mu-e conversion
COMET at J-PARC
Satoshi MIHARA
KEK, Japan
23/Oct/2009
ILC Seminar 2009
Outline
• Introduction
• COMET at J-PARC
– Proton/Muon beam at J-PARC
– Detector
– Sensitivity and Background
– R&D Status
– Schedule and Cost
• Summary
23/Oct/2009
ILC Seminar 2009
INTRODUCTION
23/Oct/2009
ILC Seminar 2009
Introduction
Lepton Flavour Violation of Charged Leptons
LFV diagram in Standard Model
Neutrino Mixing
(confirmed)
m ixing
nm
Soon!
(mn/mW)4
ne
µ
ne
nm
?
W
nt
Very Small (10-52)
LFV diagram in SUSY
m
e
Large top Yukawa coupling
t
~
m
?
m ixin g
~
e
m
e
~
B
Charged Lepton Mixing
(not observed yet)
23/Oct/2009
e
Sensitive to new Physics beyond
the Standard Model
ILC Seminar 2009
What is a m-e Conversion ?
1s state in a muonic atom
nucleus
-
-
m + (A, Z) e + (A,Z)
mmuon decay in orbit
-
-
m  e n n
nuclear muon capture
m + ( A, Z)  n m + ( A, Z - 1)
23/Oct/2009
Neutrino-less muon
nuclear capture
(=m-e conversion)
ILC Seminar 2009
lepton flavors
changes by one unit
G(m - N  e- N)
B( m N  e N) =
G(m - N  nN ' )
-
-
m-e conversion Signal
•
Eme ~ mm-Bm
–
•
Bm: binding energy of the 1s muonic atom
Comparison with meg (and m3e) from the view
point of experimental technique
Background Challenge
megand
m3e
Accidental
Detector performance
resolution, high rate
m-e
conversion
Beam
Beam background
•
•
Improvement of a muon beam is possible, both in
purity (no pions) and in intensity (thanks to muon
collider R&D). A higher beam intensity can be taken
because of no accidentals.
Potential to discriminate different models through
studying the Z dependence
23/Oct/2009
ILC Seminar 2009
R.Kitano, M.Koike, Y.Okada
P.R. D66, 096002(2002)
Comparison between
meg and m-e Conversion (Physics Sensitivity)
• Photonic and non-photonic (SUSY) diagrams
23/Oct/2009
photonic
non-photonic
•
meg
yes (on-shell)
no
•
m-e conversion
yes (off-shell)
yes
ILC Seminar 2009
MEG at PSI Status
•
•
Physics data production started in 2008
Current limit Br(meg)<3.0x10-11 (at 90% C.L.) (MEGA world record 1.2x10-11)
– Probability to obtain this limit given the average expected limit of 1.3x10-11 is 5%
•
Will reach the MEGA limit with 2009 data and go further in 2010-2011
Event Distribution in 90% Box
23/Oct/2009
ILC Seminar 2009
0nbband m-e conversion
• V. Cirigliano et al. PRL 93, 231802 (04)
RPV-SUSY
• R=B(me)/B(meg)
• RPV-SUSY
– R >> 10-2
• LRSM (Left-Right Symmetric Model)
– R~O(1)
LRSM
• Important to measure R to extract
m0nbb from G0ngg
23/Oct/2009
ILC Seminar 2009
cLFV search and LHC
Masiero et al. JHEP03 (2004) 046
23/Oct/2009
ILC Seminar 2009
History of m-e conversion search
Year
Muon source
Target
Upper bound
1952
Cosmic rays
Cu, Sn
4×10−2
1955
Nevis cycl.
Cu
5×10−4
1961
Berkeley synchroc.
Cu
4×10−6
1961-62
CERN synchroc.
Cu
2.2×10−7
1972
Virginia SREL synchroc.
Cu
1.6×10−8
1977
SIN
S
7×10−11
1984
TRIUMF
Pb
Ti
4.9×10−10
4.6×10−12
1992
PSI
Pb
4.6×10−11
1993-97
(only in conference proc.)
Ti
6.1×10−13
Au
7×10−13
2006
23/Oct/2009
ILC Seminar 2009
The SINDRUM-II Experiment (at PSI)
Published Results
SINDRUM-II used a continuous muon
beam from the PSI cyclotron. To
eliminate beam related background
from a beam, a beam veto counter was
placed.
23/Oct/2009
ILC Seminar 2009
The MELC and MECO Proposals
•
MELC (Russia) and then
MECO (the US)
•
To eliminate beam
related background,
beam pulsing was
adopted (with delayed
measurement)
•
To increase a number of
muons available, pion
capture with a high
solenoidal field was
adopted
•
For momentum
selection, curved
solenoid was adopted
The MECO Experiment
at BNL
Cancelled in 2005
 mu2e @ Fermilab
23/Oct/2009
ILC Seminar 2009
Mu2E @ Fermilab
Fermilab Accelerators
Target
• The mu2e Experiment at Fermilab.
– Proposal has been submitted.
• CD-0
Giese Road
Detector
AP-10
– After the Tevatron shut-down
• uses the antiproton accumulator
ring
• the debuncher ring to manipulate
proton beam bunches
AP-30
AP-50
Extracted
Beam Line
From
Debuncher
MI-8
22 batches = 1. 467s MI cycle
Booster Batches
NEUTRINO PROGRAM
MUONS
4.61012 p/batch
Accumulator
(NuMI +Muons)
Recycler
56 1012 p/sec
(NuMI)
Debuncher (Muons)
44.61012 p/1467ms = 12.5 1012 p/sec
(Alternative: 24 batches=1.6s MI cycle  11.5 1012 p/s)
0.1s
23/Oct/2009
ILC Seminar 2009
1.367s
cLFV Search Experiment
• cLFV search is as important as high-energy frontier
experiments (and n oscillation measurements) to find a clue
to understand
– SUSY-GUT
– Neutrino See-saw
• MEG is expected to reach and go beyond the current limit
pretty soon
• Need other experiment(s) to confirm it
– Using “different” physics process (with better sensitivity)!
• COMET (COherent Muon Electron Transittion)
– Submitted a proposal to J-PARC in 2008 and a CDR in 2009,
– and obtained Stage-1 approval in July 2009
23/Oct/2009
ILC Seminar 2009
An Experimental Search For Lepton Flavor Violating
m- - e- Conversion at Sensitivity of 10-16
COMET
23/Oct/2009
ILC Seminar 2009
Collaboration
as of Jul/2009
Y.G. Cui, R. Palmer
Department of Physics, Brookhaven National Laboratory, USA
Y. Arimoto, Y. Igarashi, S. Ishimoto, S. Mihara, T. Nakamoto, H. Nishiguchi, T. Ogitsu,C. Omori, N. Saito, M. Tomizawa, A. Yamamoto, K. Yoshimura
High Energy Accelerator Research Organization (KEK), Tsukuba, Japan
P. Dornan, P. Dauncey, U. Egede, A. Kurup, J. Pasternak, Y. Uchida,
Imperial College London, UK
Y. Iwashita
Institute for Chemical Research, Kyoto University, Kyoto, Japan
V. Kalinnikov, A. Moiseenko, D. Mzhavia, J. Pontecorvo, B. Sabirov, Z. Tsamaiaidze, and P. Evtukhouvich
Joint Institute for Nuclear Research (JINR), Dubna, Russia
M. Aoki, Md.I. Hossain, T. Itahashi, Y. Kuno∗, E. Matsushita, N. Nakadozono, A. Sato, S. Takahashi, T. Tachimoto, M. Yoshida,
Department of Physics, Osaka University, Osaka, Japan
M. Koike, J. Sato, M. Yamanaka
Department of Physics, Saitama University, Japan
Y. Takubo
Department of Physics, Tohoku University, Japan
D. Bryman
Department of Physics and Astronomy, University of British Columbia, Vancouver, Canada
R. D’Arcy, M. Lancaster, M. Wing
Department of Physics and Astronomy, University College London, UK
E. Hungerford
Department of Physics, University of Houston, USA
T. Numao
TRIUMF, Canada
* Contact
23/Oct/2009
ILC Seminar 2009
Overview of the COMET Experiment
23/Oct/2009
ILC Seminar 2009
PROTON BEAM
23/Oct/2009
ILC Seminar 2009
Requirements for the Muon Beam
•
Backgrounds
– Beam Pion Capture
p-+(A,Z)  (A,Z-1)*  g+ (A,Z-1)
g  e+ e• Prompt timing  good Extinction!
•
–
•
m- decay-in-flight, e- scattering, neutron
streaming
Requirements from the experiment
– Pulsed
– High purity
– Intense and high repetition rate
Muon Capture(MC)
m- nuclei
Muon Decay in Orbit (MDO)
SIGNAL
23/Oct/2009
ILC Seminar 2009
Requirements for the Proton Beam
•
Proton beam structure for the mu-e conversion search
–
–
–
–
•
100nsec bunch width, 1.1msec bunch-bunch spacing
8GeV to suppress anti-proton background
< 1011 proton/bunch, limited by the detector performance
Repetition rate as high as possible within tolerable CR background
Extinction
–
Residual protons in between the pulses should be < 10-9
1.17ms (584ns x 2)
NP : total # of protons (~1021)
Rext : Extinction Ratio (10-9)
Yp/P : p yield per proton (0.015)
Ap : p acceptance (1.5 x 10-6)
Pg : Probability of g from p (3.5x10-5)
A : detector acceptance (0.18)
100ns
0.7 second beam spill
BR=10-16, Nbg < 0.12 
Extinction < 10-9
1.5 second accelerator cycle
23/Oct/2009
Nbg =
NP x Rext x Yp/P x Ap x Pg x A
ILC Seminar 2009
Proton Acceleration at J-PARC
•
Proton acceleration in
– LINAC
– Booster (RCS)
– Main Ring
•
8 filled buckets out of 9 buckets
Nominal scheme
– RCS: h=2
– MR:h=9
• 8 buckets filled
• 1 empty bucket, used for kicker
excitation
•
MR RF cavities are designed for
this scheme
– h=18 optional by removing
capacitors on cavities
– Need long shutdown to change
the configuration
23/Oct/2009
ILC Seminar 2009
Proton Acceleration for COMET
•
•
•
RCS: h=2 with one empty bucket
MR:h=8(9) with 4(3) empty buckets
Bunched slow extraction
– Slow extraction with RF cavity ON
Realization of an empty bucket in
RCS by using the chopper in Linac
•Simple solution
•No need of hardware modification
•Heavier heat load in the scraper
•Possible leakage of chopped beam in
empty buckets
23/Oct/2009
ILC Seminar 2009
LINAC Chopper
Two Cavities
池上雅紀
高エネルギーニュース vol.25 No.4
RCS, MR Injection
池上雅紀
高エネルギーニュース vol.25 No.4
• Acceleration and extraction in 20msec
• Bunch configurations
– Micro bunch 324 MHz LINAC
– Intermediate bunch ~1MHz Chopper
– Macro bunch 25Hz RCS
RCS
MLF
MR
MUON/PION PRODUCTION
23/Oct/2009
ILC Seminar 2009
Pion Production Target
•
low-E pions
•
pion yield is proportional to Tproton
•
High-Z Metal Rod like tungsten or gold
– for low-E muons to stop
– Backward extraction
– pion yld is proportional to Beam Power
– 12-mmfx 16-cm
– 3-4 kW on the target
– Water cooling
•
•
23/Oct/2009
0.3mm water layer  DT=56K
1.0mm water layer  DT=42K
ILC Seminar 2009
Pion Capture
MARS simulation
•
•
•
23/Oct/2009
m-yield vs. Bmax
ILC Seminar 2009
5 T at the target position
–
capture pt < 120 MeV/c
Radiation Shield
< 100 W on SC coil
–
–
–
3-4 kW @ target
35 kW @ W Shield
2x10-5 W/g @ coil
Yields
–
0.05(p+m)/8-GeV-proton
p-Capture Solenoid
• Heat-load density : 2 x 10-5 W/g behind W shield
• Utilize Al stabilized SC cable to reduce a heat load to the cold mass.
–
Cable dimension: 15mm x 4.7mm
Length(mm)
Thickness (mm)
rCurrent (A/mm2)
Coil 1
1200
90 (6 layers)
53.0
Coil 2
1400
30 (2 layers)
53.0
Coil 3
600
30 (2 layers)
53.0
Coil 4
300
60 (4 layers)
62.9
Al-SC: one of world leading expertise of KEK
23/Oct/2009
ILC Seminar 2009
MUON TRANSPORT
23/Oct/2009
ILC Seminar 2009
Muon Transport
Guide p’s until decay to m’s
Suppress high-p particles
• m’s : pm< 75 MeV/c
Beam Blocker
• e’s : pe < 100 MeV/c
Beam collimator
23/Oct/2009
ILC Seminar 2009
High-p Suppression
• A center of helical trajectory of
charged particles in a curved
solenoidal field is drifted by
– This effect can be used for charge
and momentum selection.
See “Classical Electrodynamics”, J.D.Jackson Ch.12-Sec.4
• This drift can be compensated by
an auxiliary field parallel to the
drift direction
dp/dx = 1 MeV/c/cm
23/Oct/2009
ILC Seminar 2009
Spectra at the End of the Muon Transport
• Preliminary beamline design
– main magnetic field
– compensation field
– Inner radius of transport
magnet cryostat (175 mm)
• Transport Efficiency
Spectra at the end of the beamline
(top left) total momentum
(top right)pt vs pL
(bottom left) time of flight
(bottom right) beam profile
.
75 MeV/c
# of m /proton
0.0071
# of stopped muons/proton
0.0035
# of muons with p>75MeV/c
/ proton
10-5
Dispersion on the
muon beam just
before the collimator.
23/Oct/2009
ILC Seminar 2009
The COMET Detector
23/Oct/2009
ILC Seminar 2009
The COMET Detector
under a solenoid
magnetic field.
to detect and
identify 100 MeV
electrons.
to stop muons in the
muon stopping target
to eliminate low-energy beam
particles and to transport only
~100 MeV electrons.
23/Oct/2009
ILC Seminar 2009
Muon Stopping Target
• Light material for delayed measurement (1st choice)
– Aluminum : tm- = 0.88 ms
• Thin disks to minimize electron energy loss in the target
– R = 100 mm, 200mmt, 17 disks, 50 mm spacing
• Graded B field for a good transmission in the downstream curved section.
• Good m-Stopping efficiency: e=0.66
–
–
23/Oct/2009
Muon rate 1.5x1011/sec
stopped-muon yields:~0.0023 m’s/proton
ILC Seminar 2009
Curved Solenoid Spectrometer
60-MeV/c DIO electrons
•
Torus drift for rejecting low energy DIO electrons.
–
•
rejection ~10-6: < 10kHz
Good acceptance for signal electrons (w/o
including event selection and trigger acceptance)
–
20%
105-MeV/c m-e electron
23/Oct/2009
ILC Seminar 2009
Electron Detectors
•
•
Rate < 800 kHz
Straw-tube tracker to measure electron momentum
–
5 Planes with 48cm distance, sp = 230 keV/c
•
•
–
–
•
One plane has 2 views (x and y) with 2 layers per view.
A straw tube has 25mm thick, 5 mm diameter.
should work in vacuum and under a magnetic field.
<500mm position resolution.
Crystal calorimeter for Trigger
–
GSO, PWO, or LYSO
23/Oct/2009
ILC Seminar 2009
Tracking Detector
•
Main detector to measure Ee
– Thickness should be about 0.01 radiation-length to
suppress % backgrounds.
– Spatial resolution < 0.5 mm
– Hit multiplicity ~ 1 per plane per event
•
Straw tube tracker
–
–
–
–
•
•
5mm&, 208 tubes per sub-layer
four layers per station
anode readout (X, X’, Y, Y’)
five stations, 48 cm apart
Amps and Digitizers (DRS) are in vacuum
Optical-link to the outside of the vacuum vessel
23/Oct/2009
ILC Seminar 2009
Trigger Calorimeter
•
Trigger Source
–
•
Energy measurement
–
–
–
•
Better trigger condition
Redundancy to Ee measurement
E/p cut to cosmic muons
Position measurement
–
–
•
Timing for Tracker
Improve track recognition.
f = 92 kHz per crystal
Possible Photon Detectors (operational in vacuum
and magnet)
–
–
APD
MPPC
23/Oct/2009
ILC Seminar 2009
Experimental Space
A possible layout
•
Discussion in the task force
– Target and beam dump outside the hall
– Share the upstream proton transport line with the high p beam line
– External extinction device in the switch yard
23/Oct/2009
ILC Seminar 2009
Signal Sensitivity
2x107 sec running
• Single event sensitivity
– Nm is a number of stopping muons in the muon stopping target. It is 2.0x1018
muons.
– fcap is a fraction of muon capture, which is 0.6 for aluminum.
– Ae is the detector acceptance, which is 0.031.
total protons
muon yield per proton
muon stopping efficiency
8.5x1020
0.0035
0.66
# of stopped muons
2.0x1018
Single event sensitivity
2.6 x 10-17
23/Oct/2009
90% C.L. upper limit
6.0 x 10-17
ILC Seminar 2009
Potential Background Events
• Background rejection is the most important in searches for rare
decays.
• Types of backgrounds for m-+Ne-+N are,
Intrinsic
backgrounds
muon decay in orbit
originate from muons stopping in
radiative muon capture
the muon stopping target.
muon capture with particle emission
Beam-related
backgrounds
radiative pion capture
muon decay in flight
caused by beam particles, such pion decay in flight
as electrons, pions, muons, and
beam electrons
anti-protons in a beam
neutron induced
antiproton induced
Other
backgrounds
23/Oct/2009
caused by cosmic rays
ILC Seminar 2009
cosmic-ray induced
(pattern recognition error)
Intrinsic Background (from muons)
• Muon Decay in Orbit
– Electron spectrum from muon decay in
orbit
– Response function of the spectrometer
included.
– 0.05 events in the signal region of 104.0
- 105.2 MeV (uncorrected).
Energy spectrum of electrons from decays in
orbit in a muonic atom of aluminum, as a
function of electron energy. The vertical axis
shows the effective branching ratio of m-e
conversion.
• Radiative Muon Capture with Photon
Conversion
– Max photon energy 102.5 MeV
– < 0.001 events
• Muon Capture with Neutron Emission
• Muon Capture with Charged Particle
Emission
– <0.001 events for both.
23/Oct/2009
ILC Seminar 2009
DIO Background
a number of events for
1.1 x1018 stopped muons.
<0.05 events
23/Oct/2009
ILC Seminar 2009
Beam Related Background Rejection
Rejection of beam related (prompt) backgrounds can be done by a
combination of the following components.
Momentum Selection at the Muon Transport
(pm < 75 MeV/c)
Electron Energy Cut
(104.0 - 105.2 MeV uncorrected)
Electron Transverse Momentum Cut
(pT> 52 MeV/c)
Timing Cut and Beam Extinction
(10-9)
Beam Channel Length
(pion decay)
23/Oct/2009
ILC Seminar 2009
Background Estimation Summary
Background
Events
Comments
Radiative Pion Capture
0.05
Beam Electrons
<0.1 MC stat limited
Muon Decay in Flight
<0.0002
Pion Decay in Flight
<0.0001
Neutron Induced
0.024 For high E n
Delayed-Pion Radiative Capture
0.002
Anti-proton Induced
0.007 For 8 GeV p
Muon Decay in Orbit
0.15
Radiative Muon Capture
<0.001
Muon Capture with n Emission
<0.001
Muon Capture with Charged Part. Emission
<0.001
Cosmic-Ray Muons
0.002
Electrons from Cosmic-Ray Muons
0.002
Total
23/Oct/2009
0.34
ILC Seminar 2009
Assuming proton
beam extinction < 10-9
R&D Status
23/Oct/2009
ILC Seminar 2009
R&D Status
• Straw-tube tracker
– Done by Osaka group for MECO
• Crystal calorimeter
• Transport Solenoid
• Extinction Measuring Device R&D
– Gas Cherenkov + Gated PMT
• Extinction measurement at J-PARC MR
23/Oct/2009
ILC Seminar 2009
Straw-tube Tracker R&D
• Seamless type tube (I.S.T.)
–
–
–
–
• Prototype Straw-tube chamber
Thickness : 25 mm
Diameter : 5mm
Material : polyimide+carbon
Resistance : 6MW/sq
Beam test using 2.0GeV/c
pion beam in 2002
23/Oct/2009
ILC Seminar 2009
Calorimeter R&D
•
30mm x 32mm x 120mm
Stack of 5 x 8 x 4 crystals
GSO crystal for PET use
– Stack in 3D
– Light loss across the
connection
•
•
Agreement btw data & MC
Need further study
– Stacking method
– Readout
– Radiation hardness?
23/Oct/2009
ILC Seminar 2009
Transport Solenoid Design
23/Oct/2009
ILC Seminar 2009
R&D on the Transport Solenoid
•
Transport Solenoid would be easier than the capture solenoid:
–
–
•
•
•
NbTi copper stabilized conductor, which is widely used in MRI magnets and commercially available,
will be used.
The solenoid will be constructed by arranging coil “pancakes” electrically and thermally connected
in series along the curve.
We are studying basic parameters by a prototype comprising of 3 pancakes.
–
–
–
•
field magnitude ~ 2 T
smaller radiation load
cooling performance
electromagnetic forces between pancakes
quench back system
A new high-Tc superconductor, MgB2, will be used for one of the coils for the first time.
–
MgB2 will be used for the electron-spectrometer and detector solenoids
23/Oct/2009
ILC Seminar 2009
Extinction Measuring Device
• Monitor extinction level online by using a Gas
Cherenkov detector with gating PMTs
– Blind to proton beam core, active only in between
proton pulses
280V, 10kHz, fall t ~100nsec
(nsec)
23/Oct/2009
Aiming at
•Repetition ~1MHz
•On/off ratio <10-6
•Long-term operation
ILC Seminar 2009
Extinction Measurement using Secondary Beam
•
Measure secondary particle time
structure relative to a reference
signal from the MR
– MR RF signal in the experimental
area
– Beam line hodoscope counters at
the K1.8BR line
• Support by E15/E17 group
p
proton
– MR operation with empty buckets
– Bunched slow extraction
– Count the number of secondary
particles as a function of time
• Particle identification
– TOF
• Integration for ~103 seconds
supposing 1MHz counting rate
Delayed reference signals
23/Oct/2009
ILC Seminar 2009
Counter signal
Extinction Measurement
•
•
•
Utilize beam monitor in the abort line
Single bunch operation of the MR
–
–
Look at the empty bucket before the filled one
Detector that can count the number of protons
Two layers of 2mmt scintillator hodoscopes
–
–
–
Support by thin carbon fiber plates
Read by Multi-anode PMT through optical fibers
Operated in the beam line vacuum
23/Oct/2009
ILC Seminar 2009
First Shot !
•
•
RCS single bunch operation
Fast extraction to the abort line, First the empty bucket and then filled bucket
Kicker excitation signal
Abort line monitor HV 300V
Outside counter normal gain
Beam intensity monitor
23/Oct/2009
ILC Seminar 2009
Chopper Phase Optimization
•
Factor of 2.5 improvement  Extinction < 2 x 10 -5
–
–
•
Estimated only from pulse height
Perhaps better than this because of scintillator saturation
The chopper (two cavities) is currently driven by a single power supply. Operation with two
independent power supply is planned. This is expected to improve.
23/Oct/2009
ILC Seminar 2009
Schedule and Construction Cost
Funding starting
1st year
design &
order of SC wires
2nd year
3rd year
4th year
5th year
engineering run
6th year
physics run
23/Oct/2009
ILC Seminar 2009
Summary
• New experiment (COMET) to search for mu-e conversion at JPARC
• COMET aims at achieving a sensitivity of 10-16
– High-intensity, high-purity pulsed proton beam at J-PARC
– Curved solenoid muon transport/spectrometer to suppress
backgrounds efficiently
• R&D work in progress
– Detector
– Magnet
– Proton beam
• Beam structure
• Extinction
23/Oct/2009
Do Join COMET
Now!
ILC Seminar 2009
Backup
23/Oct/2009
ILC Seminar 2009
High-frequency Chopper
23/Oct/2009
ILC Seminar 2009
Additional Extinction Means
•AC-dipole
•@ primary beamline
•fextinction ~ 1/100
•collaboration with mu2e
•Bunch Cleaner
•in MR
•tested at AGS for MECO
23/Oct/2009
ILC Seminar 2009