Rejection Power for the Hadron-cluster Background with the upgraded - - PowerPoint PPT Presentation

Rejection Power for the Hadron-cluster Background with the upgraded KOTO CsI calorimeter

2019/12/23
 Osugi Mayu @ Year End Presentation

1

• Introduction


KOTO Experiment
 Hadron-cluster background & both-end readout

• Data samples
• Timing Calculation
• Performance of both-end readout
• Summary

Contents

2

3

KOTO Experiment

KL → π0ν¯ ν

Search for

CP violating decay Highly suppressed in SM (BR : 3.0 × ) Small theoretical uncertainties (2%) 10−11

Sensitive to new physics Signal π0→ 2γ: @ CsI caloriemeter nothing : @ other detectors

CsI

KL

ν ¯ ν

π0

γ γ

Hadron-cluster background

4

Hadron-cluster Background

KL → π0ν¯ ν

PMT

CsI

γ

PMT

CsI

Upstream

beam halo neutron

Downstream

γ

50cm

radiation length ~ 2cm

50cm

interaction length ~ 40cm

Background source # of events hadron-cluster background 0.24 ± 0.17 0.05 ± 0.02 0.02 ± 0.02

• ther source

0.11 ± 0.04 Total 0.42 ± 0.18

KL → 2π0 KL → π+π−π0 Background estimation (2015 data)

Largest contribution Should be suppressed

Hadron-cluster background

5

Hadron-cluster Background

KL → π0ν¯ ν

PMT

CsI

γ

PMT

CsI

ΔT(MPPC-PMT)=small Upstream

beam halo neutron

ΔT(MPPC-PMT)=large Downstream

γ

MPPC

S13360-6050CS (HPK)

66 !!"

MPPC

depth of the interaction

ΔT

deep

6

Data Samples

Used data taken in 2019 run Gamma Sample

• Used

decay (BR: 19.52%)

Neutron Sample

• Used special run data as a neutron samples

→ Enhance the neutron events by placing Al plate KL → 3π0

CsI

n n

Al plate

Special run setup n

・ Physics data Special run data Deposit energy 2000 Arbitrary Units Deposit energy [MeV]

Timing Calculation (1)

7

Time (clock=8 ns)

10 20 30 40 50 60

ADC counts

200 400 600 800 1000 1200 1400 1600 1800 2000

Peak height + pedestal Pedestal

0.5*Peak height + pedestal

tCF

Channel Timing 
 → Constant Friction Time (tCF)

Energy

Event Display of the CsI calorimeter Cluster Calculate channel timing Calculate cluster timing (used in analysis)

Timing Calculation (2)

8

100MeV

50MeV 30MeV 5MeV

Event Display of the CsI calorimeter

Cluster Timing (MPPC)


Used 2 methods ➡ Max energy channel timing ( ) ➡ Energy weighted timing ( )
 (only for E>10 MeV channels)

Cluster Timing (PMT)

➡ Energy Weighted Timing

ΔT

➡ ➡ TMax Energy

MPPC

TEnergy Weighted

MPPC

ΔTM.E. = TMax Energy

MPPC

− TPMT ΔTE.W. = TEnegy Weighted

MPPC

− TPMT

Performance of ΔT Cut

9

ΔT range : ΔT : 25 ns < ΔT < 31.05 ns

gamma efficiency = 89.9%
 hadron-cluster background is suppressed to (2.1±0.1)%

T (ns) Δ 26 28 30 32 34 36 38 40 42 44 Arbitrary Units 0.01 0.02 0.03 0.04 0.05 0.06

deltaT

cut at = 31.05 ns 0.1)% ± neutron back ground (2.1 gamma efficiency = 89.9

hadron-cluster control sample (Data)


(w/kinematic cut, w/tight all veto)

KL → 3π0 (Data)

(w/kinematic cut)

v

Max distribution( ) ΔTE.W. ΔTE.W. = TEnergy Weighted

MPPC

− TPMT (downstream one) Max ΔT 89.9%

0.2 0.4 0.6 0.8 1

2 −

10

1 −

10 1

efficiency

gamma efficiency Hadron-cluster efficiency

ΔTM.E. ΔTE.W.

efficiency (log)

Efficiency of ΔT cut

10

Energy weighted timing( ) has better performance 
 than maximum energy channel timing ΔTE.W.

By changing the higher threshold of ΔT, efficiency (γ, hadron-cluster) are calculated

ΔTM.E. = TMax Energy

MPPC

− TPMT ΔTE.W. = TEnegy Weighted

MPPC

− TPMT

Correlation (Pulse Shape Related Cut)

11

Correlation btw. ΔT cut and other neutron cut Pulse Shape Cut : Fourier Pulse Shape Discriminator (FPSD)

Fourier PSD

Xk = xne

−2πi N kn n=0 N−1

∑

[peak-10, peak+17] Phase : 2π 28

3

Gamma Neutron

28 Fourier Template

Fourier Transformation Template γ and neutron Waveform

Minimum FPSD Value 0.2 0.4 0.6 0.8 1 Arbitrary Units

4 −

10

3 −

10

2 −

10

1 −

10 1

hadron-cluster control sample (Data)

FPSD Value

KL → 3π0 (Data)

Gamma Like Neutron Like Neutron background is suppress to ~4% (w/ 90% signal efficiency)

Correlation (FPSD vs ΔT)

12

Small correlation but 
 hadron-cluster BG is suppress to 4%

2.1% 4.0%

ΔT (w/o FPSD cut) ΔT (w/ FPSD cut) Apply FPSD cut

Any correlation btw. FPSD vs ΔT ?

• Performance of ΔT 


Hadron-cluster background is suppressed to 2.1%
 (w/90% gamma efficiency)
 Energy weighted timing has better performance than maximum energy channel timing.

• Correlation btw. pulse shape cut vs ΔT


Correlation can be seen but enough small.


Summary

13

N/S Ratio

14


 (MPPC channel energy > 10MeV)

ΔTM.E. = TMPPC Maximum Energy − TPMT ΔTE.W. = TMPPC Enegy Weighted − TPMT

N/S 
 (neutron efficiency/gamma efficiency)

ΔTM.E. ΔTE.W.

By changing the higher threshold of ΔT, N/S ratio is calculated. higher threshold of ΔT (ns)