1. NEW APPROACH TO THE MUON g2 AND EDM EXPERIMENT AT J-PARC
Hiromi Iinuma (KEK) for
　New collaboration
Motivation and goal
Outline of new experiment
Summary

2. Motivation for Muon g-2 and EDM
Standard model
+ + +
      
Electric Dipole Moment
Scaling from electron EDM
EDM~1025 e.cm
A recent theoretical work
EDM = 1022~ 1024 e.cm ?
G. Hiller et al. hep-ph/ v1 (2010, Aug.)
EDM upper limit:1.8 1019 e.cm
(95% C.L.)
PRD 80, (2009)
Go beyond at least order of magnitude!
Recent theoretical work
HLMNT10(preliminary)
3.2 (Tau2010)
Latest experimental result：
PRD73, (2006)
Muon anomalous magnetic moment is one of clear probe to new physics.
There have been repeatedly measured for several tens years at CERN and Brookhaven National Lab, and they made great improvements to reduce experimental uncertainty down to sub-ppm.
Careful comparison between experimental result and standard model calculation give us a hint of new physics.
Deference between experiment and theory is 3.2-sigma from particle data book.
Very recent theoretical study by HLMNT collaboration is shown in the Tau2010 conf. which is two weeks ago,
(This new value is obtained by including the new BaBar ISR data as well as the new KLOE radiative return data.
They are now preparing a paper for this.)
Anyway, I would say we need one more step to conclude whether we find new physics or not.
And we are now working on new super precise measurement of muon g-2.
We are also thinking to measure muon electric dipole moment.
As everybody knows, non-zero EDM is a direct evidence of new physics.
However, it is difficult to judge which level of sensitivity is enough to conclude EDM is absolutely zero or not.
And I think we are not still reaching to the enough precision to discuss this.
Therefore we are also planning to improve muon EDM measurement at least order of magnitude!
2010/9/28

3. Measure two frequency vectors separately!!
How to measure a =(g2)/2 and EDM?
Measure the muon precession frequency in the uniform magnetic field B
< 0.1 ppm B
How to….?
Measure……
In the magnetic field perpendicular to the screen, muon spin and momentum vectors rotate on the screen-plane as this way.
Green arrow depicts muon motion and red arrow depicts muon spin.
Both angular momenta are perpendicular to the screen and we can take the difference of them.
This difference, called omega-a is linear to the magnetic field and directly connected to the a_mu.
Therefore, to measure a_mu, we simply measure omega_a and magnetic-field B, very precisely.
In case of non-zero EDM presence, muon precession frequency is re-written as this way.
It is clear that we can decompose g-2 term and EDM term by measuring…
“Non-zero EDM case”
Measure two frequency vectors separately!!
2010/9/28

4. Great E821 and beyond the horizon
Magic momentum + beam cancels focusing electric field term
EDM=0
2/a=4.4sec
150 (g-2) Cycles in Positron Time Spectrum! 　
PRD73, (2006)
More muons! Statistics uncertainty < 0.1 ppm (0.46ppm)
For the better B/B <0.1ppm ( 0.17ppm);
Compact storage ring to reduce the field volume
For the better a/a <0.1 ppm ( 0.21ppm);
Ultra-cold muon beam, even “off-magic” momentum
High granularity detector
Measure two frequency vectors separately for g2 and EDM measurements
Independent experimental approach provides independent       systematic studies for even clearer physics understanding!! e+
In this slide, I would like to mention about experimental method of great E821 briefly.
They use magic momentum muon beam to cancel this term.
This term comes from electric focusing field E.
And they treated EDM term is zero in their analysis.
This is a storage ring.
They detect decay positrons beyond proper energy threshold, and plot its time spectrum,
to measure precession frequency.
This is called as wiggle plot.
Statistical and systematic uncertainties are in the following precisions.
This is a very beautiful experiment and, I would like to break the world record.
To do this, we need more muons to improve statistical uncertainty.
For the better systematic control and measurement of the field, we should use compact storage ring.
For the better precession frequency measurement, we need ultra-cold muon beam,,,
We basically do not use focusing electric field .
…………………..
Finally, I would emphasize that,…..
So, let’s go beyond the horizon!
B=1.45 Tesla、diameter 14m、45m round
12 magnet yoke pieces
Ee+　> 1.8 GeV calorimetry
2010/9/28

5. High Intensity Muon Beam in JAPAN!
J-PARC
Thursday 10:00 Prof. N. Saito
Bird’s eye photo in Feb. 2008
J-PARC Facility
(KEK/JAEA)
P, 3GeV
KEK 
-beam
High intensity muon beam in Japan, J-PARC.
Details of J-PARC will be given by Prof. Saito on Thursday.
Tokyo
P, 30GeV
2010/9/28

6. Step1: Ultra-Cold + Source and LINAC
3 GeV proton beam
( 333 uA)
New Muon g-2/EDM Experiment at
J-PARC with Ultra-Cold Muon Beam
Graphite target (20 mm)
Surface muon beam
(28 MeV/c, 4x108/s)
Muonium Production
(300 K ~ 25 meV⇒2.3 keV/c)
Surface muon
Ultra Cold m+ Source
Muon
storage
Muon LINAC (300 MeV/c)
Step1: Ultra-Cold + Source and LINAC
Laser
Proton beam
(3 GeV, 1MW, 25 Hz)
This is a conceptual drawing of the muon beam line for the new g-2/EDM experiment.
There are 3 steps.
The first step is Ultra-….
3 GeV proton beam hit the target.
Surface muons of 28MeV/c are produced and they hit the 2nd target, which is room temperature, to produce muonium.
High power laser waiting just another side of the 2nd target to hit and ionize muonium.
Then we have ultra-cold positive muons of 2.3 KeV/c.
And accelerate before it decays…up tp 300MeV/c.
As this way we have ultra-cold straight muon beam.
Muon Linac
(300 MeV/c)
(2.3 keV/c)
(28 MeV/c)
2010/9/28

7. Step2: Injection & storage
3 GeV proton beam
( 333 uA)
New Muon g-2/EDM Experiment at
J-PARC with Ultra-Cold Muon Beam
Graphite target (20 mm)
Surface muon beam
(28 MeV/c, 4x108/s)
Muonium Production
(300 K ~ 25 meV⇒2.3 keV/c)
Surface muon
Super Precision Magnetic Field
(3T, ~1ppm local precision)
Ultra Cold m+ Source
Muon
storage
Muon LINAC (300 MeV/c)
Step2: Injection & storage
+ beam
Step3: Detect decay e+
Step two is beam injection and storage.
I am going to discuss this later, but I would emphasize the diameter is only 0.66m.
This is a cute compact ring compared to the E821 storage ring, which is 14m in diameter.
3rd step is to detect and measure momentum of positron.
Detector is inside of the storage ring with presence of high magnetic field.
In order to survive high event rate condition, we use double sided Si Tracker.
Basic design is from Belle sensor with SiLC –collaboration based front-end-electronics.
(International Collaboration of Silicon tracker for the Linear Collider)
By use of this silicon tracker system, we obtain wiggle plots.
0.66 m diameter
=3 and B=3 [T]
2010/9/28
(note: 14 m for E821)

8. Expected precession signal from J-PARC within a Snowmass year beam time !
Ee> 200MeV
Default 50% pol.
Study for high pol. Is ongoing.
This is an expected precession signal from J-PARC within a single Snowmass year beam time.
Statistical uncertainty is 1ppm.
15 wiggles are there in 5 life-times in 3 tesla storage magnetic field.
2010/9/28

9. Example EDM wiggle, if EDM=210-20 e.cm
Expected sensitivity for EDM would be better than
210-20 e.cm
Let’s think of EDM is finite value, say 2 times 10 to the minus 20.
If we take up-down time spectra separately, we can see such plots.
EDM effect appears in amplitude of the up-down asymmetry.
This amplitude can be calculated by analytically, and consistent with this simulation result.
In this way, we can measure EDM effect and extract pure g-2 effect from precession frequency.
These two wiggles are our goal plots.
2010/9/28

10. R&D items are running! Step1: Ultra-cold + source and LINAC
Test experiments to search the best Mu production target are ongoing at TRIUMF and RAL !
Hi Power Ly- Laser System R&D
LINAC R&D
Step2: Injection & storage
See next few pages
Step3: Detect decay e+
Setup spin dependent muon-decay including EDM term in GEANT4
Positron detector design
Hi-rate Si Tracker
Belle sensor with SiLC based FEE
This slide is a short summary of our current R&D activities.
We expect a big progress of ultra-cold muon beam production study from
test experiments at TRIUMF and RAL.
As for positron detector, conceptual design has been done by use of GEANT4.
I contribute some development of GEANT4 source code to set up proper environment for
spin-dependent muon decay in the electromagnetic field, including EDM term.
We are going to have proto-type sensors in the next Spring.
Now, I would discuss about injection and storage for the rest of my slides.
GEANT4
2010/9/28

11. Back to Step2: Injection & storage
Technical difficulties:
3T is too high to cancel fringe field by inflector,
Required kick angle (~60 mrad) is too big.
3D-spiral injection +
We do not apply horizontal beam injection.
Because of technical……
Instead of horizontal injection, we try 3D spiral Injection Scheme:
Long axis field, like a solenoid, is needed.
We make use of radial fringe field to deflect vertical motion to horizontal.
And then, we apply vertical soft kick to "stop" the vertical motion completely in the storage region.
Concept is simple
but real design ?
2010/9/28

12. Storage ring magnet (ver.0 design)
3T super conductive solenoid magnet
Uniformity in the beam storage region<0.1ppm
Careful design of fringe field for stable beam injection
Tunnel for + beam
Upper end cap (Iron)
Pole tip (Iron)
Cylindrical return yoke (Iron)
1.8m
Apply MRI technology!
Super conductive Main coils
Inner radius 1.6m
Conceptual idea of 3-D spiral injection is simple, but real design is a bit difficult.
We are planning to use 3 tesla… and the magnet need to satisfy two requirements at a time.
Note that, storage volume is compact, but magnet itself is very big, to achieve such high performance.
But if you look at inner radius of main coils, they are 1.6m.
This size is similar to the medical MRI and we apply its technology very much!
Upper half
4.6m
1ppm level at storage volume is achieved
2010/9/28

13. Storage ring magnet (ver.0 design)
3 tesla super conductive solenoid magnet
Uniformity in the beam storage region<0.1ppm
Careful design of fringe field for stable beam injection +
1.8m
Now, I have a single muon beam from the outside of the magnet, through the tunnel, and fringe field region, and then, storage region.
I use OPERA for this study.
Upper half
4.6m
+ beam orbit diameter 0.66m
2.1m round (cyclotron period=7.4 nsec)
2010/9/28

14. Beam acceptance study is ongoing+
y<0
Solenoid axis
y>0
y>0
y<0 y
Start
Beam acceptance
Beam acceptance study is ongoing.
I tried many initial condition by changing start position and initial angle.
And compare these tracks with the basic track to check beam acceptance.
Colored region is a group of acceptable trajectories.
Beam transport from the end point of the LINAC will be designed to meet this acceptance..
2010/9/28

15. Vertical kicker (ver.0) Helmholtz type Br(t)=Bpeak sin(t) Coil ±10cm
This is a conceptual design of vertical kicker.
We will use two pairs of coils to apply….within few tens of turns.
Proto…
±10cm
STOP!
Apply radial magnetic field to reduce beam vertical momentum
Prototype kicker is being designed for test .
Coil
2010/9/28

16. Summary A new muon g2 and EDM experiment at JPARC:
Off-magic momentum
Ultra-cold muon beam + compact g-2 ring
Independent experimental approach provides independent systematic uncertainty
Complementary to New
Saturday 10:00 Prof. B. Lee Roberts
Active R&D efforts are ongoing!
My next step: Verification test for injection + kicker system
2010/9/28

17. J-PARC g-2/EDM collaboration
71 members (…still evolving)
M. Aoki, P. Bakule, B. Bassalleck, G. Beer, A. Deshpande, S. Eidelman, D. E. Fields, M. Finger, M. Finger Jr., Y. Fujirawa, S. Hirota, H. Iinuma, M. Ikegami, K. Ishida, M. Iwasaki, T. Kakurai, T. Kamitani, Y. Kamiya, N. Kawamura, S. Komamiya, K. Koseki, Y. Kuno, O. Luchev, G. Marshall, M. Masuzawa, Y. Matsuda, T. Matsuzaki, T. Mibe, K. Midorikawa, S. Mihara, Y.Miyake, J. Murata, W.M. Morse, R. Muto, K. Nagamine, T. Naito, H. Nakayama, M. Naruki, H. Nishiguchi, M. Nio, D. Nomura, H. Noumi, T. Ogawa, T. Ogitsu, K. Ohishi, K. Oide, A. Olin, N. Saito, N.F. Saito, Y. Sakemi, K. Sasaki, O. Sasaki, A. Sato, Y. Semeritzidis, K. Shimomura, B. Shwartz, P. Strasser, R. Sugahara, K. Tanaka, N. Terunuma, D. Tomono, T.Toshito, K. Ueno, V. Vrba, S. Wada, A. Yamamoto, K. Yokoya, K. Yokoyama, Ma. Yoshida, M. H. Yoshida, and K. Yoshimura
18 Institutions
Academy of Science, BNL, BINP, UC Riverside, Charles U., KEK, NIRS, UNM, Osaka U., RCNP, STFC RAL, RIKEN, Rikkyo U., SUNYSB, CRC Tohoku, U. Tokyo, TRIUMF, U. Victoria
6 countries
Czech, USA, Russia, Japan, UK, Canada
Thank you!
2010/9/28

18. Backups
2010/9/28

19. Requirements for Kicker
Br(t)=Bpeak sin(t)
=/Tkick
Field strength：
Bpeak = 1Gauss ~ 10 Gauss
Time distribution：
Tkick=150 nsec (c.f. 20 cyclotron periods)
Spatial distribution：
33cm±5mm in radial direction, 1% uniformity
±10cm in solenoid axis direction to reduce distortion o f
beam bunch shape
Minimal effect for positron detector：
Quench protection
Space problem
Eddy currents on cryostat wall
e+ detector
(Silicon tracker)
2010/9/28

20. Kicker circuit and eddy current study
R=60cm cryostat wall
Start DAQ
2010/9/28

21. Injection orbit stability: Shallow vs. deep angle orbit
Smaller angle orbit gets bigger integral field effect
Need to decrease
510-4 Tesla Br
0.8 deg.
2 deg.
20% 5%
Small angle orbit requires “smooth” Br distribution
Smooth Br provides easy requirements for beam and kicker
2010/9/28

22. “good field region” weak magnetic focus ?
Focus condition　0<n<1
n<1: for radial direction
n>0 :solenoid axis direction
If we require:
We have n=3E-5
3D field measurement is needed…
2010/9/28

23. Field inside beam tunnel
Cross-section1
Gauss
Inside tunnel
Cross section2
Cross section3
Cross section1
Outside
Absolute field strength along the tunnel
2010/9/28

24. Field vectors We are trying square and circle shapes Cross section1
Cross section2 8cm
10cm
We are trying square and circle shapes
Cross section3 50cm
2010/9/28

25. Field measurement =NMR frequency measurement
Measure  and  well better than 0.1ppm
Muonium hyper splitting experiment to improve 
(ongoing at J-PARC)
2010/9/28

26. Recent ee experimental data made big contributions (BaBar, KLOE….)
K. Hagiwara et al., PLB 649 (2007)
 5.1
Theoretical calculation
aSM =
10-10
QED
 0.0
Dominant uncertainty
T.Kinoshita, M.Nio
693.2  5.1
QCD
K.Hagiwara D. Nomura
Dominant component~99%
(ee ~73%)
QED, QCD and Weak interaction contribute to a_mu. Dominant component is from QED, which is very precisely calucurated. Among QCD diagrams, this is a dominant one.
In particular, e+e- going to pi+pi- is major contributer.
Contribution of weak interaction is weak, true to its name.
If you look at the uncertainty, QCD is the biggest and
These two diagrams have major uncertainty.
As for this LO diagram, recent…made….
On the other hand, this diagram is difficult to calculate and still model dependent.
Weak
15.4  0.2
Recent ee experimental data made big contributions (BaBar, KLOE….)
2010/9/28

27. ! is a known function (PRD69 093003), Nambu-Goldstone boson’s exchange
K. Hagiwara et al., PLB 649 (2007) !
is a known function (PRD ),
This diagram is LO term of hadronic component.
This diagram is e+e- going to hadron.
This term is calculated from this diagram but require experimental results as input for these equations.
Recent experimental data made a great help to reduce the uncertainty from this diagram.
However, on the other hand, this term called light-by-light term is still model dependent calculation.
Some progress is reported recently.
As you can see, there are discrepancy between the SM calculation and experimental result.
Nambu-Goldstone boson’s exchange M
J. Prades, arXiv: v1
0, ,’
2010/9/28

28. BNL, FNAL, and J-PARC BNL-E821 Fermilab J-PARC Muon momentum
3.09 GeV/c
0.3 GeV/c
gamma
29.3 3
Storage field
B=1.45 T
3.0 T
Focusing field
Electric quad
None
# of detected m+ decays
5.0E9
1.8E11
1.5E12
# of detected m- decays
3.6E9 -
Precision (stat)
0.46 ppm
0.1 ppm
2010/9/28

29. 4x104 ultra-cold muon/spill with p=2.3keV/c
Muon source
Requrements:
40000 times more muons, and
Cooler muon than RAL
670 times higher surface muon per spill at J-PARC
2.4 x 104/spill  1600 x 104/spill (25 spill/sec)
Room temperature target (hot tangsten  silica aerogel?)
2000K (15keV/c)  300K (2.3keV/c)
100 times intense laser
1mJ 100mJ
4x104 ultra-cold muon/spill with p=2.3keV/c
2010/9/28

30. g-2:Stored Energy / Cold Mass K. Sasaki, T. Ogitsu, et al.
Not an extreme, but requires serious efforts
Material : NbTi /Copper
Cu/Sc ratio : 4
Central Field:3T
Peak Field on Cable: 5.4 T
Nominal current : 417 A
Stored Energy : 23 MJ
Inductance : H
Total mass : 3.7 t
Well within
current
Technology !
2010/9/28

31. g-2 silicon tracker Detector area Number of sensors Number of channels
0.12 * number of vanes [m2]
2.9 m2 for 24 vanes
Number of sensors
384 for 24 vanes
Number of channels
Assume 0.2 mm pitch
115k for 24 vanes*
*288k for multi-segments readout
576 mm
580 mm
g-2 silicon tracker
2010/9/28

32. Silicon strip module Support Readout chip DSSD sensors front back
2010/9/28 32

33. Silicon sensor Sensor type： Double-sided SSD Chip size : ~12 cm x 6 cm
Thickness: um
Readout: AC-couple
Depletion voltage : 80 V
Detector capacitance : ~100pF*
Strip pitch : 200um*
* to be determined by further studies.
From Belle SVD page
p-side
n-side
2010/9/28

34. Spin equation (T-BMT equation + EDM)
Our case:
2010/9/28