSTRIUMF, Sept 13, 2012 28
Engineering Challenges for DIC• Deliver components through narrow passage-
ways or down shafts– 4-meter diameter coil will NOT fit!
• KamLAND drive-in access, drift height < 4 meters• Creighton mine shaft (SNO+) is ~3.8 x 1.5 m
– Pole pieces too heavy for hoist• Creighton hoist has 12 ton limit
• Power and cooling • Neutron shielding
DAEALUS
DAEALUS
Luciano CalabrettaINFN-LNS
For the DAEALUS Collaboration
Cyclotron Design and Simulation Studies
DAEALUS30
Extraction trajectory
Coil
Electrostatic deflectors
TRIUMF, Sept 13, 2012
Injector Cyclotron
arXiv: 1207.4895
• 60 MeV/amu, 5 emA H2+
• Normal conducting coils ~ 4 meter coil diameter • Axial injection
(spiral inflector) • Electrostatic
extraction channel
DAEALUS31
Main parameters are extrapolated from
existing commercial compact cyclotron: C30
and TR-30
Injector Cyclotron ParametersEinj 35 keV/n Emax 60 MeV/nRinj 55 mm Rext 1.99 m
<B> at Rinj 0.97 T <B> at Rext 1.16 TSectors N. 4 Cavities N. 4
RF 49.2 MHz Harmonic 6th V-inj > 70 kV V-ext 240 kV
<DE/turn> 1.3 MeV DR at Rext 14 mmDR at Rinj >56 mm Turns N. < 952 Electrostatic Deflector (ED) + Magnetic ChannelsE. D. Gap 14 mm E.D. field 35 kV/cm
Courtesy of R. Doelling
Injection energy and acceleration voltage are higher vs. C30/TR-30
60 MeV/n after 100 turn vs. 150, or 12 msec vs. 8.1 msec
DAEALUS32
Beam size vs. turns number for different beam current
Particle’s distribution at last turn
[m]
1 MeV/n
61.7 MeV/n
See, Jianjun Yang presentation at ECPM 2012
DAEALUS33
Beam emittance 6.4 p mm.mrad, off center + 3mm at orbit of 1 MeV/n
DAEALUS34
0 10 20 30 40 50 600.9
0.95
1
1.05
1.1
1.15
E [MeV/amu]
n r
nr
50 52 54 56 58 60 620.9
0.95
1
1.05
1.1
1.15
E [MeV/amu]
n r
nr
DAEALUS35
See, Jianjun Yang presentation at ECPM 2012
Power lost on the septum of E.D. 120 W @ 600 kW!
DAEALUSCAARI – Aug 6, 2012
DAEALUS Superconducting Ring Cyclotron• H2+ injected in the ring cyclotron at 60MeV/n• Last closed orbit at 800MeV• Isochronous field. Average field ≈2T• Simulation done with OPERA3D:
– Design the structure– Define materials– Design coils– Define mesh– Solve through Tosca (FEM)– Elaborate and visualize results with Post-Processor – Optimization process
• To accelerate H2+ the magnetic field is 2 times higher than
for protons superconducting coils• Losses have to be limited to a few hundred watts in total
DAEALUSCAARI – Aug 6, 2012
Outer Yoke
Pole
hill
Inner Yoke
8 SUPERCONDUCTING sector magnets:
Outer radius ≈7m
Half height 2.8m
Sector Weight
<450tons
Pole gap= 80mm
view of1/16 of the cyclotron: median plane is above the upper hill
DAEALUS Superconducting Ring Cyclotron
DAEALUSCAARI – Aug 6, 2012
DAEALUS –SRC COILS
RIKEN SRC sector
RIKEN
The longest straight arms of the coils will be conncted with a plate
passing between the upper and lower hill. This will be necessary to
cancel the expansion magnetic forces which tend to make the coil
round
On the coil surface Bmodmax≈6T
DAEALUSCAARI – Aug 6, 2012
DAEALUS –SRC COILS
Current density 3400A/cm2, Area 30x24cm2 or 15x48cm2
The winding of the cable could be a blocking point for cooling reasons, Huge forces developped
The different cross section is needed to guarantee enough space in the center region for the RF Cavities installation (and maybe is
not enough!)
Front view Top view
DAEALUSCAARI – Aug 6, 2012
DAEALUS –SRC OPTIMIZATION RESULTS
±0.5% between isochronous and
calculated
All dangerous resonances are crossed
quicklyWalkingshaw resonance NOT crossed at the end
of acceleration
We need 10-4
DAEALUSCAARI – Aug 6, 2012
DAEALUS –SRC LAYOUTBmax on the patch on the median plan 6.05TBmin on a patch on the median plan -2.5T
1 Electrostatic Deflector + 4
Magnetic channels for the injection
DAEALUSCAARI – Aug 6, 2012
DAEALUS –SRC CAVITIESTwo solutions analyzed
Double gap cavity multi sistem cavity like SC
Compact Cyc
Single gap cavity like 590MeV PSI cyc
With the courtesy of L. Piazza INFN-LNL
DAEALUS43
Vertical beam size along the acceleration in the radial range from 4 to 4.9 m, snapshot at 0° azimuth. The left figure has no charge space effects, 0 mA. The right figure is evaluated with a 5 mA beam H2+ current
Simulation made by J. Yang and A. Adelman, using OPAL code
Space charge effects have
negligible effects during acceleration in
the ring cyclotron
DAEALUS
Histogram of 5 mA H2+ beam at the stripper foil position, simulation include space charge effects (OPAL code)
44
The radial size of the beam depend mainly by the energy gain
per turn
DE/E% phase (deg)
Z (mm) R (mm)
DAEALUS45
Effect due crossing Walkinshaw resonance
DAEALUSTRIUMF, Sept 13, 2012 46
Steering /focusing Magnetic Channel
Extraction trajectories energy spread< ±0.6%
Stripper foils
2°proton beam
2° Stripper
DAEALUS47
Stripper foil: 5 MW beam @ 800 MeV Xing a stripper foil 1 mg/cm2 thick release 45 W due to nuclear interaction! The electrons removed by the strippers have a full power of 5 MW*Me/MH2=5/(2*1826)=1370 W !
Electrons are the main source of stripper damageBut, electrons can be stopped before strike the stripper
Stripper Thickness can be also thinner, because:H- =(p+e+e) p, e, e is a two steps process (p+e+e) H + e p, e, eH2+ =(p+p+e)p, p, e is a single step process, lower probability for H2+H + pThe neutral H can be removed by an additional stripper foil, 10 cm later
Stripper foil
emergingprotons
H2+ beam
electronscatcher
If B=0.4 T Re=4.5 mm
DAEALUS
[cm
]injection
yoke
extraction
DAEALUS
BEAM LOSSES
DAEALUS
3 major blocking points in the design
the variable cross-section that could cause
complications on the LHe cooling system
Huge residual radial FORCE that pushed
outward the coil
Not enough SPACE left between two contiguous
sectors to install RF cavities PSI like
DAEALUS
Alternative solutions AWe need an exact symmetry 8 only above 400MeV/amu -> the coil was divided in two zones. Fixing a point on the center of the hill where there is the last closed orbit, the tip of the coil was tilted by 4deg. Same with the contiguous coil ring, which was tilted by -4deg
Zone 2 symmetry 8
Zone 1symmetry 4
cavity
cavity
DAEALUS
Alternative solutions BThe stopping band resonance for a six sector machine is g=3
Why don’t use a 6-sector cyclotron?
No more issue for the RF cavities installation Easier study for injection system
Sector shape similar to Riken
DAEALUSTRIUMF, Sept 13, 2012 53
0 100 200 300 400 500 600 700 8000
0.5
1
1.5
2
2.5
3
nr
nz
E [MeV/amu]
0 100 200 300 400 500 600 700 800-20
020406080
100
770 MeV/n
E [MeV/amu]
RF phase [degre]
DAEALUSTRIUMF, Sept 13, 2012 54
DAEALUSTRIUMF, Sept 13, 2012 55
Preliminary design study made by Joe Minervini Group at MIT-PSFC
DAEALUS
Proposed layout for an ADS plant driven by superconducting LINAC
S. Henderson, Thorium Energy Conference 2011
DAEALUSRochester, 03/04/2012 57
2.5 MW
2.5 MW
5 MW 5 MW
5 MW 5 MW
5 MW 5 MW
5 MW 5 MW
5 MW 5 MW 2.5 MW
2.5 MW
12.5 MW
12.5 MW
12.5 MW
12.5 MW
Layout of a production plant
5 cyclotrons drive 4 ADS
Each ADS is driven by 3 accelerators. Beam stability increases and decreases the number of beam trips
DAEALUS
6.25 MW 6.25 MW
6.25 MW 6.25 MW
6.25 MW 6.25 MW
6.25 MW 6.25 MW
12.5 MW
12.5 MW
12.5 MW
12.5 MW
Cyclotron OFF
In case of failure or beam trips in a cyclotron, it is possible to increase the beam current delivered by each cyclotron to maintain the beam power at 12.5 MW!
DAEALUSTRIUMF, Sept 13, 2012 59
IsoDAR for Radioisotopes?
• Would 10 mA of 60 MeV protons be of interest to the radioisotope community?– Interesting target problems– OK with good beam splitting system?
• Could also be good source of 60 MeV (15 MeV/amu) alphas– Need good source of He++
DAEALUS
DAEALUS
Project Status and Roadmap
DAEALUSTRIUMF, Sept 13, 2012 61
Project Status• Central region tests being set up at BEST
Cyclotrons (Vancouver)
Ion Source
HV Cage70 kV
Wien Filter“Cyclotron”
(Buncher)
Beam Stop
DAEALUSTRIUMF, Sept 13, 2012 62
Goals and Schedule• To benchmark ion source and buncher
performance for H2+ ions
– Input validation for simulation codes• To explore space-charge effects and {x,y}
coupling of space-charge dominated beams• Components being built• Catania shipment in February• Run tests in late spring 2013.
DAEALUSTRIUMF, Sept 13, 2012 63
On Other Fronts…• Near-term funds being sought for further engineering
design for IsoDAR• Longer-range funding for engineering studies of
DAEALUS SRC• Ion source development plans taking shape
– Vibrational states not an issue for IsoDAR– Loosely-bound H2
+ states will Lorentz dissociate in 6T at 800 MeV/amu• Must develop source that quenches these states
• Erice Workshops (12/2011, 11/2012) study design issues with cyclotron experts
DAEALUSTRIUMF, Sept 13, 2012 64
Summary
• H2+ ions can be a key to high-power cyclotrons
for many applications• Compactness and relative cost of cyclotrons
could open doors to several important neutrino experiments
• Exciting times ahead!