Transcript Acceleration Scenarios for the Muon Collider

FNAL Feasibility Study on a Neutrino
Source Based on a Muon Storage Ring
Fardog Summary, Jan. 6th ‘00
Norbert Holtkamp
• Results from recent technical discussions
• some nice technical solutions
• some results with impact on the physics
study
• Status of the study
• Timing
The Task
• A design concept for a muon storage ring and associated
support facilities that could, with reasonable assurance,
meet performance goals required to support a compelling
neutrino based research program.
• 2.Identification of the likely cost drivers within such a
facility.
• 3.Identification of an R&D program that would be
required to address key areas of technological uncertainty
and cost/performance optimization within this design, and
that would, upon successful completion, allow one to
move with confidence into the conceptual design stage of
such a facility.
• 4.Identification of any specific environmental, safety, and
health issues that will require our attention.
Choice has been made !
Parameters for the Neutrino Source
- Energy of the ring
- Number of neutrinos /
straight
- no polarization
GeV
50
2x1020/y
- capability to switch
between m m
- FERMI to SLAC / LBNL
• Basic Calculation
– 1/3 of the muons decay in the straight section (38 %)
– 10 protons for 1 m into the storage ring (>10; >20-50)
– 2x107 sec
• 2x1013 proton on target per pulse @ 16 GeV and 15 Hz
– 3x1013 proton because of carbon target
• 2x1012 m per pulse to be accelerated and injected into the
ring
– cooling channel not successfully simulated < 2x1012
• longer bunch in the proton driver and on target (1 nsec  3)
– helps
• ring tilt angle is 13deg ( 22 %) instead of 35 deg (57 %)
– ring with these params. Is not a cost driver at all
The Neutrino Source
• Approach:
– go more conventional where ever possible
– Oak Ridge, FHML, Brookhaven  the target
• most people bought the solid target
– Jefferson Lab / Cornell  sc rf and re-circulating linacs
• biggest disconnect right now
– LBNL , DUBNA  induction linacs (talk tomorrow)
• goes much better than expected but not cheap
– IHEP Protvino  sc solenoid channels
• so far very good job
– specific design and engineering (cooling channel, target
collection, beam manipulation, beam tracking and
simulation)  Muon Collider group (12 people @FNAL)
+ the collaboration
• ( thank Andy for the enormous support)
– general engineering (large scale rf systems, sc magnets, sc
solenoid channels, ps, vacuum, beam lines, tunnel, water)
(20 FTE for 6 month)
• very good support from FESS and TD
• not easy to convince somebody in BD that anything else than
TEVATRON is useful at all.
The Neutrino Source
• First experiment based on an intense muon
source -> does it have to be 50 GeV ??
– 10 GeV and 50 kT or more magnetized water detector: Goal:
Balance detector cost with Accelerator: E*kT*I=const.
– Start with 2x1019/year (Sessler, Geer) and still good physics
Parameters for the Muon Storage Ring
Energy
GeV
50
decay ratio
%
>40
inv. Emittance
m*rad
0.0032
m
160
 in straight
12
10
6
Nm/pulse
mrad
2.0
typical decay angle of m
mrad
0.2
Beam angle ((
m
3x105
Lifetime c*

Neutrino Source Study @ FERMI
•
•
•
Application of a “Generic Neutrino Source” to specific site
Base the study on specific set of Parameters
6 month period of time to define the R&D program and develop a
layout to investigate the scope of such a complex
Generic Layout
collaboration
paper
“deviate wherever
necessary or useful”
Physics Study in parallel
H. Schellmann / S. Geer
Footprint for a 50 GeV Neutrino Source
• Infrastructure is very close together ...
R & D Issues for the
Proton Driver Design Study
• R & D groups:
• Goal:
–4 x 0.75x 10-13 = 3x 10-13 @ 15 Hz
–8 GeV versus 16 GeV versus higher energies ?
•Very hard to get people going on this: Only Chuck A.
•Will probably not appear in the report in detail
• Why?
–Power bill is dominated acceleration
–at low Frequ. Tfill is large compared to T pulse
–most of the time power source is on for filling
– -> higher rep rate less efficient; peak current is limit too.
–Go to higher energies (more acceleration): Higher Proton
beam energy and smaller rep rate is more efficient way to
produce beam power
What changes compared to MC
• The target
– Ptarget is still of the order of 1.5 (graphite) with upgrade
possibility to 4 MW (probab. Liquid)
– RF after target only gives increase in Polarization from
28% to 40% -> it is not worth it !! A. Blondel from CERN
– Target dissipation is only 30 kW or so -> radiation cooled
– Radiation damage is a major constraint for sc coil. 0.5 - 1
year is most probable lifetime so far
– 20 T can only be achieved with sc +nc coil.
– Disagreement between people on the fatigue limits
calculated and the actually NuMI test.
30 - 60 MHz rf
~ 5 MV/m
Target for a Neutrino Factory
• 1 - 2 MW target
• Reduce power in the target -> low Z
• Solid Target: loose x 1.5 in yield -> more protons ->
ok at this intensity
• Magnet radius is too small, especially with Copper
inside
Induction Linacs and Long
Solenoidal Channels
50 m drift before f rotation
For carbon target:
0.10 m/p between 225 - 240 MeV
0.13 m/p between 220 - 250 MeV
0.18 m/p between 200 - 270 MeV
Trade off:
•Energy Spread after rotation drift channel length
•Particle capturelength(voltage) in induction linac
[loss]
[loss]
 simplest solution offers comparable yield
Induction Linac Layout
• Strong Effort at LBL for DAHRT
• A little bit of expertise at Fermi
– higher field 2-3 T and smaller cores may be better solution
– saturation in the cores is under control
– switching is the main problem
Small effort at
FNAL: “old”
expert
from DUBNA no
working in TD.
Mainly on
solenoid
channel design
LBNL Status
• Induction Cell
– basically ok. Trade off between: operational and
investment cost
• Pulser system:
– still not so clear: 4 pulse per burst…
– asymmetric voltage
–
The Heart of the Cooling Channel for a
Neutrino Factory
• IIT, BNL, LBNL, FNAL: go through an engineering
design faster
• Goal: Do all the cooling with one set of hardware
RF may be, solenoids no
• Analytical (Courant Snyder type description of the
motion in Solenoidal channel) LBL
Bz ~ 5 T max
Eacc ~ 15 MV/m @ 200 MHz
The Cooling Channel
• +/- 3.5 T up to 5T or more Lattice with 15 MV/m 175 MHz rf
• period gets shorter and shorter
Reduces to about 100-140 m of cooling channel
200 MHz Cavity + Power Source
• Engineering Layout required for the study
• Want to build and test it, once study is done
1/8th of the full
accelerating cell
~ 0.65 m
Enhance the E Field
on Axis by using a grid
Goal: 15 MV/m
nc cavities
~ 0.65 m
Acceleration based on RLA
• Why not nc?
– Peak power limited already
– in normal conducting cavity: too much power required to
build up the gradient
– gradient is not a free parameter for optimization:
• muon decay +longitudinal acceptance
• SC structures at 200 MHz (100 MHz) and
10 MV/m (7,5 MV/m real estate)
– almost no power to build up gradient --> beam
– loaded Q’s are similar to nc structures ->fill time short but
coupler ?
-> comparatively efficient !
– 4 x 30 bunches per RLA, 2.5x1012 total
– Plinac ~ 6 MW, PRLA1 ~ 2 MW, PRLA2 ~ 12 MW
– ~20 MW power for RF acceleration at ~25 % overall
efficiency (AC  RF)
– Problem is the required peak power not average power !
RLA
Energy
Ltot
m
Llinac
m
Ibeam Rloaded
mA M/m
2-10
10-50
800
2400
2x135
2x700
120
40
83
252
Qloaded
80 x103
240 x103
Nr of
Tpulse =
turns Tfill+Tacc /ms
4
4
132+10
400+32
Peak
Power
MW/m
1.2
0.4
Optimization of the Storage Ring
• The cheapest way to produce muons in the straight
section is to make them as long as possible !
h=
Nr of m decaying in straight section
Nr of m injected
L=length of straight
h=
1
2 (1+pr /L)
=
1
2 (1+0.2)
B / Tesla
• Problem with dynamic aperture due to short
Magnets and large aperture (10x15 cm) -> not fixed
yet
The SC Large Bore Magnets
• Low field quality helps reduce price although large aperture
• main heat load due to m-decay, probably not optimum
• 1 cm tungsten liner instead of 3 cm
Large Bore Magnets
• Similar magnets for the RLA Arcs with less
tungsten shielding
• Bigger magnets in the Beam Spreaders of the RLA
arcs + combined function
• Energy Acceptance of the Arcs ??????
Storage Ring Arc Cell
The Storage Ring Location@ FNAL
Storage Ring Layout
• Site layout for the Storage Ring and the Arcs
• Experimental beam lines and halls
• Cryo space requirement
Goal & Schedule
• 6 Month study:  “10 pages of paper per
subsystem+ 1 schedule + 1 cost”
– Internal Review Feb. 15th and 16th for the Accelerator
part, Feb. 17th and 18th for Geer/Shellman to align the
different contributions
– Documents in by beg. of march
– Report out by March 30th if that’s acceptable
• Most risky
– induction linac (no, only expensive ->R&D)
– cooling channel design and performance
• what is the minimum emittance achievable
• it is still the most unreliable item
– acceleration: ( largest cost driver)
• Jefferson Lab  Cornell  CERN  SLAC  Fermi
• after Feb. 15th & 16th have working meeting with these
groups to go through: Cryo, cavities, couplers and power
sources
• not sop clear how to organize it
Cost
• Hot Topic: Preliminary result.
Cost Total for each Sub-System
Sub-systems
Storage Ring
RLA2
RLA 1
Capture Linac
Cooling Channel
Adiabatic Capture
Mini Cooling
Induction Linac
Decay Channel
Target Systems
Proton Driver
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
30.0%
35.0%
percent of total
Civil
Systems
ES&H
Utilities
Cryo
Diagn.
PS
Vacuum
RF Cav
RF Source
Magnets
0.0%
10.0%
20.0%
percent of total
30.0%
Questions
• What comes after the study ?
– Review the results
– Does the lab want to do this ?
– If done or obvious: Define the R & D program
• remember: 10 different subsystem and all need strong R&D
program … a lot of people, time and money.
– Develop efficient and cost effective accelerating systems
at low frequency
– Prepare a site where this experimental program can be
pursued
– leading role for Muon Cooling
Directors office