Design and status of the Mu2e crystal calorimeter · Mu2e experiment design The Mu2e Calorimeter, R.Donghia 2 PS TS DS May 10, 2018 Detector Solenoid: stopping target and detectors

Design and status of the Mu2e crystal calorimeter

Raffaella Donghia LNF-INFN and Roma Tre university

On behalf of the Mu2e calorimeter group

May , 2016 XVII LNF Spring School

"Bruno Touschek”

Talk overview

Mu2e Electromagnetic calorimeter •  Requirements •  Components •  Single Channel Tests •  “Module-0” performance •  Production phase

The Mu2e Calorimeter, R.Donghia 1 May 10, 2018

Mu2e experiment design

The Mu2e Calorimeter, R.Donghia 2

PS TS

DS

May 10, 2018

Detector Solenoid: stopping target and detectors •  Stops μ- on Al foils

•  Events reconstructed by detectors optimized for 105 MeV/c momentum

•  CLFV strongly suppressed: Branching Ratio ≤10-54 à Observation would indicate New Physics

•  CLFV@Mu2e:μ- e conversion in a nucleus field à discovery sensitivity to many NP models

•  Goal: μ-e conversion in the presence of a nucleus

< 8.4 x 10-17

Nuclear captures of muonic Al atoms

µe Al

ECE = 104.97 MeV

(@ 90% CL, with ~ 1018 stopped-µ)

Calorimeter requirements


•  σE/E = 𝓞(5%) for CE •  σT < 500 ps for CE •  σX,Y ≤ 1 cm

•  Fast scintillation signals (τ

Calorimeter Design

The Mu2e Calorimeter, R.Donghia

2 annular disks with 674 undoped CsI (34 x 34 x 200) mm3

square crystals/each disk

o  RIN = 374 mm, ROUT = 660 mm o  Depth = 10 X0 (200 mm), Distance 70 cm o  Readout: 2 UV-extended SiPMs/crystal o  FEE on the SiPMs , Digital readout on crates o  RA source for energy calibration o  Laser system for monitoring

4 May 10, 2018

BETTER PICTURE!!

Simulated performance


Simulation of CsI+SiPM performanceSimulation includes full background and digitization and cluster-finding, with split-off and pileup recovery

Energy resolution

Dependence on LRUand photostatistics

Specification is LRU

Small prototype: Test Beam


•  Small prototype tested @ BTF (Frascati) in April 2015, 80-120 MeV e-

•  3×3 array of 30×30×200 mm2 undoped CsI crystals coupled to one Hamamatsu SiPM array (12x12) mm2 with Silicon optical grease

•  DAQ readout: 250 Msps CAEN V1720 WF Digitizer

JINST 12 (2017) P05007

FOTO matrice

Log-normal fit

1 year long R&D phase for the final test of the option CsI + UV extended SiPM

72 crystals + 150 SiPM + 150 FEE chips completed in 2016

Small prototype: Time and Energy resolution


Significant leakage contribution due to block dimensions w.r.t. the shower

σT ~ 110 ps at 100 MeV σE ~ 6.5% at 100 MeV

JINST 12 (2017) P05007

PRE-PRODUCTION

hEntries 0Mean 0RMS 0

Energy [GeV]0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09

[ns]

tσ

0

0.05

0.1

0.15

0.2

0.25

0.3

hEntries 0Mean 0RMS 0

/ ndf 2χ 38 / 17a 0.00015± 0.0049 b 0.0033± 0.087

/ ndf 2χ 38 / 17a 0.00015± 0.0049 b 0.0033± 0.087

b⊕ = a/E tσ

Single crystal @ 0 degAll crystals above 10 MeV @ 0 deg

CosmicsNeighboring crystals @ 50 degSingle crystal @ 50 deg

Most energetic crystal @ 50 degAll crystals above 10 MeV @ 50 deg

Figure 24. Time resolution summary plot.

leakage due to the small dimensions of the prototype. The time resolution �t as a function of theenergy deposition has been measured using three di�erent techniques that consistently show that�t ranges from 250 ps at 22 MeV to about 120 ps above 50 MeV. These results satisfy the Mu2erequirements and also significantly improve the timing resolution achievable when using undopedCsI at these low energies compared to present results in literature [6, 7] thanks to the use of SiPMand high frequency waveform digitizer boards.

Acknowledgment

The authors express their sincere thanks to the operating sta� of the Beam Test Facility in Frascati(Italy), for providing us a good quality electron beam, and the technical sta� of the participatinginstitutions. This work was supported by the US Department of Energy; the Italian Istituto Nazionaledi Fisica Nucleare; the US National Science Foundation; the Ministry of Education and Science ofthe Russian Federation; the Thousand Talents Plan of China; the Helmholtz Association of Germany;and the EU Horizon 2020 Research and Innovation Program under the Marie Sklodowska-CurieGrant Agreement No.690385.

References

[1] L. Bartoszek et al., Mu2e Technical Design Report, 3rd ed. arXiv:1501.05241 [physics.ins-det],2014.

[2] N. Atanov et al., Design and status of the Mu2e electromagnetic calorimeter, NIM A 824, 695 - 698,2016.

[3] K. A. Olive et al, Review of Particle Physics, Chin. Phys. 14, 2014.

Total energy [GeV]0.07 0.075 0.08 0.085 0.09 0.095 0.1 0.105

/E [%

]Eσ

44.5

55.5

66.5

77.5

88.5

9

/ ndf 2χ 2.866 / 3a 0.3253± 1.38 b 1.092± 4.911

/ ndf 2χ 2.866 / 3a 0.3253± 1.38 b 1.092± 4.911

data

Monte Carlo

b⊕ E [GeV]

a = EEσ

Figure 17. Energy resolution obtained from the data (black) taken at 0 degrees compared with the MonteCarlo (red).

Figure 18. Distribution of reconstructed energy obtained from the data overlaid with the Monte Carlo forthe run with beam energy of 100 MeV and 50 degrees incidence angle. Blue line represents a fit to the datawith a log-normal function.

obtained with this technique, Method 1 was used to measure the time resolution in the same events.Figure 22 shows the time residual between tcrystal(1,1) and tscint. Subtracting in quadrature the tscintjitter of 100 ps results in a time resolution of about 209 ps that is compatible with the result obtainedwith Method 3.

/MeVpeN100 110 120 130 140 150 160 170 180 190 200

Entr

ies

0

2

4

6

8

10

LRU (%)0 1 2 3 4 5 6 7 8 9 10

Entr

ies

0

2

4

6

8

10

12

(%)µ/σ10 11 12 13 14 15 16 17 18 19 20

Entr

ies

0

2

4

6

8

1012

14

16 SICCAS

Amcrys

Saint Gobain

F/T (%)60 65 70 75 80 85 90 95 100 105

Entr

ies

0

2

4

6

8

10

12

14

Pre-production Crystals


•  24 crystals from three different vendors: SICCAS, Amcrys, Saint Gobain •  Optical properties tested with 511 keV γ’s along the crystal axis •  Crystals wrapped with 150 µm of Tyvek and coupled to an UV-extended PMT

Energy resolution

Light Yield Longitudinal Resp. Uniformity

Q(200 ns)/Q(3000 ns)

RMS/MEAN of Light

Output values

along axis

Fast/Total

Un-doped CsI crystals perform well

•  Excellent LRU and LY: -100 pe/MeV with PMT readout -LRU < 5% §  τ of 30 ns with small slow component

§  Radiation hardness OK

Smaller than 40% LY loss @ 100 krad

Entries 102900Mean 90.04Mean y 33.18RMS 63.51RMS y 33.26

Time [ns]0 50 100 150 200

Am

plitu

de [m

V]

020406080

100120140160



Time [ns]0 50 100 150 200

Am

plitu

de [m

V]

0

100

200

300

400

500 Entries 102900Mean 90.09Mean y 60.63RMS 63.51RMS y 110.6

Pre-production test: SiPMs (1)


Gain spread – vendor 1

• Relative spread at the level of 2%• All tested devices satisfy the requirement

6/19/2017G. Pezzullo @ Calorimeter CRR for CsI and SiPMs14

Vop spread – vendor 1

• 35 devices tested• Relative spread at the level of 0.07%• All Mu2e SiPMs from this vendor satisfy this requirement

6/19/2017G. Pezzullo @ Calorimeter CRR for CsI and SiPMs4

150 sensors: 3×50 Mu2e pre-production SiPMs from Hamamatsu, SenSl and AdvanSiD •  3×35 were fully characterized for all six cells in the array

Operating Voltage

Gain

Mu2e custom silicon photosensors: à 2 arrays of 3 6 x 6 mm2 UV-extended SiPMs: total area (12x18) mm2 The readout series configuration reduces the overall capacitance and allows us to generate faster signals

~ 150 V

i1≈ i2 ≈ i3Ctot ≈ C1/3

6x6 mm2

K1

A1

A1-1

A1-2

Single cell of 6 x 6 mm2

Series of 3 cells

Module 0

• The outer shell has been initially machined in Lecce. Now in INFN Padova where they have EDM

• Front plate in Lecce• FEE plate in Pisa• FEE holders order out• Crystals being wrapped• SiPM tested• FEE on the way

FabioHappacher7

Module 0


Large EMC prototype: 51 crystals, 102 SiPMs, 102 FEE boards

Mechanics and cooling system similar to the final ones! Goals:

•  Integration and assembly procedures •  Test beam May 2017, 60-120 MeV e-

(beam perpendicular and @ 50°)

•  Work under vacuum, low temperature, irradiation test

3/29-30/2017Fabio Happacher | MDR Review14

1.91335e+06

128053 93017.6

81310.4

99920.4112888

135539

67022.8 102991 98354.5

103895

108076

123601

151623128720130298

94299.6

88138.3

86969.5

136475 79070.2 127412 110208

103408

134562

111132

134589

103162

10689712596287092.6103181

26479

100319

122841

104284

120963

97679.7

76132.1

106792

79394.9

104968

104676

96822.4107369

98298.7

108205

121208

101734

122820

107887

x [mm]200− 150− 100− 50− 0 50 100 150 200

y[m

m]

200−

150−

100−

50−

0

50

100

150

200

Entries 663136Mean x 0.01988Mean y 0.1337− Std Dev x 62.82Std Dev y 55.23 0 0 0 0 0 0 0 0 0

1

10

210

310

410

510

610

Entries 663136Mean x 0.01988Mean y 0.1337− Std Dev x 62.82Std Dev y 55.23 0 0 0 0 0 0 0 0 0

Module 0 Energy resolution


σE ~ 5.4 % (entire matrix)

@ Ebeam = 100 MeV

§  Calibration - cosmic rays - beamàcompleted for second

ring around central crystal §  Data quality good,

resolution in agreement with 3x3 prototype

Entries 1811

/ ndf 2χ 4.943 / 4

η 0.104± 0.277

σ 0.240± 5.075

µ 0.34± 93.14

N 104.1± 3168

E [MeV]50 60 70 80 90 100 110 120

Ent

ries/

2 M

eV

0

50

100

150

200

250Entries 1811

/ ndf 2χ 4.943 / 4

η 0.104± 0.277

σ 0.240± 5.075

µ 0.34± 93.14

N 104.1± 3168

Orthogonal runEbeam=100 MeV

Orthogonal runEbeam=100 MeV

Module 0 Time resolution


TimeHist / ndf 2 106.6 / 20

0.0444 0.6526 1.21 17.12 0.7 216.6

N 640.2 8782

t [ns]0 200 400 600 800 1000

Am

plitu

de [m

V]

0

50

100

150

200

250

300

350 TimeHist / ndf 2 106.6 / 20

0.0444 0.6526 1.21 17.12 0.7 216.6

N 640.2 8782

Entries 1531

/ ndf 2 19.56 / 15

Constant 7.6 240.9

Mean 0.0049 0.1664

Sigma 0.0035 0.1874

t [ns] 1.5 1 0.5 0 0.5 1 1.5

Ent

ries

/ (0.

075

ns)

0

50

100

150

200

250Entries 1531

/ ndf 2 19.56 / 15

Constant 7.6 240.9

Mean 0.0049 0.1664

Sigma 0.0035 0.1874

Fit Example Central crystal time distribution

(TSiPM 1 – TSiPM2)

σ (T1+T2)/2 ~ 94 ps @ Ebeam = 100 MeV

•  Selection on single particle •  Log-normal fit on leading edge •  Constant Fraction method used à CF = 5%

Δ

ησµ

Module 0 Results


Test Beam data analysis results satisfy Mu2e requirements à @ 100 MeV

•  σE ~ 5.44 ± 0.25 % •  σT < 100 ps, both @ 1 GHz and 250

MHz sampling rate à MC simulation still on going

Energy [MeV]0 20 40 60 80 100

[ns]

Tσ

0

0.05

0.1

0.15

0.2

0.25

0.3

/ ndf 2χ 1.506 / 4a 0.2903± 6.217 b 0.009087± 0.08504

/ ndf 2χ 1.506 / 4a 0.2903± 6.217 b 0.009087± 0.08504

/ ndf 2χ 1.997 / 2a 0.4899± 4.449 b 0.007553± 0.07

/ ndf 2χ 1.997 / 2a 0.4899± 4.449 b 0.007553± 0.07

Perpendicular Beam

Beam at 50Â°

0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12 [MeV]depE

0123456789

10

[%]

dep

/Eσ

a 0± 0.6

b 0.02119± 0.3277

c 0.2454± 4.006

a 0± 0.6

b 0.02119± 0.3277

c 0.2454± 4.006

�EE

=apE

� bE

� c

σT = a/E + b

Calorimeter Assembly status


New laboratory built at FNAL. QA tests of all components started on march 2018

Summary and Conclusions


•  The Mu2e calorimeter has concluded its prototyping phase satisfying the Mu2e requirements:

•  Un-doped CsI crystals perform well §  Excellent LRU and LY 100 pe/MeV ( PMT+Tyvek wrapping ) §  τ of 30 ns with negligible slow component §  Radiation hardness OK for our purposes: 40% LY loss at 100 krad

•  Mu2e SiPMs quality OK, high gain, high PDE, small Idark, small spread inside array •  SiPM performance after irradiation OK •  SiPM MTTF > 0.6 million hours

•  Single calorimeter channel: timing resolution of 215 ps /MIP •  Small prototype tested with e- beam

§  Good time and energy resolution achieved @ 100 MeV •  Module 0 built and first tests done.

•  TB data analysis results satisfies the requirements •  Good time and energy resolution achieved @ 100 MeV, in agreement with the small prototype

•  Calorimeter production phase started •  Detector installation expected for beginning of 2020

Spares

Raffaella Donghia LNF-INFN and Roma Tre university

On behalf of the Mu2e calorimeter group

May , 2016 XVII LNF Spring School

"Bruno Touschek”

Charged Lepton Flavor Violation


•  CLFV strongly suppressed in SM: Branching Ratio ≤10-54 à Observation would indicate New Physics

•  CLFV@Mu2e:μ- e conversion in a nucleus field à discovery sensitivity to many NP models

•  Goal: 104 improvement w.r.t. current limit (SINDRUM II)

(@ 90% CL, with ~ 1018 stopped muons in 3 years of running)

μ-e conversion in the presence of a nucleus

< 8.4 x 10-17


µe

Al

ECE = mμc2 – Eb – Erecoil= = 104.97 MeV

Mu2e experiment design


25 m Production Solenoid / Target •  Protons hitting target and

producing mostly π

Transport Solenoid •  Selects and transports low

momentum μ-

PS

TS

DS

May 10, 2018

1.  Generate low momentum μ- beam

2.  Stop the muons in an Al target à trapped in orbit around the nucleus 3.  Look for an excess around 105 MeV/c in the electron spectrum

Detector Solenoid: stopping target and detectors •  Stops μ- on Al foils •  Events reconstructed by detectors optimized

for 105 MeV/c momentum

Mu2e

Charged Lepton Flavor Violation


•  CLFV strongly suppressed in SM: BR ≤10-54 à Observation indicates New Physics

•  CLFV@Mu2e:μ- e conversion in a nucleus field à discovery sensitivity on many NP models

•  Goal: 104 improvement w.r.t. current limit (SINDRUM II)

(@ 90% CL, with ~ 1018 stopped muons in 3 years of running)

Contact dominated

Loop dominated

μ-e conversion in the presence of a nucleus

< 8 x 10-17


µe

Al

arXiv: 1303.4907

Pre-production test: SiPMs (2)


•  1 sample per vendor has been exposed to neutron flux up to 8.5×1011 n1MeVeq/cm2 (@ 20 °C)

•  5 samples per vendor have been used to estimate the mean time to failure value Requirement: obtain an MTTF of 1 million hours when operating at 0 °C

]2/cm1MeV

Integrated flux [n0 100 200 300 400 500 600 700 800

910

I [m

A]

0

10

20

30

40

50

60

•  MTTF evaluated operating SiPMs @ 50 °C for 3.5 months

•  No dead channels observed MTTF ≥ 6×105 hours

•  SiPMs will operate @ 0 °C: a decrease of 10 ˚C in SiPMs temperature corresponds to a Id decrease of 50%

•  Lower Vop also helps to decrease the Id

MTTF Neutron test

HamamatsuSenSl AdvanSiD

SG crystal + Hamamatsu SiPM + FEE Optical coupling in air. •  22Na source

-  TRG: small scintillator readout by a PMT -  Study distance effect for air-coupling

Single channel slice test


h1Entries 4443222

Mean 145.5

Mean y 10.67

RMS 43.59

RMS y 12.41

Time [ns]80 100 120 140 160 180 200 220

Am

plitu

de [m

V]

0

5

10

15

20

25

h1Entries 4443222

Mean 145.5

Mean y 10.67

RMS 43.59

RMS y 12.41

h2Entries 4375308

Mean 500

Mean y 2.651

RMS 288.4

RMS y 5.918

h3Entries 4128348

Mean 500

Mean y 2.812

RMS 288.4

RMS y 5.987

Hamamatsu 4 - DIstance from Crystal0 mm1 mm2 mm

onda3-1.3:time {tmean2-tmean3>-40 && tmean2-tmean3120 && charge2

Single channel Cosmic Rays Test


•  TRG time resolution ~ 170 ps •  Constant fraction method used •  Pulse height correction applied (slewing)

After jitter subtraction: SiPM 1 – σT ~ 330 ps SiPM 2 – σT ~ 340 ps

T(SiPM1 - SiPM2)/2 à ~ 215 ps @ ~ 23 MeV energy deposition

(MIP energy scale from Na22 source peak)

Timing result well compares with old tests: à  Reduced light output/SiPM

(22 vs 30 pe/MeV) à 2 SiPMs/crystal à  LY of 44 vs 30 à 215 ps (now) vs 250 ps (old).

Entries 727

Constant 6.2± 133.6

Mean 0.0161± 0.1234

Sigma 0.012± 0.429

Time [ns]-3 -2 -1 0 1 2 3

Entr

ies

/ 25

ps0

20

40

60

80

100

120

140

Entries 727


Mean 0.0161± 0.1234

Sigma 0.012± 0.429

SiPM 1 - SiPM 2Entries 727


Mean 0.012± -0.297

Sigma 0.0083± 0.3073

Time [ns]-3 -2 -1 0 1 2 3

Entr

ies

/ 25

ps

0

20

40

60

80

100

120

140

160

180

Entries 727


Mean 0.012± -0.297

Sigma 0.0083± 0.3073

/2Σ(SiPM1 + SiPM2)/2 -

Pre-production test: Crystals (2)


Few samples per vendor have been exposed both to ionizing dose and neutrons •  Irradiation test up to 100 krad •  Requirement:

normalized LY after 10/100 krad > 85/60%

•  Radiation Induced Noise (RIN) @ 1.8 rad/h required is < 0.6 MeV •  All 72 samples tested. All OK apart some Amcrys crystals that do not satisfy the required limit

•  Negligible LY and LRU variation after 1.6 x 1012 n1MeV/cm2 integrared flux •  Neutron RIN is also smaller than the one from dose

Most crystals have LY larger than 100 p.e./MeV after 100 krad

(40% max. loss), promising a robust CsI calorimeter

85%

60%

PId

The Mu2e Calorimeter, R.Donghia 24 May 10, 2018 24

With a CRV inefficiency of 10-4 an additional rejection factor of ~ 200 is needed to have < 0.1 fake events from cosmics in the signal window

A rejection factor of 200 can be achieved with ~ 95% efficiency for CE

Calibration source and laser

25 May 10, 2018

+ cosmics

•  Liquid source FC 770 + DT generator: 6 MeV + 2 escape peaks •  Laser system to monitor SiPM performance

The Mu2e Calorimeter, R.Donghia

Calorimeter Trigger


•  Calo info can provide additional trigger capabilities in Mu2e: •  Calorimeter seeded track finder

•  Factorized into 3 steps: hit pre-selection, helix serach and track fit •  ε ~ 95% for background rejection of 200

•  Standalone calorimeter trigger that uses only calo info •  Ε ~ 65% for background rejection 200

Calorimeter seeded track finder


Calorimeter radiation damage


•  Calorimeter radiation dose driven by beam flash (interaction of proton beam on target)

•  Dose from muon capture is x10 smaller •  Dose is mainly in the inner radius •  Highest dose ~10 krad/year •  Highest n flux on crystals ~ 2×1011 n/cm2/year •  Highest n flux on SiPM ~ 1011 n1MeVeq/cm2/year • 

•  Qualify crystals up to ~ 100 krad, 1012 n/cm2

•  Qualify SiPM up to ~ 1012 n1MeVeq/cm2

This includes a safety factor of 3 for a 3 year run

Calorimeter radiation damage


Three years run Expectation by full Simulation


Mu2e Collaboration, November 2013

May 10, 2018

CLFV Lagrangian


Mu2e Collaboration, November 2013

μN→eN

μ→eγ

μ→eee

Loops dominate for κ > 1

κ

μN→eN

μ→eγ

μ→eee

Λ (

TeV

)

May 10, 2018

Documents

Design and status of the Mu2e crystal calorimeter · Mu2e experiment design The Mu2e Calorimeter, R.Donghia 2 PS TS DS May 10, 2018 Detector Solenoid: stopping target and detectors