Development of LGAD Sensors at FBK - A. Bisht, G. Borghi, M. Boscardin, M. Centis Vignali1, F. Ficorella, O. Hammad Ali, G. Paternoster Fondazione ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Development of LGAD Sensors at FBK
A. Bisht, G. Borghi, M. Boscardin, M. Centis Vignali1 ,
F. Ficorella, O. Hammad Ali, G. Paternoster
Fondazione Bruno Kessler
25.02.2022
16th Vienna Conference on Instrumentation
1 mcentisvignali@fbk.eu
Fondazione Bruno Kessler
LGAD technologies: Start Batch
Standard 2015 UFSD1
2017 UFSD2
Double sided (inverted)
2018 UFSD3
AC coupled (RSD) RSD1
Trench isolated 2019 HD0
UFSD3.2
MOVEIT
TI-LGAD RD50
2020
PSI iLGAD
HADES
RSD2
2021 UFSD4
Space LGADs
6 inch (150 mm) Today: selected results and projects
Custom CMOS-like process
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 2
(Standard) Low Gain Avalanche Diodes
Silicon detectors with charge multiplication Improve SNR of the system
Gain ≈ 10 (When the sensor shot noise is not
Gain layer provides high-field region dominating)
No-gain region ∼ 30 − 80 µm Noise and power consumption
Time resolution ∼ 30 ps ↔ thin ∼ 50 µm sensor ⇒ low gain
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 3
High Luminosity LHC
Development within the UFSD project
[ATLAS simulation]
Use time coordinate to mitigate pile-up
Track time resolution ≈ 30 ps
Radiation resistance to few 1015 neq /cm2
Hit time resolution at end of life ≈ 50 ps
[H. Sadrozinski et al. Rept. Prog. Phys. 81 (2018) 026101]
Application described in:
Daniel Spitzbart talk on Wednesday
Frank Filthaut talk on Wednesday
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 4
Radiation Hardening of LGADs
Gain layer doping
[M. Ferrero TREDI2021] [M. Moll PoS Vertex2019 (2020) 027]
Acceptor removal:
Sii + Bs → Bi
Bi + Oi → Bi Oi (donor level)
Carbon ⇒ Competing reaction:
Sii + Cs → Ci
Ci + Oi → Ci Oi (neutral)
Initial B concentration
→ higher concentration favored
→ narrower B distribution
Carbon coimplantation
NB (φeq ) = NB (0) exp {−cφeq } → optimized dose found
c = c(NB (0))
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 5
Radiation Hardening of LGADs
Removal constant
[M. Ferrero TREDI2021] [M. Moll PoS Vertex2019 (2020) 027]
Acceptor removal:
Sii + Bs → Bi
Bi + Oi → Bi Oi (donor level)
Carbon ⇒ Competing reaction:
Sii + Cs → Ci
Ci + Oi → Ci Oi (neutral)
Initial B concentration
→ higher concentration favored
→ narrower B distribution
Carbon coimplantation
NB (φeq ) = NB (0) exp {−cφeq } → optimized dose found
c = c(NB (0))
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 5
Radiation Hardening of LGADs
[G. Paternoster VERTEX2021]
[M. Moll PoS Vertex2019 (2020) 027]
Acceptor removal:
Sii + Bs → Bi
Bi + Oi → Bi Oi (donor level)
Carbon ⇒ Competing reaction:
Sii + Cs → Ci
Ci + Oi → Ci Oi (neutral)
Initial B concentration
Gain layer position: → higher concentration favored
→ narrower B distribution
“shallow” → higher B concentration
Carbon coimplantation
“deep” → easier compensation of B loss by increasing
→ optimized dose found
bias
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 5
Radiation Hardness Results
[M. Ferrero Vertex2021]
[R. Arcidiacono et al. NIMA 978 (2020) 164375]
Time resolution < 40 ps for 2.5 · 1015 neq cm−2
Time resolution < 50 ps for 3 · 1015 neq cm−2
Demonstrated radiation resistance and time resolution for HL-LHC
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 6
Radiation Hardness Results
[A. Howard 37th RD50 Workshop]
[R. Arcidiacono et al. NIMA 978 (2020) 164375]
Time resolution < 40 ps for 2.5 · 1015 neq cm−2
Time resolution < 50 ps for 3 · 1015 neq cm−2
Demonstrated radiation resistance and time resolution for HL-LHC
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 6
UFSD4
ATLAS CMS
Both “shallow” and “deep”
gain layers
Different pad layouts
Sensors up to ∼ 2 × 2 cm2
Wafers and sensors for qualification for ATLAS and CMS timing detectors
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 7Single Event Burnout SEB
[N. Cartiglia 39th RD50 Workshop]
Thin irradiated PiN and LGADs at high bias
E > 11 V/µm
All tested producers affected
⇒ “fundamental” property
Catastrophic failure due to dense energy deposition
(hard collision)
Observed with:
[G. Medin 38th RD50 Workshop]
lasers [G. Medin 38th RD50 Workshop] [G. Medin 38th RD50 Workshop]
beam tests [R. Heller 38th RD50 workshop]
Mitigation:
further hardening the gain layer
thicker substrates
quenching resistors [N. Cartiglia 39th RD50 Workshop]
Further details: Gordana Medin recorded presentation
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 8HADES Experiment
[R. Holzmann 54. Winter Meeting on Nuclear Physics]
Fixed target experiment at GSI
TOF used for particle identification
(among other methods)
T0 detector
Based on diamond detectors
Beam monitoring
TOF start
Replace diamond with LGADs
[J. Pietraszko et al. Eur. Phys. J. A 56 (2020) 183]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 9LGADs for HADES
Strip geometries
Sensor dimension up to ∼ 2 × 2 cm2
Different active thicknesses: 35, 55, 85 µm
Wafers thinned down to 200 µm total
Dicing after thinning
First results:
Wilhelm Krueger talk on Monday
[W. Krueger VCI 2022]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 10Hadron Therapy
[A. Vignati Picosecond Workshop 2018] Development within the MoVeIT project
Beam monitoring during treatment
Quality assurance
Energy (TOF measurement)
Particle count
Beam profile
Sensor requirements similar to HEP but TOF
measurement using several particles
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 11MoVeIT Energy Measurements
[A. Vignati PSD12]
Results from LGAD strip sensors, 590 µm pitch
Short signal duration → rate capabilities
Time resolution → energy resolution
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 12MoVeIT Energy Measurements
[A. Vignati PSD12]
Results from LGAD strip sensors, 590 µm pitch
Short signal duration → rate capabilities
Time resolution → energy resolution
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 12LGADs for MoVeIT
Strip geometries
Particle counting sensors ∼ 2.7 × 2.7 cm2
Energy meas. sensors ∼ 6.5 × 4 mm2
First characterization of the particle counting
sensors: [O. Marti Villarreal PSD12]
[A. Vignati PSD12]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 13Space Applications
Timing in space experiments
[AMS]
Particle ID using TOF
Identification of incoming or outgoing
particles
Identification of calorimetric showers and
backsplash
σt ∼ 50 − 100 ps
Particle rate + limited power
⇒ channel size ∼ 1 cm2
→ sensor capacitance
Timing with LGADs?
Project in collaboration with
INFN Perugia and Torino
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 14Space Applications
[M. Duranti et al. Instruments 2021, 5, 20] Timing in space experiments
Particle ID using TOF
Identification of incoming or outgoing
particles
Identification of calorimetric showers and
backsplash
σt ∼ 50 − 100 ps
Particle rate + limited power
⇒ channel size ∼ 1 cm2
→ sensor capacitance
Timing with LGADs?
Project in collaboration with
INFN Perugia and Torino
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 14Thickness and Gain Optimization, Jitter Simulation
LTspice simulation
Sensor capacitance
Uniform charge deposition
(no Landau noise)
Simplified gain layer model
Assumed saturated velocities
Noise from amplifier and sensor
Simple model favors thickness > 100 µm and gain > 100
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 15Thickness and Gain Optimization, Jitter Simulation
LTspice simulation
Sensor capacitance
Uniform charge deposition
(no Landau noise)
Simplified gain layer model
Assumed saturated velocities
Noise from amplifier and sensor
Simple model favors thickness > 100 µm and gain > 100
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 15Thickness and Gain Optimization, Jitter Simulation
LTspice simulation
Sensor capacitance
Uniform charge deposition
(no Landau noise)
Simplified gain layer model
Assumed saturated velocities
Noise from amplifier and sensor
Simple model favors thickness > 100 µm and gain > 100
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 15Space LGADs Production
In fabrication
Active thickness: 50, 100, 150 µm
Strip sensors for daisy chains
Single pads up to 1 cm2
Target gain of 100 to 200
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 16Double Sided (Inverted) LGADs
Continuous gain area in the active region ⇒ 100% fill factor
Double sided process → active thickness is the wafer thickness ⇒ not optimal for timing
Readout side is ohmic
Readout side separated from LGAD side ⇒ no restrictions on channel dimensions
[G.F. Dalla Betta et al. NIM A 796 (2015) 154]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 17X-ray Detection
Development in collaboration with PSI
[Wikipedia CC BY-SA 2.0 ]
Detection of soft X-rays: 250 eV - 2 keV
K-edges of bio elements
→ pharmaceuticals, cell imaging
L-edges of 3d-transition metals
→ magnets, superconductors, quantum
materials ...
Use LGADs:
Gain to lower the detection limit of photon
counting detectors
Advantages of LGADs demonstrated in:
Gain to improve SNR of integrating detectors
[Andrae et al. J.Synchrotron Rad. 26 (2019)
1226-1237] Thin entrance window and gain structure
must be developed
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 18Double Sided LGADs for PSI
Gain with LED (STD gain structure)
10 GT_01
GT_02
GT_03
8 GT_04
GT_05
6
Gain
4
2
0 0 50 100 150 200 250 300
Bias [V]
Several pixel and strip geometries
Thin entrance window
Several gain structure designs → make as thin as possible
Thickness 275 µm
First results with x-rays at TREDI next week
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 19Double Sided LGADs for PSI
Gain with LED (STD gain structure)
10 GT_01
GT_02
GT_03
8 GT_04
GT_05
6
Gain
4
2
0 0 50 100 150 200 250 300
Bias [V]
Several pixel and strip geometries
Thin entrance window
Several gain structure designs → make as thin as possible
Thickness 275 µm
First results with x-rays at TREDI next week
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 19AC Coupled LGADs (RSD)
Continuous gain area in the active region ⇒ 100% fill factor
Readout channels capacitively coupled and resistive layer to limit signal spreading
No restrictions on channel dimension
[M. Mandurrino et al. IEEE EDL Vol 40 Issue 11 (2019) 1780-1783]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 20AC Coupled LGAD Productions
RSD1 RSD2
Several pixel and strip
geometries
Electrode geometries to
exploit signal propagation
Variations of resistive layer
Variations of coupling
dielectrics
Project in collaboration with INFN Torino
Results in Federico Siviero talk (this session)
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 21Trench Isolated LGADs
Trenches substitute the isolation structures
Trench width about 1 µm ⇒ fill factor close to 100%
[G. Paternoster et al. IEEE EDL Vol 41 Issue 6 (2020) 884-887]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 22RD50 TI-LGADs
55 µm Pixel 50-100 µm Strip
Second TI-LGAD run
Project within the RD50 collaboration
Several pixel and strip geometries
Different gain structure layouts
Variations in trench depth and fabrication process
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 23First Characterization
[A. Bisht Picosecond Workshop 2021]
Stable trench structures
Breakdown due to gain layer
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 24First Characterization [A. Bisht Picosecond Workshop 2021]
Interpad 3-10 µm with laser
∼ 10× improvement from STD LGAD
Some process and layout combination:
→ gain in trench region > gain of pad
→ negative interpad
More characterization and irradiation results:
Matias Senger talk (next)
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 25First Characterization [A. Bisht Picosecond Workshop 2021]
Interpad 3-10 µm with laser
∼ 10× improvement from STD LGAD
Some process and layout combination:
→ gain in trench region > gain of pad
→ negative interpad
More characterization and irradiation results:
Matias Senger talk (next)
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 25Upcoming Projects
EXFLU
Silicon sensors for extreme fluences
Variation of thickness and gain layer designs
Valentina Sola talk on Thursday
[N. Cartiglia 39th RD50 Workshop]
DC Resistive Silicon Detectors
Position interpolation through resistive charge division
Improved spatial resolution with large readout pitch
Luca Menzio recorded presentation
[N. Cartiglia 39th RD50 Workshop]
SEB resistant LGADs
AC-coupled readout
Quenching resistors to protect the sensor
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 26Thank you for your attention M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 27
Backup Material M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 1
[G. Paternoster et al. NIMA 987 (2021) 164840] M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 2
Segmentation: Fill Factor
Focused 20 keV x-ray beam
Measured FF: ≈ 40%
Impact on detection efficiency
Signal vs position for 3 strips
[M. Andrae, J. Zhang, et al. J. Synchrotron Rad. (2019)]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 3Interpad Distance TI-LGADs
[A. Bisht Picosecond Workshop 2021]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 4Low Energy X-ray Detection
Photon counting strip detectors, fluorescence X-rays
PiN sensor LGAD sensor
E > 8 keV visible E < 3.3 keV visible
Improvement in detection threshold
[A. Bergamaschi TREDI2019]
[M. Andrae, J. Zhang, et al. J. Synchrotron Rad. (2019)]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 5SEB Resistant LGADs
[N. Cartiglia 39th RD50 Workshop]
M. Centis Vignali Development of LGAD Sensors at FBK 25.02.2022 6You can also read
