Neutron Induced Radiation Damage of KETEK SiPMs M. Centis Vignali 1 - - PowerPoint PPT Presentation

SLIDE 1

Neutron Induced Radiation Damage

• f KETEK SiPMs
• M. Centis Vignali1c
• V. Chmillb
• E. Garuttia
• R. Klannera
• M. Nitschkea
• J. Schwandta
• S. Sondera

aHamburg University bSamara University cCERN

22.11.2016

1matteo.centis.vignali@cern.ch

SLIDE 2

SiPMs, MPPCs

> Silicon PhotoMultipliers > Multi Pixel Photon Counters > Array of avalanche photodiodes (APD) operated in Geiger mode > Quenching resistor (Rq) in series to stop avalanche > APD in series with Rq ≡ pixel > Light intensity sensitivity → count firing pixels > Pixel size: 5 − 100 µm (side) > Avalanche zone: 0.5 − 2 µm > Operating voltage: < 100 V > Gain: 105 − 106

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 2 / 18

SLIDE 3

IV Characteristic

Breakdown voltage (Vbd) Minimum bias voltage to achieve Geiger discharges in the pixels. The SiPM is operated a few voltages above the breakdown voltage IV curve for a non-irradiated SiPM Different illumination conditions

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 3 / 18

SLIDE 4

Charge Spectra Measurements

Pixel model Cpix Rq > The capacitor is charged by the bias voltage > A Geiger discharge discharges the capacitor through Rq G = Cpix(V − V ∗

bd)/q0

Pixels firing with LED illumination Distribution of the integrated pulses

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 4 / 18

SLIDE 5

Dark Count Rate, Cross Talk, After Pulse

Dark count rate (DCR) Charge carriers reach the avalanche region and cause a Geiger breakdown. Thermal generation, tunneling, diffusion, etc. Cross talk (XT) Photons are emitted during a Geiger avalanche and interact in a neighboring pixel. After pulse(AFP) Charge carriers created during an avalanche are trapped and released after some time causing a second avalanche.

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 5 / 18

SLIDE 6

Motivation

HEP applications of SiPMs > Calorimetry, timing, ... > Are SiPMs rad. hard enough? Strategy > Determine relevant parameters > Build models > Improve design Measurement Cpix Rq Vbd DCR G CN High fluence Charge (QDC) spectra X X X X X X No IV

• X

X X X X Yes CV-f X X

• Yes

Pulse height measurements do not work for highly irradiated SiPMs This talk focus on IV and CV-f measurements V > Vbd > Idark = DCR G(1 + CN)q0 > ILED = NγA∗

probG(1 + CN)q0

> G = Cpix(V − V ∗

bd)/q0

> G → gain > CN → correlated noise (XT, AFP) > Nγ → LED photon rate > A∗

prob → avalanche prob. for light

The results are specific to a SiPM model, the methods are general

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 6 / 18

SLIDE 7

Sensors

Sensor > Produced by KETEK > 1 mm diameter, 4384 pixels > Pitch: 15 µm in each direction > Depletion depth ≈ 0.75 µm > Vbd ≈ 27 V Irradiation > Bare Si die (unpackaged SiPM) > Neutrons → only bulk damage > Φeq = 109, ..., 1012 cm−2

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 7 / 18

SLIDE 8

Cpix and Rq Extraction

Data measured at 25◦C, 26.5 V (≈ Vbd − 0.5 V) Pixel model Cpix Rq Parallel capacitance

Cp = CpixNpix

Series resistance

Rs = Rq/Npix

Parameters optimized till agreement The full electrical model describes the frequency dependence of the data The relevant parameters can be extracted

Circuit model: C. Xu et al. NIM A762 (2014) 149-161 An additional inductance is needed to describe high freq.

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 8 / 18

SLIDE 9

Cpix and Rq vs Φeq

Ratio of the parameters after/before irradiation

Φeq = 0, 109, 1010, 1011, 5 · 1011, 1012 cm−2

Quenching resistance Rq

]

• 2

[cm

eq

Φ

9

10

10

10

11

10

12

10 = 0)

eq

Φ (

q

) / R

eq

Φ (

q

R 0.9 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1

> Rq(Φeq = 0) ≈ 575 kΩ > 3% increase at Φeq = 1012 cm−2 Pixel capacitance Cpix

]

• 2

[cm

eq

Φ

9

10

10

10

11

10

12

10 = 0)

eq

Φ (

pix

) / C

eq

Φ (

pix

C 0.9 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08 1.1

G = Cpix(V − V ∗

bd)/q0

> Cpix(Φeq = 0) ≈ 18 fF > Systematic shift for all fluences → effect not due to irradiation Cpix and Rq change by less than 3% with irradiation

V ∗

bd can be different than the breakdown voltage

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 9 / 18

SLIDE 10

IV measurements

T = 0◦C, LED off

Bias [V] 5 10 15 20 25 30 35 40 Current [A]

13 −

10

12 −

10

11 −

10

10 −

10

9 −

10

8 −

10

7 −

10

6 −

10

5 −

10

4 −

10

3 −

10

2 −

10

• 2

= 0 cm

eq

Φ

• 2

cm

9

= 10

eq

Φ

• 2

cm

10

= 10

eq

Φ

• 2

cm

11

= 10

eq

Φ

• 2

cm

12

= 10

eq

Φ

Φeq = 1012 cm−2, LED off

Bias [V] 5 10 15 20 25 30 35 40 Current [A]

13 −

10

12 −

10

11 −

10

10 −

10

9 −

10

8 −

10

7 −

10

6 −

10

5 −

10

4 −

10

3 −

10

2 −

10

30 C 25 C 20 C 15 C 10 C 5 C 0 C

• 5 C
• 10 C
• 15 C
• 20 C
• 25 C
• 30 C

> IV characteristics measured as a function of T, Φeq, LED on/off > Extraction of: Vbd, photodetection signal, DCR > Investigation of DCR origin

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 10 / 18

SLIDE 11

Breakdown Voltage

ILD =

• d ln(I)

dV

−1 G = Cpix(V − V ∗

bd)/q0

> ILD of LED photocurrent > min(ILD) ⇔ V = Vbd > No irradiation effects > Linear dependence on T

Φeq = 0, 109, 1010, 1011, 5 · 1011, 1012 cm−2

]

• 2

[cm

eq

Φ

9

10

10

10

11

10

12

10 ) [V]

eq

Φ (

bd

= 0) - V

eq

Φ (

bd

V 0.1 − 0.08 − 0.06 − 0.04 − 0.02 − 0.02 0.04 0.06 0.08 0.1 Temperature [C] 30 − 20 − 10 − 10 20 30 [V]

bd

V 26.2 26.4 26.6 26.8 27 27.2 27.4 27.6 27.8

• 2

= 0 cm

eq

Φ

• 2

cm

9

= 10

eq

Φ

• 2

cm

10

= 10

eq

Φ

• 2

cm

11

= 10

eq

Φ

• 2

cm

12

= 10

eq

Φ

The breakdown voltage measured using LED photocurrent is not affected by irradiation

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 11 / 18

SLIDE 12

Signal from Photodetection

Measurements: Idark, I∗

LED

ILED → LED photocurrent Inorm

LED → charge / 1 photon

Inorm

LED = ILED(V)/ILED(7.5V|M∗ = 1)

V < Vbd ILED = NγM∗q0 Nγ → photon rate M∗ → multiplication LED light Inorm

LED = M∗

V > Vbd ILED = NγA∗

probG(1 + CN)q0

G → gain CN → correlated noise (XT, AFP) A∗

prob → avalanche prob. LED light

Inorm

LED = A∗ probG(1 + CN)

> Inorm

LED spans several orders of

magnitude > Variation of Inorm

LED (Φeq=1012cm−2)

Inorm

LED (Φeq=0)

< 9% > A∗

probG(1 + CN) changes < 9%

Photon detection changes by less than 9% with irradiation

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 12 / 18

SLIDE 13

Dark Count Rate from IV measurements

Measurements: Idark, I∗

LED

ILED → LED photocurrent Inorm

LED → charge / 1 photon

Inorm

LED = ILED(V)/ILED(7.5V|M∗ = 1)

V > Vbd Idark = DCR G(1 + CN)q0 Inorm

LED = G(1 + CN)A∗ prob

G → gain CN → correlated noise (XT, AFP) A∗

prob → avalanche prob. LED light

Idark q0Inorm

LED

= DCR A∗

prob

Φeq = 0, 1012 cm−2

25 30 35 40 103 104 105 106 107 108 109 1010 1011 20°C 0 neq 20°C 1E12 neq 0°C 0 neq 0°C 1E12 neq

• 20°C 0 neq
• 20°C 1E12 neq

DCR / A* [Hz] Voltage [V]

> At 20◦C DCR increases of 5 orders of magnitude > 1011 Hz ⇒ 104 pulses / 100 ns gate DCR strongly affected by irradiation Using IV, rates higher than 107 Hz are accessible

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 13 / 18

SLIDE 14

Interlude

> The photodetector characteristics of a SiPM are not affected by irradiation > The DCR increases significantly with irradiation DCR scaling using QDC spectra If the DCR has been measured before irradiation (QDC spectra) ⇒ The DCR of the irradiated SiPM is DCR(Φeq) = DCR(Φeq = 0) Idark(Φeq) Idark(Φeq = 0) Factor 2 difference wrt method shown in previous slide What is the origin of increased DCR? Bulk generation? Diffusion? Surface generation?

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 14 / 18

SLIDE 15

Arrhenius Plot V < Vbd

V = 5 V ⇒ M = 1

]

• 1

1/kT [eV 38 40 42 44 46 48 Current [A]

15 −

10

14 −

10

13 −

10

12 −

10

11 −

10

10 −

10

9 −

10

• 2

= 0 cm

eq

Φ

• 2

cm

9

= 10

eq

Φ

• 2

cm

10

= 10

eq

Φ

• 2

cm

11

= 10

eq

Φ

• 2

cm

12

= 10

eq

Φ

> Identify origin of DCR using activation energy Ea (or trap level) > Fit function from SRH bulk generation I = I0T 2e−

Eg 2kT /cosh

• Ea−

Eg 2

kT

• V = 5, 15 V, Φeq = 0, 1012 cm−2

]

• 1

1/kT [eV 38 40 42 44 46 48 Current [A]

15 −

10

14 −

10

13 −

10

12 −

10

11 −

10

10 −

10

9 −

10 , 5 V bias

• 2

= 0 cm

eq

Φ , 15 V bias

• 2

= 0 cm

eq

Φ , 5 V bias

• 2

cm

12

= 10

eq

Φ , 15 V bias

• 2

cm

12

= 10

eq

Φ

> No description for Φeq = 1012 cm−2, 15 V → multiplication effects → eh gen. in multiplication region > No multiplication for Φeq = 0 cm−2 → eh gen. outside mult. region Activation energy changes by 0.034 ± 0.006 eV with irradiation For Φeq = 0 cm−2 current generation mainly outside multiplication region

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 15 / 18

SLIDE 16

Arrhenius Plot V < Vbd

V = 5 V ⇒ M = 1

]

• 2

[cm

eq

Φ 1 10

2

10

3

10

4

10

5

10

6

10

7

10

8

10

9

10

10

10

11

10

12

10 [eV]

a

E 0.595 0.6 0.605 0.61 0.615 0.62 0.625 0.63 0.635 0.64

> Φeq = 0 cm−2: Ea = 0.63 eV or trap level

• Ea − Eg

2

• = 0.07 eV

> Φeq = 1012 cm−2: Ea = 0.60 eV or trap level

• Ea − Eg

2

• = 0.04 eV

> σEa < 0.01 eV V = 5, 15 V, Φeq = 0, 1012 cm−2

]

• 1

1/kT [eV 38 40 42 44 46 48 Current [A]

15 −

10

14 −

10

13 −

10

12 −

10

11 −

10

10 −

10

9 −

10 , 5 V bias

• 2

= 0 cm

eq

Φ , 15 V bias

• 2

= 0 cm

eq

Φ , 5 V bias

• 2

cm

12

= 10

eq

Φ , 15 V bias

• 2

cm

12

= 10

eq

Φ

> No description for Φeq = 1012 cm−2, 15 V → multiplication effects → eh gen. in multiplication region > No multiplication for Φeq = 0 cm−2 → eh gen. outside mult. region Activation energy changes by 0.034 ± 0.006 eV with irradiation For Φeq = 0 cm−2 current generation mainly outside multiplication region

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 15 / 18

SLIDE 17

Arrhenius Plot V > Vbd

Arrhenius plot at fixed excess voltage V − Vbd = 2 V

]

• 1

1/kT [eV 38 40 42 44 46 48 Current [A]

10 −

10

9 −

10

8 −

10

7 −

10

6 −

10

5 −

10

4 −

10

3 −

10

• 2

= 0 cm

eq

Φ

• 2

cm

9

= 10

eq

Φ

• 2

cm

10

= 10

eq

Φ

• 2

cm

11

= 10

eq

Φ

• 2

cm

12

= 10

eq

Φ

> Compared at constant excess voltage G = Cpix(V − V ∗

bd(T))/q0

> Naive generation model does not describe the data > Other effects to be considered e.g. temperature dependence of ionization coefficients

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 16 / 18

SLIDE 18

Summary

> KETEK SiPM irradiated with neutrons up to Φeq = 1012 cm−2 > No change in Vbd, Cpix, and Rq with irradiation, the signal from photodetection is not affected by irradiation > DCR can be estimated from IV measurements and increases greatly with irradiation Φeq = 1012 cm−2, T = 20◦C, Vex = 3 V: DCR/A∗

prob = 10 GHz

> Activation energy (or trap level) for current generation at low voltages (M = 1) determined Outlook > You can use our data: Phenomenological model of IV curves (LED off) Φeq = 0, 1012 cm−2, −30 ≤ T ≤ 30◦C above and below Vbd

• S. Sonder Dark Current of Neutron Irradiated SiPMs Hamburg (2016)

> Complete the measurements of the other fluences > Model IV curves using TCAD and semi-analytic models

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 17 / 18

SLIDE 19

Bibliography

> C. Xu et al. Influence of X-ray irradiation on the properties of the Hamamatsu silicon photomultiplier S10362-11-050C NIM A762 (2014) 149-161 > V. Chmill et al. Study of the breakdown voltage of SiPMs NIM A (2016) > V. Chmill et al. On the characterisation of SiPMs from pulse-height spectra arXiv (2016) > E. Garutti et al. Silicon Photomultiplier characterization and radiation damage investigation for high energy particle physics applications JINST 9 (2014) 03 C03021 > M. Nitschke Characterization of Silicon Photomultipliers Before and After Neutron Irradiation Master thesis Hamburg (2016) > S. Sonder Dark Current of Neutron Irradiated SiPMs Bachelor thesis Hamburg (2016)

Acknowledgments We would like to thank Florian Wiest and his colleagues from KETEK for providing the SiPMs samples and for fruitful discussions.

Thank you for your attention!

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 18 / 18

SLIDE 20

Backup Material

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 19 / 18

SLIDE 21

Calculation Avalanche Probability

T = 20◦C, w = 0.75 µm

Bias [V] 26 28 30 32 34 36 38 40 Avalance probability 0.2 0.4 0.6 0.8 1 Mean avalanche prob. Avalanche prob. e at w

SiPM with 25 × 25 µm2 pixels > Constant E field approximation > Avalanche prob. model from Mc Intyre 1973 > Ionization coefficients from van Overstraeten-de Man (Synopsis TCAD manual) Red → e− at the p++n+ junction Blue → ¯ Aprob (dark current)

• V. Chmill et al. Study of the breakdown voltage of SiPMs NIM A (2016)

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 20 / 18

SLIDE 22

Calculation Multiplication Coefficient

T = 20◦C, w = 0.75 µm

Bias [V] 5 10 15 20 25 Multiplication factor 2 4 6 8 10 12 14 16 18 20

Average M M electron at w Ratio

SiPM with 25 × 25 µm2 pixels > Constant E field approximation > Multiplication model from Mc Intyre 1966 > Ionization coefficients from van Overstraeten-de Man (Synopsis TCAD manual) Red → e− at the p++n+ junction Blue → M (dark current)

• V. Chmill et al. Study of the breakdown voltage of SiPMs NIM A (2016)

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 21 / 18

SLIDE 23

Setup

Measurements: > Current > Capacitance > Resistance Environment: > Temperature control > LED illumination (470 nm) > Dry atmosphere

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 22 / 18

SLIDE 24

LCR SiPM Model

Z =

• 1

Rpar + jωCpar + Npix

• 1

jωCpix + Rq 1 + jωCqRq −1−1 > 1 pixel about to experience breakdown: C∗

pix, Rbd, C∗ q, R∗ q

> Npix − 1 pixels: Cpix, Cq, Rq > Parasitic contributions: Cpar, Rpar

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 23 / 18

SLIDE 25

LCR SiPM Model

Z =

• 1

Rpar + jωCpar +

• jωLs +

1 Npix

• 1

jωCpix + Rq 1 + jωCqRq −1−1 > 1 pixel about to experience breakdown: C∗

pix, Rbd, C∗ q, R∗ q

> Npix − 1 pixels: Cpix, Cq, Rq > Parasitic contributions: Cpar, Rpar > Ls inductance to improve data description

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 23 / 18

SLIDE 26

DCR scaling

T = 20◦C, Φeq = 109 cm−2 DCR(Φeq) = DCR(Φeq = 0) Idark(Φeq) Idark(Φeq = 0)

• M. Nitschke Characterization of Silicon Photomultipliers Before and After Neutron Irradiation

Master thesis Hamburg (2016) Fit for DCR extraction:

• V. Chmill et al. On the characterisation of SiPMs from pulse-height spectra, arXiv (2016)

22.11.2016 Neutron Induced Radiation Damage of KETEK SiPMs 24 / 18