1. Mu2e experiment at Fermilab
Yuri Davydov
DLNP, JINR, Dubna
Erevan, March 2015

2. m- N  e- N Introduction What is Mu2e?
A search for Charged-Lepton Flavor Violation via
Will use current Fermilab accelerator complex to reach a sensitivity better than current world’s best
Will have discovery sensitivity over broad swath of New Physics parameter space
m- N  e- N
March 2015

3. We’ve known for a long time that quarks mix  (Quark) Flavor Violation
In last 15 years we’ve come to know that neutrinos mix  Lepton Flavor Violation (LFV)
Why not Charged Lepton Flavor Violation (CLFV)?
In general, new physics models have to work hard to *suppress* FNCN. And, of course, empirically, we all know that the other fermions in the SM mix… the quarks mix, the neutral leptons mix – so, to me, it would be surprising if the charged leptons *didn’t* mix. The key questions, then, are 1) at what level do they mix? And 2) what can we learn from measuring CLFV rates?
Cabbibo-Kobayashi-Maskawa Pontecorvo-Maki-Nakagawa-Sakata
March 2015

4. CLFV in the Standard Modelq m- e- W g nm ne
Strictly speaking, forbidden in the SM
Even in ν-SM, extremely suppressed
(rate ~ Δmν2 / Mw2 < 10-50)
However, most all NP models predict rates observable at next generation CLFV experiments
… now, this last statement shouldn’t surprise you.
March 2015

5. The BR of CLFV processes in the Standard Model
Mu2e : SM prediction and New Physics
The BR of CLFV processes in the Standard Model
Mu2e sensitivity
is 6*10-17 NP
Sensitive to mass scales up to O(10,000 TeV)
March 2015
Flavour Physics of Leptons and Dipole Moments, Eur.Phys.J.C57:13-182,2008

6. Mu2e – High Priority for U.S. HEP
In the 2008 P5 report Mu2e was strongly supported:
“Mu2e should be pursued in all budget scenarios considered by the panel”
In 2010 P5 reiterated their support of the 2008 plan and the priorities specified therein.
In 2013 the Facilities Panel gave Mu2e the highest endorsement:
“The science of Mu2e is Critical to the DOE OHEP mission and is Ready to Construct.”
In the 2014 P5 report Mu2e is strongly supported:
Recommendation 14, “Complete the Muon (g-2) and Mu2e Projects.”
March 2015

7. How does Mu2e work?
March 2015

8. Mu2e Concept Generate a beam of low momentum muons (
Stop the muons in a target
Mu2e plans to use aluminum
Sensitivity goal requires ~1018 stopped muons
The stopped muons are trapped in orbit around the nucleus
In orbit around aluminum: Al = 864 ns
Large N important for discriminating background
Look for events consistent with mNeN
March 2015

9. The process is a coherent one
Mu2e Signal
The process is a coherent one
The nucleus is kept intact
Experimental signature is an electron and nothing else
Energy of electron: Ee = m - Erecoil - E1S-B.E.
For aluminum: Ee= MeV
Important for discriminating background
March 2015

10. Design goal: single-event-sensitivity 2.4 x 10-17
Mu2e Sensitivity
Design goal: single-event-sensitivity 2.4 x 10-17
Requires about 1018 stopped muons
Requires about 1020 protons on target
Requires extreme suppression of backgrounds
Expected limit: Rme < 6 x 90% CL
Factor 104 improvement
Discovery sensitivity: all Rme > 1 x 10-16
Covers broad range of new physics theories
March 2015

11. µ-e conversion experiment concept
The great-grandparents of the Mu2e
( MELC, 1992; MECO, 1997)
V.M. Lobashev and R.M. Djilkibaev
March 2015

12. Baseline Mu2e Apparatus
8-GeV protons from
delivery ring
(about 25 meters end-to-end
Particles produced
from tungsten target
Muons stop on Al target,
which emits an electron isotropically
2.0 T
2.5 T
1.0 T
4.6 T
Detector Region:
Uniform Field 1T
Graded B for most of length
S-shaped solenoid:
transports particles to detector area, and
allows remaining pions to decay to muons
collimator selects negatively-charged particles
Tracker/calorimeter detect
electron signature
March 2015

13. The Measurement Method
Ee = m - Erecoil - E1S-B.E.
Ee= MeV

14. Some Mu2e numbers Every 1 second Mu2e will By the time Mu2e is done…
Send 7,000,000,000,000 protons to theProduction Solenoid
Send 26,000,000,000 ms through the Transport Solenoid
Stop 13,000,000,000, ms in the Detector Solenoid
By the time Mu2e is done…
Total number of stopped muons
1,000,000,000,000,000,000
March 2015

15. Mu2e Proton Beam
Time (ns)
Proton pulse on
Production target
Muons at
Stopping
target
1700 ns
700 ns
900 ns
Live Window
I mentioned that the long lifetime of the muonic Al. atom was important in discriminating background. I’d like to take a minute to explain why before discussing the background sources in more detail. Mu2e will use a pulsed proton beam (previous mu2e experiments used a DC beam).
Mu2e will use a pulsed proton beam and a delayed live gate to suppress prompt backgrounds
March 2015

16. Will employ straw technology
The Mu2e Tracker
Will employ straw technology
Low mass
Can reliably operate in vacuum
Robust against single-wire failures
5 mm diameter straw
Spiral wound
Walls: 12 mm Mylar + 3 mm epoxy
+ 200 Å Au Å Al
25 mm Au-plated W sense wire
33 – 117 cm in length
80/20 Ar/CO2 with HV < 1500 V
March 2015

17. The Mu2e Tracker Self-supporting “panel” consists of 100 straws
plane
station
Self-supporting “panel” consists of 100 straws
6 panels assembled to make a “plane”
2 planes assembled to make a “station”
Rotation of panels and planes improves stereo information
>20k straws total
March 2015

18. The Mu2e Tracker 18-20 “stations” with straws transverse to beam
Naturally moves readout and support to large radii, out of active volume
3 meters
38 cm
70 cm
March 2015

19. Inner 38 cm is purposefully un-instrumented
The Mu2e Tracker
Escape
thru center
of Tracker
Fully
fiducial
Inner 38 cm is purposefully un-instrumented
Blind to beam flash
Blind to >99% of DIO spectrum
March 2015

20. Mu2e Calorimeter Crystal calorimeter Compact Radiation hard
Good timing and energy resolution
March 2015

21. Mu2e Calorimeter Baseline design : Barrium Flouride (BaF2)
Radiation hard, very fast, non-hygroscopic
BaF2
Density (g/cm3)
4.89
Radiation length (cm)
2.03
Moliere Radius (cm)
3.10
Interaction length (cm)
30.7
dE/dX (MeV/cm)
6.52
Refractive index
1.50
Peak luminescence (nm)
220 (300)
Decay time (ns)
1 (650)
Light yield (rel. to NaI)
5% (42%)
Variation with temperature
0.1% (-1.9)% / deg-C
March 2015

22. Mu2e Calorimeter Will employ 2 disks ( radius = 36-70 cm)
~2000 crystals with hexagonal cross-section
~3 cm diameter, ~20 cm long (10 X0)
Two photo-sensors/crystal on back (APDs or SiPMs)
March 2015

23. Veto system covers entire DS and half TS
Mu2e Cosmic-Ray Veto PS TS
Without the veto system, ~1 cosmic-ray
induced background event per day
Veto system covers entire DS and half TS
March 2015

24. Will use 4 overlapping layers of scintillator
Mu2e Cosmic-Ray Veto
Will use 4 overlapping layers of scintillator
Each bar is 5 x 2 x ~450 cm3
2 WLS fibers / bar
Read-out both ends of each fiber with SiPM
Have achieved e > 99.4% (per layer) in test beam
March 2015

25. Backgrounds Stopped Muon induced
Muon decay in orbit (DIO)
Out of time protons or long transit-time secondaries
Radiative pion capture; Muon decay in flight
Pion decay in flight; Beam electrons
Anti-protons
Secondaries from cosmic rays
Mitigation:
Excellent momentum resolution
Excellent extinction plus delayed measurement window
Thin window at center of TS absorbs anti-protons
Shielding and veto
March 2015

26. Designed to be nearly background free
Mu2e Backgrounds
Category
Source
Events
 Decay in Orbit
0.20
Intrinsic
Radiative Capture
<0.01
Radiative  Capture
0.02
Beam electrons
 Decay in Flight
Late Arriving
 Decay in Flight
Anti-proton induced
0.05
Miscellaneous
Cosmic Ray induced
0.10
Total Background
0.37
(assuming 6.7E17 stopped muons in 6E7 s of beam time)
Designed to be nearly background free
March 2015

27. Mu2e Intrinsic Backgrounds
Once trapped in orbit, muons will:
Decay in orbit (DIO): N --> e- eN
For Al. DIO fraction is 39%
Electron spectrum has tail out to MeV
Accounts for ~55% of total background
March 2015

28. Mu2e Intrinsic Backgrounds
Once trapped in orbit, muons will:
Capture on the nucleus:
For Al. capture fraction is 61%
Ordinary Capture
NZ --> N*Z-1
Used for normalization
Radiative capture
NZ --> N*Z-1 + 
(# Radiative / # Ordinary) ~ 1 / 100,000
E kinematic end-point ~102 MeV
Asymmetric  -->e+e- pair production can yield a background electron
March 2015

29. Prompt Background Suppression
Happens around the time, when the beam arrives at the target.
Sources
beam electrons,
muon decay in flight,
pion decay in flight,
radiative pion capture
May create electrons with energies in the signal region
Prompt background can be suppressed by not taking data during the first 670 ns after the peak of the proton pulse.
However, this prompt background cannot be eliminated entirely, since some of the protons arrive “out of time”.
A ratio of is required for the beam between pulses vs. the beam contained in a pulse.
670 ns
925 ns
1695 ns
The lifetime of a muon in an Al orbit is 864 ns
March 2015

30. Mu2e Late Arriving Backgrounds
Contributions from
Radiative Capture
NZ --> N*Z-1 + 
For Al. RC fraction: 2%
Eextends out to ~m
Asymmetric e+e- pair production can yield background electron
Beam electrons
Originating from upstream and decays
Electrons scatter in stopping target to get into detector acceptance
Muon and pion Decay-in-Flight
Taken together these backgrounds account for ~10% of the total background and scale linearly with the number of out-of-time protons
March 2015

31. Backgrounds for 3 Year Run
Source
Events
Comment
μ decay in orbit (DIO)
0.20 ± 0.06
Anti-proton capture
0.10 ± 0.06
Radiative π- capture*
0.04 ± 0.02
From protons during detection time
Beam electrons*
0.001 ± 0.001
μ decay in flight*
0.010 ± 0.005
With e- scatter in target
Cosmic ray induced
0.050 ± 0.013
Assumes 10-4 veto inefficiency
Total
0.4 ± 0.1
All values preliminary; some are stat error only.
* scales with extinction: values in table assume extinction = 10-10
March 2015

32. Mu2e Miscellaneous Backgrounds
Several additional miscellaneous sources can contribute background - most importantly:
Anti-protons
Proton beam is just above pbar production threshold
These low momentum pbars wander until they annihilate
A thin mylar window in beamline absorbs most of them
Annihilations produce high multiplicity final states e.g.  can undergo RC to yield a background electron
Cosmic rays
Suppressed by passive and active shielding
 DIF or interactions in the detector material can give an e- or  that yield a background electron
Background listed assumes veto efficiency of 99.99%
March 2015

33. Signal Sensitivity for 3 Year Run
~4x1020 protons on target
(3 year 8 KW)
Stopped μ: 1018
Nμe = 3.94 ± For R = 10-16
NDIO = 0.20 ± 0.06
NOther = 0.2
SES = (2.5 ± 0.1) × 10-17
Errors are stat only
Reconstructed e- Momentum
March 2015

34. LYSO 5x5 matrix beam tests
March 2015

35. JINR CONTRIBUTION (1) Mu2e Electromagnetic calorimeter
Evaluation of crystal samples including LGSO, LSO, LYSO, BaF2 and CsI. Tests of the crystals on the gamma sources, cosmic muons and accelerator beams. Tests of the optical uniformity, scintillation yield and yield collection uniformity.
Upgrade of the facility for the crystal testing at DLNP JINR.
Simulation of the electromagnetic calorimeter elements and calorimeter in whole.
Participation in the radiation hardness of the crystal and front-end electronic tests on the neutron sources of JINR and/or collaborator institutes.
Participation in the production of crystals for the experiment in cooperation with Kharkov.
Participation in quality control of the crystals by testing their performance.
Participation in calorimeter construction and integration in the full detector on the experimental site.
Maintenance of the calorimeter during the data taking period to ensure efficient operation.
March 2015

36. JINR CONTRIBUTION (2) Mu2e Cosmic Ray Veto system
Participation in the simulation activity of Mu2e experiment to define the final demands to Cosmic Ray Veto system and choose the optimal geometry of this system.
Upgrade of the facility for scintillation counters tests at DLNP JINR and Fermilab.
Design, build and test the prototype scintillation counters for Cosmic Ray Veto system at the created stands.
Participation in the prototype scintillator counters tests on the neutron facilities of JINR and/or collaborator institutes to define sensitivity of neutron registration and radiation hardness.
Participation in the mass production and tests of the scintillation counters for Cosmic Ray Veto system in the cooperation with Kharkov, Fermilab and University of Virginia (UVA).
Participation in the CRV test stand construction at Fermilab.
Participation in the CRV construction and integration in the full detector on the experimental site.
Maintenance of this system during the data taking period to provide its efficient operation.
March 2015

37. JINR CONTRIBUTION (3)
March 2015

38. Experimental setups for crystals and strips tests
March 2015

39. LYSO – SICCAS: 22Na, coincidence
Two left frames: linear scales, the bottom frame shows 1275 keV and keV sum peaks
Right bottom frame shows spectrum in log scale along with peaks fitted with Gaussian

40. JINR Mu2e: Comparison of 3 crystals - Energy resolution and Linearity of the energy response

41. Sample from ISMA NASU* with triangular cross-section and inner hole
The triangle scintillation strips with a hole in the center (diam. 2.6 mm) are similar to those developed for MINERvA
Water-filling hole
plug
Geometry for the our strip sample:
Base: 33 mm
Height: 17 mm
Length: 2000 mm
For water-filling holes were drilled in the side faces of the sample and plugs are made (see photo).
Fiber end
* ISMA NASU – Institute for Scintillation Materials, National Academy of Sciences of Ukraine

42. Test conditions
Measurements of the light yields for the same sample of the scintillation strip with/without distilled water inside were done;
Light collected by using d = 1.2 mm Kuraray WLS fiber. The strip’s far end was covered by Aluminum foil, near end was closed by black plastic cap for light isolation. Just hole left for coupling the WLS fiber to the PMT;
The WLS fiber was inserted into the hole and fixed by glue on both ends;
Distilled water pumped through the sample, no air bubbles were observed.
PMT EMI 9814B was used, HV= 2200 V. Four pairs of trigger counters were used;
Absolute calibration method was used to calculate the light yield 2 2 4

43. Results: comparison of the light collected of the same sample using the wls fiber with/without water
Fit function f(x)=expp0+p1*x
Measuring points (см) :
Increase of the yield of light (%) :

44. Some publications
L. Bartoszek et al., “Mu2e Technical Design Report”, arXiv: (2015).
K. Afanaciev et al., “Response of LYSO:Ce scintillation crystals to low energy gamma-rays”, “Particles and Nuclei, Letters”, 2015, issue 2.
G. Pezzullo et al., “Progress status for the Mu2e calorimeter system”, Journal of Physics: Conference Series 587(2015),
G. Pezzullo et al., “The LYSO crystal calorimeter for the Mu2e experiment”, 2014 JINST 9 C03018.
O. Sidletskiy et al., “Evaluation of LGSO:Ce scintillator for high energy physics experiments”, Nucl. Instr.&Meth. A735(2014)
J. Budagov et al., “The calorimeter project for the Mu2e experiment”, Nucl. Instr.&Meth. A718(2013)
Z. Usubov, “Light output simulation of LYSO single crystal”, arXiv: (2013).
Z. Usubov, “Electromagnetic calorimeter simulation for future µ→e conversion experiments”, arXiv: (2012).
R.J. Abrams et al., “Mu2e Conceptual Design Report”, arXiv: (2012).
March 2015

45. Mu2e Technical Schedule
CD-1
CD-3a
CD-2/3
Superconductor R&D
Fabricate and QA Superconductor
Solenoid Design
Solenoid Fabrication and QA
Field Mapping
Install Detector
Operations
Detector Hall Design
Site Work
Detector Hall Construction
Solenoid Infrastructure
Solenoid Installation
Detector Construction
Accelerator and Beamline
FY FY FY FY FY FY FY FY20
March 2015

46. What next?
Precision Measurement if necessary
Measure conversion rate as a function of Z
Higher Sensitivity search
Mu2e Signal?
Yes No
Accelerator
Upgrade
A next-generation Mu2e experiment makes sense in all scenarios
Push sensitivity or
Study underlying new physics
Will need more protons  upgrade accelerator
March 2015

47. Improves sensitivity by a factor of 104
Summary
The Mu2e experiment:
Improves sensitivity by a factor of 104
Provides discovery capability over wide range of New Physics models
Is complementary to LHC, heavy-flavor, and neutrino experiments
March 2015

48. BACKUP
March 2015

49. March 2015

50. Mu2e / COMET comparison Mu2e COMET approval/ funding Operation
ranked among the very top priorities for the U.S. HEP program
P5 Report 2014
“Complete the Mu2e and g-2 projects.”
Mu2e is fully funded in all budget scenarios
The DOE is committed to completing Mu2e
COMET phase-II funding has not been identified and Japan has clearly stated that their top priorities are Belle-II, long baseline neutrinos, and ILC
Operation
Condition
Mu2e will run simultaneously with NOvA and the short-baseline neutrino program at Fermilab
COMET cannot run simultaneously with the JPARC neutrino program.
It forces COMET to plan for higher beam power in order to minimize the amount of required beam time
Detector
Straight Solenoid with gradient field
Tracker and Calorimeter
C-shape Solenoid with gradient field
the COMET solenoids, particularly their production solenoid, is more technically risky (e.g. they will use a 9 (!) layer coil, Mu2e is 3 layer).
Because the Mu2e DS is straight, it is equally sensitive to e- and e+ and thus have an additional physics channel μ-N  e+ N’ and will be able to measure in situ some background components (e.g. π-N  γN’ e-e+N’)
March 2015

51. March 2015

52. Mu2e / COMET comparison
Mu2e has consistently been ranked among the very top priorities for the U.S. HEP program
P5 Report 2008
Facilities Report 2012
P5 Report 2014
“Complete the Mu2e and g-2 projects.”
Mu2e is fully funded in all budget scenarios
The DOE is committed to completing Mu2e
COMET phase-II funding has not been identified and Japan has clearly stated that their top priorities are Belle-II, long baseline neutrinos, and ILC
Mu2e solenoids are technically less risky and employ fabrication and cooling techniques that have been used before
the COMET solenoids, particularly their production solenoid, is more technically challenging (e.g. they will use a 9 (!) layer coil, Mu2e is 3 layer).
March 2015

53. Status of CLFV Searches
March 2015