Radiation hardness simulation of silicon thin detectors · 2004. 5. 7. · *F. Moscatelli, et al.Nuclear Instruments and Methods in Physics Research B 186 (2002) Università degli

Università degli Studi Università degli Studi di Perugiadi Perugia

Radiation hardness simulation of silicon thin detectors

F.Moscatelli1,2 , M. Petasecca1 , G.U. Pignatel1

1 DIEI - Università di Perugia, via G.Duranti,93 - Italy 2IMM-CNR sez.di Bologna, via Gobetti 101 - Italy

Università degli Studi Università degli Studi di Perugiadi PerugiaOutline

• Radiation damage modelling (F. Moscatelli)

• Thin structures (M. Petasecca)

Università degli Studi Università degli Studi di Perugiadi PerugiaSimulation Tool

• Simulation tool:–ISE-TCAD – discrete time and spatial solutions to equations

• Damage modelling:- Deep levels: Nt , Et, σn and σp

- SRH statistics.- Donor removal mechanism- Other effects: high density defect concentration (clusters) produces anincrease of the leakage current.

Università degli Studi Università degli Studi di Perugiadi PerugiaRadiation Damage ModelRadiation Damage Model

•• Four levels*:Four levels*:

-- VV22--/0/0 located at E=Elocated at E=ECC--0.420.42

-- CCiiOOii located at E=Elocated at E=EVV+0.36+0.36

-- VV22OO located at E= Elocated at E= ECC--0.500.50

-- E(70)E(70) located at E=Elocated at E=ECC--0.450.45

•• Direct Direct chargecharge exchangeexchange betweenbetween VV22--/0/0 and E(70) to reproduce cluster effectsand E(70) to reproduce cluster effects..

•• Donor removal mechanism.Donor removal mechanism.

•• Reproduce variation of the Reproduce variation of the VVdepdep and and IIleakageleakage as a function of the fluenceas a function of the fluence

•• Over Over Φ=2×1014 n/cm2 computational problems

*F. Moscatelli, et al.Nuclear Instruments and Methods in Physics Research B 186 (2002)

Università degli Studi Università degli Studi di Perugiadi PerugiaISE-TCAD Damage Model

•• Three levelsThree levels**::

-- VV22--/0/0 located at E=Elocated at E=ECC--0.420.42

-- CCiiOOii located at E=Elocated at E=EVV+0.36+0.36

-- VV22OO located at E= Elocated at E= ECC--0.500.50

•• TTo reproduce cluster effectso reproduce cluster effects, we use, we use increased increased VV22--/0/0 occupancyoccupancy

•• Donor removal mechanismDonor removal mechanism..

*D. Passeri, P. Ciampolini, G. Bilei and F. Moscatelli, IEEE Trans. Nucl. Sci., vol. 48, pp. 1688, 2001.

Università degli Studi Università degli Studi di Perugiadi PerugiaThree-level model

1010--1616 cmcm221010--1515 cmcm228 8 ⋅⋅1010--1515cmcm22σσpp

1 cm1 cm--110.1 cm0.1 cm--1126 cm26 cm--11ηη

1010--1515 cmcm221010--1616 cmcm221010--16 16 cmcm22σσnn

EEvv+0.36eV+0.36eVEEcc --00.50eV.50eVEEcc --0.42eV0.42eVEE

CiOiVV22OO

• Level characteristics:

VV22--/0/0


Effective Doping Concentrationand Leakage Current

∆∆I/VI/Volumeolume = = αΦαΦ, with , with αα =(2.9=(2.9÷÷10) 1010) 10--1717A/cm.A/cm.


Simulations results as a function of T

100 150 200 250 3000

3

6

9

12

15

V dep [V

]

Temperature [K]

measurements simulations

Φ=2.2×1013

protons/cm2

100 150 200 250 300

20

40

60

80

100

120

140

Vde

p [V

]

Temperature [K]

measurements simulations Φ=4.7×1014

protons/cm2

180 200 220 240 260 280

1E-15

1E-14

1E-13

σ p V2 [c

m2 ]

Temperature [K]

180 200 220 240 260 280

1E-15

1E-14

σ pV2O

[cm

2 ]

Temperature [K]

Università degli Studi Università degli Studi di Perugiadi PerugiaThin detectors

• Thin detectors have been proposed to investigate the possibility to get a low depletion voltage and to limit the leakage current of heavily irradiated silicon devices


Simulation setup

Guard Ring Diode Guard Ring

6µm40µm15µm9µmD

Back

Simulated structures:

=m

mD

µµ

30058

• Simulated device structure and parameters

• Doping profiles: – n-doped substrate (7×1011 cm-3)

6kΩcm.– Charge concentration at the silicon-

oxide interface of :• 4 ×1011 cm-3 pre-irradiation• 1 ×1012 cm-3 post-irradiation

thin devicethick device

Università degli Studi Università degli Studi di Perugiadi PerugiaSimulation setup

Guard RingDiode- Variable mesh definition:

the mesh is better refined in correspondence of the critical points of the device to improve simulator performance.

- The typical electric field distribution at the full depletion voltage of the diodes: red areas correspond to the maximum


Simulation results

Simulated Depletion Voltage as function of the fluence

0.0 1.0x1014 2.0x10140

100

200

300

400

Vde

p [V

]

Fluence [(1MeV) n/cm2]

300 µm

0.0 1.0x1014 2.0x1014

0

4

8

12

16

Vde

p [V]

Fluence [(1MeV) n/cm2]

58 µm

- Vdep in thin structures is one order of magnitude lower than in thick one- Vdep of thin diode at a fluence of 1×1015 n/cm2 is about 120 V while in thick diode is more than 3000 V !


CCE Simulation results

MIP: 80 e-h pairs/ µmCylinder diameter= 2µm

dttIQ ∫= )(
















Simulated CCE as function of the fluence

Exp.CCE = 20-30%CCE = 95%CCE = 27%

Q@120V= 0.57 fCQ@VBD=300V= 1 fCΦ=1e15 n/cm2

Exp. [1, 2] CCE = 42%

CCE = 98%CCE = 45%

Q@14V= 0.58 fCQ@300V= 1.6 fCΦ=2e14 n/cm2

ThinThick

For Φ=1e15 n/cm2

Qthin=57% QthickThin @ 120V NO Breakdown risk and full depleted

[1] L.Beattie et al./ NIM 412A (98)

[2] M.Bruzzi et al./ NIM 61B(98)

Università degli Studi Università degli Studi di Perugiadi PerugiaConclusions

• Irradiated thin and thick diodes have been analyzed considering a three levels simulation model until Φ=1e15 n/cm2

• Thin features:

– Vdep in thin structures is one order of magnitude lower

– CCE at very high fluence (1015 n/cm2) is 95% for thin structures.

– Qthin=57% Qthick

• Next step is to simulate thin structures at higher fluences (1e16 n/cm2) and measure irradiated ones.

Documents

Radiation hardness simulation of silicon thin detectors · 2004. 5. 7. · *F. Moscatelli, et al.Nuclear Instruments and Methods in Physics Research B 186 (2002) Università degli