MCTF
Steve Geer SLAC/LBNL November, 2009 1
Muon Colliders
MCTF
Physics Landscape
Steve Geer SLAC/LBNL November, 2009 2
MCTF
Decision Tree
Steve Geer SLAC/LBNL November, 2009 3
Pierre Oddone
0.5 TeV e+e-
3 TeV e+e-
3-4 TeV +-
MCTF
Muon Collider Motivation
If we can build a multi-TeV muon collider it’s an attractive option because muons don’t radiate as readily as electrons (m / me ~ 207):
- COMPACTFits on laboratory site
- MULTI-PASS ACCELERATIONCost Effective (e.g. 10 passes → factor 10 less linac)
- MULTIPASS COLLISIONS IN A RING (~1000 turns)Relaxed emittance requirements & hence tolerances
- NARROW ENERGY SPREADPrecision scans
- TWO DETECTORS (2 IPs)-Tbunch ~ 10 s … (e.g. 4 TeV collider)
Lots of time for readoutBackgrounds don’t pile up
- (m/me)2 = ~40000Enhanced s-channel rates for Higgs-like particles
Steve Geer SLAC/LBNL November, 2009 4
CO
ST
PH
YS
ICS
MCTF
Muon Colliders are Compact
Steve Geer SLAC/LBNL November, 2009 5
4 TeV
0.5 TeV3 TeV
MCTF
Narrow Energy Spread
Beamstrahlung in
any e+e- collider
E/E 2
Steve Geer SLAC/LBNL November, 2009 6
Shiltsev
MCTF
Challenges
● Muons are born within a large phase space ( → )
- To obtain luminosities O(1034) cm-2s-1, need to reduce initial phase space by O(106)
● Muons Decay (0 = 2s) - Everything must be done fast
→ need ionization cooling- Must deal with decay electrons- Above ~3 TeV, must be careful about
decay neutrinos !
Steve Geer SLAC/LBNL November, 2009 7
MCTF
Muon Collider Schematic
Steve Geer SLAC/LBNL November, 2009 8
Proton source: Upgraded PROJECT X (4 MW, 2±1 ns long bunches)
1021 muons per year that fit within the acceptance of an accelerator
√s = 3 TeV Circumference = 4.5kmL = 3×1034 cm-2s-1 /bunch = 2x1012
(p)/p = 0.1%* = 5mmRep Rate = 12Hz
MCTF
Target Facility Design
Steve Geer SLAC/LBNL November, 2009 9
• A 4MW target station design study was part of “Neutrino Factory Study 1” in 2000 ORNL/TM2001/124
• Facility studied was 49m long = target hall & decay channel, shielding, solenoids, remote handling & target systems.
• Target: liquid Hg jet inside 20T solenoid, identified as one of the main Neutrino Factory challenges requiring proof-of-principle demonstration.
• Beam dump = liquid Hg pool. Some design studies started.
4MW Target Station Design
V. Graves, ORNL
T. Davonne, RAL
Proton Hg Beam Dump
MCTF
MERcury Intense Target Experiment (MERIT)
Steve Geer SLAC/LBNL November, 2009 10
• Proof-of-principle demonstration of a liquid Hg jet target in high-field solenoid ran at CERN PS in Fall 2007.
• Successfully demonstrated a 20m/s liquid Hg jet injected into a 15T solenoid, & hit with a suitably intense beam (115 KJ / pulse !).
• Results suggest this technology OK for beam powers up to 8MW with rep. rate of 70Hz !
Hg jet in a 15T solenoidMeasured disruption length
= 28 cm
1 cm
MCTF
Front-End Specifications
Steve Geer SLAC/LBNL November, 2009 11
Parameter Drift Buncher Rotator Cooler
Length (m) 56.4 31.5 36 75
Focusing (T)
2 2 2 2.5 (ASOL)
RF f (MHz) 360 240 240 202 201.25
RF G (MV/m)
0 15 15 16
Total RF (V) 126 360 800
± →
/p within reference acceptance = 0.085 at end of cooler→ 1.5 1021 μ/year
MCTF
Front-End Simulation Results
Steve Geer SLAC/LBNL November, 2009 12
Neuffer
MCTF
Ionization Cooling
Steve Geer SLAC/LBNL November, 2009 13
Must cool fast (before muons decay)
Muons lose energy by in material (dE/dx). Re-accelerate in longitudinal direction reduce transverse phase space (emittance). Coulomb scattering heats beam low Z absorber. Hydrogen is best, but LiH also OK for the early part of the cooling channel.
Cooling
XmEEds
dE
dsd NN
03
2
2 2
)GeV 014.0(1
Heating
MCTF
MuCool
Steve Geer SLAC/LBNL November, 2009 14
Developing & bench testingcooling channel components
MuCool Test Area at end of FNAL linac is a unique facility:
-Liquid H2 handling-RF power at 805 MHz-RF power at 201 MHz-5T solenoid (805 MHz fits in bore)-Beam from linac (soon)
Liq. H2 absorber
MTA42cm Be RF window
New beamline
MCTF
RF in Magnetic Field: 805 MHz Results
Steve Geer SLAC/LBNL November, 2009 15
Peak Magnetic Field in T at the Window
>2X Reduction @ required field
Data reproducible & seem to follow universal curve.
Possible solutions:-special surfaces (e.g. beryllium)-Surface treatment (e.g. ALD)- Magnetic insulation
Effect is not seen in cavities filled with high pressure hydrogen gas (Johnson & Derbenev) – possible solution (but needs to be tested in a beam – coming soon)
When vac. copper cavities operate in multi Tesla co-axial mag. field, the maximum operating gradient is reduced.
MCTF
MICE
Steve Geer SLAC/LBNL November, 2009 16
Ionization Cooling
Instrumentation
Instrumentation
GOALS: Build a section of cooling channel capable of giving the desired performance for a Neutrino Factory & test in a muon beam. Measure performance in various modes of operation.
Multi-stage expt.
First stage being installed at purpose-built muon beam at RAL (first beam to hall March 2008).
10% cooling measured to ±1%. Anticipate completed ~2011/12
Beam Line Complete First Beam 3/08 Running now
PID Installed CKOV TOF EM Cal
First Spectrometer Spring 2010
Spectrometer Solenoid being assembled
MCTF
6D Cooling
Steve Geer SLAC/LBNL November, 2009 17
RF
liquid H
2solenoid
MC designs require the muon beam to be cooled by ~ O(106) in 6D
Ionization cooling reduces transverse (4D) phase space.
To also cool longitudinal phase space (6D) must mix degrees of freedom as the cooling proceeds
This can be accomplished with solenoid coils arranged in a helix, or with solenoid coils tilted.
Palmer
Alexhin & Fernow
MCTF
6D Cooling Channel Scheme
Steve Geer SLAC/LBNL November, 2009 18
Palmer
MCTF
6D Cooling Channel Development
Steve Geer SLAC/LBNL November, 2009 19
Detailed Simulations for candidate 6D cooling schemes
Magnet develop-ment for 6D cooling channels
HCC magnet4 coil test
REQUIRES BEYOND STATE OF ART TECHNOLOGY → Ongoing R&D
FOFO Snake - AlexhinHelical Cooling Channel- Muons Inc.
MCTF
Final Cooling
Steve Geer SLAC/LBNL November, 2009 20
When the emittance is very small, to continue cooling we need very high field solenoids (to continue winning against scattering)
For fields up to ~50T, the final luminosity is ~ prop-ortional to the solenoid field at the end of the channel.
For higher fields we no longer expect to continue to win (limited by beam-beam tune shift).
MCTF
The Promise of HTS
Steve Geer SLAC/LBNL November, 2009 21
10
100
1000
10000
0 5 10 15 20 25 30 35 40 45 Applied Field (T)
JE (
A/m
m²)
YBCO Insert Tape (B|| Tape Plane) YBCO Insert Tape (B Tape Plane) MgB2 19Fil 24% Fill (HyperTech) 2212 OI-ST 28% Ceramic Filaments NbTi LHC Production 38%SC (4.2 K) Nb3Sn RRP Internal Sn (OI-ST) Nb3Sn High Sn Bronze Cu:Non-Cu 0.3
YBCO B|| Tape Plane
YYBBCCOO BB TTaappee PPllaannee
2212
RRRRPP NNbb33SSnn
BBrroonnzzee NNbb33SSnn MgB2
NNbb--TTii SSuuppeerrPPoowweerr ttaappee uusseedd iinn rreeccoorrdd bbrreeaakkiinngg NNHHMMFFLL iinnsseerrtt ccooiill 22000077
1188++11 MMggBB22//NNbb//CCuu//MMoonneell CCoouurrtteessyy MM.. TToommssiicc,, 22000077
427 filament strand with Ag alloy outer sheath tested at NHMFL
Maximal JE for entire LHC Nb Ti strand production (CERN-T. Boutboul '07)
CCoommpplliieedd ffrroomm AASSCC''0022 aanndd IICCMMCC''0033 ppaappeerrss ((JJ.. PPaarrrreellll OOII--SSTT))
44554433 ffiillaammeenntt HHiigghh SSnn BBrroonnzzee--1166wwtt..%%SSnn--
00..33wwtt%%TTii ((MMiiyyaazzaakkii--MMTT1188--IIEEEEEE’’0044))
MCTF
HTS Solenoid R&D
Steve Geer SLAC/LBNL November, 2009 22
NHMFL test coil LBL Test Coil
FNAL test cable. Test degradation of Jc in the cabling process
MCTF
Acceleration
Steve Geer SLAC/LBNL November, 2009 23
0.1 1 2 5 10 20 50
AVERAGE GRADIENT (MV/m)
MU
ON
SU
RV
IVA
L F
RA
CT
ION
Bogacz
0.2
0.4
0.6
0.8
1.0
Accelerating muons from 3 GeV to 2 TeV
Example: TESLA cavities: Real estate gradient ~31 MV/m → 97% survival
● Early Acceleration(to 25 GeV ?) couldbe the same as NF.Needs study.
● Main Acceleration –
Attractive Candidates- RLAs (extension of NF accel. scheme ?) - Rapid cycling synchrotron – needs magnet R&D- Fast ramping RLA
● Options needstudy → particletracking, collectiveeffects, cavityloading, ...
MCTF
Collider Ring
• Muons circulate for ~1000 turns in the ring
• Need high field dipoles operating in decay back-grounds → R&D
• First lattice designs exist
New ideas → conceptual designs for various options
Comparison of different schemes, choice of the baseline
Detailed lattice design with tuning and correction “knobs”
Dynamic aperture studies with magnet nonlinearities, misalignments and their correction
Transient beam-beam effect compensation
Coherent instabilities analysis
WE ARE HERE
DE
SIG
N P
RO
CE
SS
Steve Geer SLAC/LBNL November, 2009 24
MCTF
Neutrino Radiation
Steve Geer SLAC/LBNL November, 2009 25
With L ~ E2 →
OK at √s = 1 TeV OK at √s = 3 TeV if D = 200m Above 3 TeV need to pay attention (wobble
beam, lower *, higher Bring , … )
MCTF
Background from Muon Decay
Steve Geer SLAC/LBNL November, 2009 26
As the decay electrons respond to the fields of the final focus system they lose 20% of their energy by radiating on average 500 synchrotron photons with a mean energy of ~500 MeV … & are then swept out of the beampipe.
2 x 1012 muons/bunch 2 x 105 decays/m Electron decay angles O(10) rad Mean electron energy = 700 GeV
Mean energy= 700 GeV
2 2 TeV Collider
Electron Energy (GeV)
0 500 1000 1500 2000
Num
ber
of D
ecay
s
- → e- e
MCTF
Detector Backgrounds
Steve Geer SLAC/LBNL November, 2009 27
Muon Collider detector backgrounds were studied actively ~10 years ago (1996-1997). The most detailed work was done for a 22 TeV Collider → s=4 TeV.
Since muons decay (2TeV=42ms), there is a large background from the decay electrons which must be shielded.
The electron decay angles are O(10) microradians – they typically stay inside the beampipe for about 6m. Hence decay electrons born within a few meters of the IP are benign.
Shielding strategy: sweep the electrons born further than ~6m from the IP into ~6m of shielding.
MCTF
Background Simulations
Steve Geer SLAC/LBNL November, 2009 28
• Shielding design group & final focus design group worked closely together & iterated
• Used two simulation codes (MARS & GEANT), which gave consistent results
• Shielding design & simulation work done by two experts (Mokhov & Stumer) in great detail, & involved several person-years of effort.
MCTF
Final Focus Setup
Steve Geer SLAC/LBNL November, 2009 29
Fate of electrons born in the 130m long straight section: 62% interactupstream of shielding, 30% interact in early part of shielding, 2% interact in last part, 10% pass through IP without interacting.
MCTF
IP Region
Steve Geer SLAC/LBNL November, 2009 30
MCTF
More Shielding Details
Steve Geer SLAC/LBNL November, 2009 31
r=4cm
Z=4m
Designed so that, viewed from the IP, the inner shielding surfaces are not directly visible.
MCTF
4 TeV Collider Backgrounds
Steve Geer SLAC/LBNL November, 2009 32
Background calculations & shielding optimization was performed using(independently) MARS & GEANT codes … the two calculations were inbroad agreement with each other (although the designs were different in detail).
Results from Summer 1996
GEANT MARSI. Stumer N. Mokhov
MCTF
4 TeV Collider Backgrounds
Steve Geer SLAC/LBNL November, 2009 33
r (cm) n p e 5 2700 120 0.05 0.9 2.3 1.7
10 750 110 0.20 0.4 0.7
15 350 100 0.13 0.4 0.4
20 210 100 0.13 0.3 0.1
50 70 120 0.08 0.05 0.02
100 31 50 0.04 0.003 0.008
calo 0.003
muon 0.0003
GEANT (I. Stumer) Results from LBL Workshop, Spring 1997
Particles/cm2 from one bunch with 2 1012 muons (2 TeV)
MCTF
Occupancies in 300x300 m2 Pixels
Steve Geer SLAC/LBNL November, 2009 34
TOTAL CHARGED
MCTF
Vertex Detector Hit Density
Steve Geer SLAC/LBNL November, 2009 35
Consider a layer of Silicon at a radius of 10 cm: GEANT Results (I. Stumer) for radial particle fuxes per crossing:
750 photons/cm2 2.3 hits/cm2
110 neutrons/cm2 0.1 hits/cm2
1.3 charged tracks/cm2 1.3 hits/cm2
TOTAL 3.7 hits/cm2
0.4% occupancy in 300x300 m2 pixels
MARS predictions for radiation dose at 10 cm for a 2x2 TeV Collider comparable to at LHC with L=1034 cm-2s-1
At 5cm radius: 13.2 hits/cm2 1.3% occupancy
For comparison with CLIC (later) … at r = 3cm hit density about ×2 higher than at 5cm → ~20 hits/cm2 → 0.2 hits/mm2
MCTF
Pixel Micro-Telescope Idea
Steve Geer SLAC/LBNL November, 2009 36
S. Geer, J. Chapman: FERMILAB-Conf-96-375
Photon & neutron fluxes in inner tracker large but
mean energies O(MeV) & radial fluxes ~ longitudinal fluxes ( isotropic)
Clock 2 layers out at variable clock speed (tomaintain pointing) &take coincidence.
Blind to soft photon hits& tracks that don’t come
from IP
MCTF
Pixel Micro-Telescope Simulation - 1
Steve Geer SLAC/LBNL November, 2009 37
MCTF
Pixel Micro-Telescope Simulation - 2
Steve Geer SLAC/LBNL November, 2009 38
MCTF
TPC
Steve Geer SLAC/LBNL November, 2009 39
Exploit 10s between crossings
Large neutron flux – gas must not contain hydrogen: 90% Ne + 10% CF4
Vdrift = 9.4 cm/s with E = 1500 V/cm. Ion buildup E/E = 0.7%
Cut on pulse height removes photon & neutron induced recoils
V. Tchernatine
MCTF
Calorimeter Backgrounds
Steve Geer SLAC/LBNL November, 2009 40
Electromagnetic: Consider calorimeter at r=120 cm, 25 r.l. deep, 4m long,22 cm2 cells:
GEANT 400 photons/crossing with <E> ~1 MeV <ETower>~400 MeV
E ~ (2<n>) <E> = 30 MeV
For a shower occupying 4 towers: <E> = 1.6 GeV and E = 60 MeV
Hadronic: Consider calorimeter at r=150 cm, 2.5m deep (~10), covering30-150 degrees, 55 cm2 cells:
<ETower> ~ 400 MeV
E ~ (2<n>) <E> = O(100 MeV)
These estimates were made summer 1996, before further improvements tofinal focus + shielding reduced backgrounds by an order of magnitude … so guess background fluctuations reduced by 3 compared with above.
MCTF
Bethe-Heitler Muons (Z Z+-)
Special concern: hard interactions (catastrophic brem.) of energetic muons travelling ~parallel to the beam, produced by BH pair production.
Believe that this back-ground can be mitigated using arrival-times, pushing calorimeter to larger radius, & spike removal by pattern recognition … but this needs to be simulated
Steve Geer SLAC/LBNL November, 2009 41
MCTF
Comparison with CLIC
Steve Geer SLAC/LBNL November, 2009 42
• We are not yet in a position to make an apples-to-apples comparison with CLIC, but …..
hits/mm2/bunch train
30mm O(1) hit/mm2/bunch train
FROM CLIC Machine-Detector interface studies:
CLICNOT AN APPLES-to-APPLES COMPARISON … BUT … Background hit densities appear to be similar to MC … so there may be many detector design issues in common between the 2 machines
Note: CLIC shielding cone= 7o c.f. 20o for MC (but we hope to improve on this)
MCTF
MC R&D – The Next Step
• In the last few years MC-specific R&D has been pursued in the U.S. by Neutrino Factory & Muon Collider Collaboration (NFMCC) & Muon Collider Task Force (MCTF)
• Last December the NFMCC+MCTF community submitted to DOE a proposal for the next 5 years of R&D, requesting a greatly enhanced activity, aimed at proving MC feasibility on a timescale relevant for future decisions about multi-TeV lepton colliders.
Steve Geer SLAC/LBNL November, 2009 43
MCTF
NFMCC/MCTF Joint 5-Year Plan
● Deliverables in ~5 years:-Muon Collider Design Feasibility Report- Hardware R&D results → technology choice- Cost estimate- Also contributions to the IDS-NF RDR
● Will address key R&D issues, including- Maximum RF gradients in magnetic field- Magnet designs for cooling, acceltn, collider- 6D cooling section prototype & bench test- Full start-to-end simulations based on technologies in hand, or achievable with a specified R&D program
● Funding increase needed to ~20M$/yr(about 3x present level); total cost 90M$
Steve Geer SLAC/LBNL November, 2009 44
MCTF
R&D – Ongoing
NFMCC/MCTF HISTORY & FUTURE
PROPOSAL
Steve Geer SLAC/LBNL November, 2009 45
MCTF
Anticipated Progress
Key component m
odels
NOW5 YEARS
Steve Geer SLAC/LBNL November, 2009 46
MCTF
Aspirational Bigger Picture
Steve Geer SLAC/LBNL November, 2009 47
MCTF
Muon Collider R&D: A National Program
Steve Geer SLAC/LBNL November, 2009 48
MCTF
Final Remarks
• Steady progress on the Front-End develop-ment for Muon Colliders
- Cooling channel design concepts- NF R&D (IDS-NF, MERIT, MICE, … )
• The time has come to ramp up the Muon Collider specific R&D → a National Program
• There are many challenges to be overcome
- RF in magnetic fields & 6D Cooling Channel
- Cost effective acceleration scheme- Collider Ring- Detector/Backgrounds optimization
• The incentive to meet these challenges is great → “5 Year Plan” → Design Feasibility Study
Steve Geer SLAC/LBNL November, 2009 49
MCTF
3
4
4 GeVNF
25 GeVNF
Illustrative Staging Scenario
4MW multi-GeVProton Source
Accumulation &Rebunching
Steve Geer CERN Neutrino Workshop October 1-3, 2009 50
MCTF
Muon Collider Parameters
Steve Geer SLAC/LBNL November, 2009 51