Understanding Pulse Shape Discrimination in Schweisshelm Germanium - - PowerPoint PPT Presentation

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects

DPG W¨ urzburg 2018 Barbara Schweisshelm

Max Planck Institute for Physics, Munich

March 22, 2018

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 1

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Gerda Experiment

• Outstanding background

reduction

• But: some background

always remains

• Signal and background

discrimination is indispensible → Pulse Shape Discrimination (PSD)

[Gerda Homepage]

30 Broad Energy Germanium (BEGe) detectors

• High energy resolution
• Pronounced weighting

potential → simplifies PSD

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 2

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Weighting Potential

Shockley - Ramo Theroem Q = −q ϕ0( x) Q = induced charge at electrode q = drifting charge ϕ0 = weighting potential

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 3

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Pulse Shape Discrimination in Gerda BEGes

• Pronounced weighting

potential at electrode

• Distinguish single site (SSE)

and multi site events (MSE)

• Rejection of surface events

[adapted from arXiv: 1307.2610]

Amplitude / Energy (A/E) of current signal gives information

• ver event topology → A/E used to classify events

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 4

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

2D Calibration Spectrum

A/E spectrum for Th calibration data Double escape peak (DEP), single escape peak (SEP) and full energy peak (FEP) of 208Tl What influence do diffusion effects have on this spectrum? → Simulation

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 5

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Charge Diffusion in SigGen Simulation

SigGen code by David Radford,

• pen source:

https://radware.phy.ornl.gov/MJ/mjd_siggen/

• Spherical charge clouds
• Holes and e− separately
• Cloud is traveling through

the detector → velocity depends on field strength at each point

• Charge cloud size affected

by diffusion and self-repulsion

• Track drift of point charge

and cloud dimensions separately → convolution → How does the diffusion affect A/E?

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 6

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Effect on Simulated Pulses

Energy deposition at a specified position → Energy dependence of the current pulse amplitude (A)?

1000 2000 3000 4000 Energy [keV] 91 92 93 94 95 96 97 98 A/E [a.u.] diff on, size = 0mm diff on, size = 1mm diff off, size = 0mm diff off, size = 1mm

• Diffusion creates a slope
• Initial charge cloud size additionally decreases A/E

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 7

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Effect on Simulated Pulses

Energy deposition at a specified position → Initial charge cloud size dependence of pulse amplitude?

0.00 0.25 0.50 0.75 1.00 initial charge cloud size [mm] 94 95 96 97 A/E [a.u.] diff on, E = 10keV diff on, E = 2500keV diff off, E = 10keV diff off, E = 2500keV

• A/E decreases with increasing charge cloud size
• Diffusion leads to an additional decrease with energy

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 8

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

E dependence of initial single site cloud size

• σ of hits given by Geant4 Th calibration simulation
• Higher E → bigger charge cloud size → lower A/E
• Study effect of different clustering sizes

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 9

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Electronics Model

200 400 600 800 time [ns] 250 500 750 1000 1250 Charge [a.u.] Simulation Electronics 200 400 600 800 time [ns] 20 40 60 80 100 120 Current [a.u.] Simulation Electronics

• Optimization of parameters using averaged DEP events

from data

• Electronics smoothens the pulses

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 10

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Realistic Noise

200 400 600 800 time [ns] 250 500 750 1000 1250 Charge [a.u.] Simulation Noise 200 400 600 800 time [ns] 20 40 60 80 100 120 Current [a.u.] Simulation Noise

• Sharp A/E distribution from simulated pulses
• Extract noise from baseline events
• Add noise for a more realistic distribution

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 11

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Energy Dependence of A/E

• A/E distribution with electronics and realistic noise
• A determined after 50 ns moving window average filter
• Slope of SSE band determined by fit

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 12

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Results for one BEGe

Simulated slopes [10−6 keV−1] without electronics clustering size diff on diff off 0 mm

• 8.03 ± 0.07
• 8.18 ± 0.07

0.1 mm

• 7.67 ± 0.07
• 8.15 ± 0.08

0.3 mm

• 7.09 ± 0.07
• 7.58 ± 0.07

0.5 mm

• 7.68 ± 0.06
• 5.59 ± 0.21

with electronics 0 mm

• 3.07 ± 0.07
• 3.24 ± 0.06

0.1 mm

• 3.24 ± 0.23
• 2.83 ± 0.06

0.3 mm

• 3.21 ± 0.07
• 2.69 ± 0.06

0.5 mm

• 3.14 ± 0.06
• 2.11 ± 0.06
• Uncertainty given by fit
• Slope is already induced by event size (not diffusion)
• Electronics model reduces slope

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 13

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Comparison of all Detectors

Slopes of SSE band for all detectors in data and simulation

1 2 3 4 5 7 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 30 31 32 33 34 35 Detector Channel

• 10.0
• 7.5
• 5.0
• 2.5

slope [10

6 1/keV]

run 62 run 75 simulation simulation with electronics

• Simulated slope and data are same order of magnitude
• Simulation shows similar trends as data

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 14

Understanding Pulse Shape Discrimination in Germanium Detectors: Diffusion Effects Barbara Schweisshelm Motivation Diffusion in Simulation Results Conclusion

Conclusion

• BEGe detectors → powerful event by event background

identification

• Energy dependence of A/E seems to be mainly caused by

event size

• Diffusion goes in the same direction but the effect is not

significant if the event size is taken into account

• Electronics model has a larger influence than diffusion
• Electronics model can reproduce the trends within the

strings

• Pulse shape simulations help Gerda to understand A/E

better

DPG Fr¨ uhjahrstagung W¨ urzburg - March 22, 2018 15