Characterization of the prototype CMOS pixel sensor JadePix-1 for - - PowerPoint PPT Presentation

▶
characterization of the prototype cmos pixel sensor
SMART_READER_LITE
LIVE PREVIEW

Characterization of the prototype CMOS pixel sensor JadePix-1 for - - PowerPoint PPT Presentation

Nov 19, 2022 28 likes •291 views

Characterization of the prototype CMOS pixel sensor JadePix-1 for the CEPC vertex detector Liejian Chen 1,2 , Jia Tao 1,2 , Hongbo Zhu 1,2 , Ying Zhang 1,2 , Xiaocong Ai 1,3 , Yi Liu 3 , Chenfei Yang 4 , Ryuta Kiuchi 1,2 , Xin Shi 1,2 , Ke Wang 1,2

slide-1
SLIDE 1

Characterization of the prototype CMOS pixel sensor JadePix-1 for the CEPC vertex detector

Liejian Chen1,2, Jia Tao1,2, Hongbo Zhu1,2, Ying Zhang1,2, Xiaocong Ai1,3, Yi Liu3, Chenfei Yang4, Ryuta Kiuchi1,2, Xin Shi1,2, Ke Wang1,2 , Na Wang1,2, Zhenan Liu1,2, Qun Ouyang1,2, Xinchou Lou1,2

1Institute of High Energy Physics, CAS 2State Key Laboratory of Particle Detection and Electronics 3Deutsches Elektronen-Synchrotron DESY 4University of Science and Technology of China

9th International Workshop on Semiconductor Pixel Detectors for Particles and Imaging, PIXEL2018, December 10-14, 2018, Taipei

slide-2
SLIDE 2

Outline

2

▪ Introduction ▪ Pixel design ▪ Prototype performance ▪ Summary and outlook

13 Dec, 2018, PIXEL2018

slide-3
SLIDE 3

3

Introduction - CEPC & SppC

13 Dec, 2018, PIXEL2018

▪ Phase 1: Circular Electron Positron Collider (CEPC)

  • Higgs(Z) factory：Ecm≈240 GeV, luminosity ~2x1034 cm-2s-1, 2

Interaction Points(Detectors), 1M clean Higgs over 10 years + operation at Z-pole (91 GeV) and WW (160 GeV)

  • Higgs boson + EW precision measurements

▪ Phase 2: Super proton proton Collider (SppC)

  • Discovery machine for new physics: upgrade to pp collision with

Ecm≈50-100 TeV (+ ep, HI options), luminosity ~1x1035 cm-2s-1

slide-4
SLIDE 4

4

CEPC Vertex Detector

▪ Vertex detector of CEPC, essential for identification of heavy-flavor quarks and 𝜐 leptons, designed to achieve excellent impact parameter resolution： 𝜏𝑠𝜚 = 5 μm ⊕ 10 μm 𝑞(GeV) ∙ sin Τ

3 2 𝜄

▪ Baseline design: three double layers pixelated vertex detector

13 Dec, 2018, PIXEL2018 Physics driven requirements Sensor specifications Single−point resolution < 3 μm Small pixel 16 μm? Material budget 0.15% X0 per layer Thinning 50 μm Low power 50 mW/cm2 R of Inner most layer 16 mm Fast readout Radiation tolerance (Higgs mode) TID 0.93 Mrad/y NIEL 2.1x1012 1 MeV neq/cm2/y

slide-5
SLIDE 5

5

Prototype Design– JadePix-1

▪ TJ 0.18 μm CMOS image process with high resistance epi-layer ▪ Goal: sensor diode geometry optimization ▪ Design remarks:

  • diode area, footprint
  • pixel pitch

33 x 33 μm2 16 x 16 μm2

▪ Submission in Nov 2015, test system developed and validated in 2017, detailed performance characterization this year

13 Dec, 2018, PIXEL2018

slide-6
SLIDE 6

6

DAQ System

▪ Analog signal from sensor amplified on the daughter board ▪ Converted to digital signal on the mother board ▪ Data transmitted to PC via PCIe after processed on evaluation board ▪ Data took automatically with modern multi-thread C++ software

13 Dec, 2018, PIXEL2018

JadePix-1: TJ 180 nm CPS Daughter Board: 16 channels amplifier Mother Board: 16 bit ADC KC705 FPGA Board: 6 Gbps PCIe

slide-7
SLIDE 7

7

Performance of DAQ system

13 Dec, 2018, PIXEL2018

Output waveform comparison(after amplified) (VIVADO ILA and Oscilloscope) DAQ system noise distribution without chip

▪ The reference voltage of ADC Vref = 4.096 V: 1 LSB = Vref/2N-1=0.125 mV ▪ Output signal from sensor responding to 1 LSB responds: 0.125mV/8 = 15.6uV ▪ Without chip test to estimate DAQ system noise: ~170μV~3.5e- Daughter board Mother board FPGA board

slide-8
SLIDE 8

8

Tests with 55Fe

▪ Correlated Double Sampling (CDS) to suppress noise and extract signals ▪ Noise measured with/without radioactive source (exclude suspected signals and get multiple frames average)

13 Dec, 2018, PIXEL2018

slide-9
SLIDE 9

9

55Fe Calibration

▪

55Fe used to calibrate the pixel gain on the assumptions:

13 Dec, 2018, PIXEL2018 𝑂𝑓−ℎ = 𝐹𝛽 𝜜 = 1640 ▪

55Fe generate two low energy X-ray:

  • 5.9 keV (90%)
  • 6.49 keV (10%)

▪ 5.9 keV X-ray produced electron- hole pairs:

1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 Events

Collection Peak =1131 =3429

a

k =3764

b

k

the charges with X-ray hitting

  • ther place disperse slowly

towards diode on thermal diffuse to neighbor pixel the charges with X-ray hitting on diode is complete conversion

slide-10
SLIDE 10

10

Simulation

13 Dec, 2018, PIXEL2018

Build Sentaurus TCAD model Extract electric field Interpolate on regular mesh

▪ 3D TCAD simulations are used to calculate the Electric Field Map ▪ Barycentric interpolation using nearest neighbors is used to calculate results on regular mesh in AllPix2 ▪ Monte Carlo sampling algorithm used for radioactive source

slide-11
SLIDE 11

1000 2000 3000 4000 5000 Signal [ADC] 500 1000 1500 2000 2500 3000 3500 4000 Events

Collection Peak = 1230 =3352

a

k =3676

b

k

11

Simulation vs Measurement

▪ TCAD + AllPix2 combined simulation managed to re-produce most of the features observed in measurements

1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 Events

Collection Peak =1131 =3429

a

k =3764

b

k

Measurement Simulation（TCAD+AllPix2） 13 Dec, 2018, PIXEL2018

slide-12
SLIDE 12

12

Diode Surface

▪ Lager diode surface -> more effective charge collection ▪ Lager capacitance -> more sensor noise

10 15 20 25 30 35 40 45 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW 10 15 20 25 30 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW 10 15 20 25 30 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW

A2 ENC 8.2 e- A3 ENC 10.4 e- A1 ENC 7.8 e-

1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 Events Entries = 2006421 54 ± = 3429

a

k 50 ± = 3764

b

k collection peak = 1131 1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 5000 10000 15000 20000 25000 Events Entries = 2382832 39 ± = 3038

a

k 46 ± = 3330

b

k collection peak = 1035 1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 5000 10000 15000 20000 25000 30000 Events Entries = 2303028 32 ± = 2198

a

k 36 ± = 2412

b

k collection peak = 796

CCE = 33.0% CVF = 31.9 μV/e A1 A2 A3 CCE = 34.1% CVF = 28.3 μV/e CC𝐹 = 36.2% C𝑊𝐺 = 20.5 𝜈V/𝑓 13 Dec, 2018, PIXEL2018

slide-13
SLIDE 13

13

Footprint

▪ Spacing: separation between diode and readout electronics

10 15 20 25 30 35 40 45 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40 45 COLUMN 2 4 6 8 10 12 14 16 ROW

A4 ENC 7.4 e- A7 ENC 7.5 e- A1 ENC 7.8 e-

1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 Events Entries = 2006421 54 ± = 3429

a

k 50 ± = 3764

b

k collection peak = 1131 1000 2000 3000 4000 5000 Seed Pixel Charge [ADC] 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 Events Entries = 1723746 51 ± = 3514

a

k 52 ± = 3864

b

k collection peak = 1054

A1 A4 A7 CCE = 33.0% CVF = 31.9 μV/e CCE = 30.0% CVF = 32.7 μV/e CCE = 28.6% CVF = 31.6 μV/e 13 Dec, 2018, PIXEL2018

slide-14
SLIDE 14

14

Cluster Charge Collection

▪ Cluster CCE=Cluster collection peak/Seed pixel calibration peak

  • Almost complete charge collection with 5x5 clusters
  • 3X3 cluster can collect most charges

5 10 15 20 25 Number of pixels in a cluster 20 40 60 80 100 120 CCE [%]

A1 A2 A3

5 10 15 20 25 Number of pixels in a cluster 20 40 60 80 100 120 CCE [%]

A1 A4 A7

Higher CCE with larger diode Higher CCE with larger footprint 13 Dec, 2018, PIXEL2018

slide-15
SLIDE 15

15

Tests with 90Sr

▪ Scintillator+ SiPM to provide the trigger signal ▪ Charged collected by the seed pixel estimated as the most probable value derived from the Landau function fit to the charge distribution

13 Dec, 2018, PIXEL2018

e- Scintillator CMOS sensor

90Sr

(collimated)

slide-16
SLIDE 16

16

Charge Collection

13 Dec, 2018, PIXEL2018

Sector Seed Charge [e-] Cluster Charge [e-] CCE S/N

A1 1498 3893 38.48% 237 A2 1624 3973 40.87% 229 A3 1673 3784 44.22% 180 A4 1391 3822 36.39% 234 A7 1361 3985 34.15% 220

A1 A2 A3

slide-17
SLIDE 17

17

Performance After Irradiation

▪ Samples sent to a pulsed neutron reactors and irradiated to fluences of 1012, 5x1012, and 1013 1 MeV neq/cm2 ▪ Larger diode (A3 >A1) more radiation hard as expected

13 Dec, 2018, PIXEL2018 A1 A3

slide-18
SLIDE 18

18

Performance After Irradiation

13 Dec, 2018, PIXEL2018

▪ Charge collection efficiency decreases but noise increases as the neutron fluence goes higher

slide-19
SLIDE 19

19

Tests with Electron Beams

▪ Sensor characterized with the DESY electron beam in September

  • Beam energy 1-6 GeV, beam size 1x1 cm2，data taken at 4.4GeV
  • EUDET beam telescope，spatial resolution 2~3μm at DUT

JadePix-1

Mimosa26

13 Dec, 2018, PIXEL2018

slide-20
SLIDE 20

20

Track Reconstruction

▪ Raw data converted to LCIO format using a customized EUDAQ version ▪ Sparse clustering to group pixels if they are within the defined distance ▪ General Broken Lines (GBL) algorithm to align reference planes and DUT

13 Dec, 2018, PIXEL2018

JadePix-1 region MIMOSA26 region

slide-21
SLIDE 21

21

Spatial Resolutions

▪ Spatial resolutions better than 5 μm and 3.5 μm achieved for pixel sizes

  • f 33x33 μm2 and 16x16 μm2

13 Dec, 2018, PIXEL2018

slide-22
SLIDE 22

22

Summary and Outlook

13 Dec, 2018, PIXEL2018

▪ Developed the first prototype JadePix-1 for the CEPC vertex detector. ▪ Sensors characterized with radioactive resources and the DESY electron test beam using a customized DAQ system; obtained useful information for future designs ▪ Performance evaluation of the irradiated samples still ongoing

This R&D project has been jointly supported by the State Key Laboratory of Nuclear Detection and Nuclear Electronics, the IHEP Innovation Fund and Yifang Wang's science studio.

Thanks for your attention!

slide-23
SLIDE 23

23 13 Dec, 2018, PIXEL2018 Test beam participants: Back: Hongbo Zhu, Chenfei Yang, Jia Tao Front: Liejian Chen, Xiaocong Ai, Ryuta Kiuchi Not in picture: Yi Liu, Shuo Han, Yanping Huang, Yifan Hu, Ying Zhang, Ke Wang

slide-24
SLIDE 24

24

Pixel Pitch

13 Dec, 2018, PIXEL2018

  • 4 μm2 diode surface, 20 μm2 footprint
  • Small pixel larger gain and cluster size
slide-25
SLIDE 25

25

Cluster Charge Distribution

13 Dec, 2018, PIXEL2018

slide-26
SLIDE 26

26

Cluster Size

5 10 15 20 25 30 S/N 1 2 3 4 5 6 7 Cluster size

A 1 A 1 A 2 A 2 A 3 A 3 A 4 A 4 A 5 A 5 A 6 A 6 A 7 A 7 A 8 A 8 A 9 A 9

5 10 15 20 25 Cluster size 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Normlized entries

t hreshol d 100 A D C t hreshol d 150 A D C t hreshol d 200 A D C t hreshol d 250 A D C t hreshol d 300 A D C t hreshol d 350 A D C t hreshol d 400 A D C t hreshol d 450 A D C t hreshol d 500 A D C

Fe-55, A1

5 10 15 20 25 Cluster size 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 Normlized entries

t hreshol d 100 A D C t hreshol d 150 A D C t hreshol d 200 A D C t hreshol d 250 A D C t hreshol d 300 A D C t hreshol d 350 A D C t hreshol d 400 A D C t hreshol d 450 A D C t hreshol d 500 A D C

Fe-55 Sr-90, A1 Sr-90 13 Dec, 2018, PIXEL2018