CCE of heavily irradiated Si diodes operated with ... · • Measurements of CCE with DC light illumination were done with high light intensity (in “saturated” mode) ¾once the

I. Mandic, RD50 Workshop, CERN, May 5-7, 2004

1

CCE of heavily irradiated Si diodes operated with increased free carrier concentration

and under forward bias

Ljubljana group:

I. Mandić, M. Batič, V. Cindro, I. Dolenc, G. Kramberger, M. Mikuž, M. Zavrtanik


2

Motivation:

• LHC upgrade relevant fluences: 1016 n/cm2

• standard Si detectors operated in standard waywill be very difficult if not impossible to use

check the alternative way of operation with heavily irradiatedstandard detectors:

forward biasA. Chilingarov et al., NIM A399 (1997) 35.

increased carrier concentration with DC light illuminationV. Eremin et al., NIM A360 (1995) 458G. Lutz, NIM A377 (1996) 234.G. Kramberger et al., NIMA A497 (2003) 440.

cryogenic temperaturesV. Palmieri et al., NIM A413 (1998) 475.K. Borer et al., NIM A440 (2000) 5.

measure CCE with 90Sr source


3

Samples and Setup:

• 4 diodes: S = 0.5x0.5 cm2, D = 300 µm, ρ = 15 kΩcm

• irradiated with neutrons in TRIGA reactor in Ljubljana.Equivalent fluences: 5x1014, 1x1015, 4x1015 and 8x1015 n/cm2

• diodes were kept for 14 days at room temperature after irradiation.After that they were stored at T = -17°C

• CCE was measured on our TCT setup:• liquid nitrogen cryostat which allows illumination of diodes with laser• 90Sr source with set of collimators• trigger provided by scintilator + PMT (rate ~ 5 events/s)• MITEQ AM-1309 wide band current amplifier• digital oscilloscope • pulses were stored in computer and integrated in 18 ns time window offline• spectra of integrated pulses were fitted with convolution of Landau and Gaussian distribution

• collected charge measured with unirradiated diode used as normalizationto calculate CCE


4

Setup:

red lasercontinuousred lasercontinuous

optical cryostatJanis VPF−100

attenuator

beam positioner

red light (DC)

Sr 90 source

lead

block

PM tube

block

LeadScintilator


5

Reverse bias, DC illumination of n side with red light (hole injection):

• injection of holes into irradiated detector influences theoccupation probability of deep levels:

it reduces the concentration of negative space charge (trapping of holes)lower full depletion voltage effective space charge sign can be reinverted to positive

More in: G. Kramberger et al., Nucl. Instr. and Meth. A497 (2003) 440.

• Measurements of CCE with DC light illumination were done with high light intensity (in “saturated” mode)

once the light intensity is high enough high field region moves from n-side top-side (the space charge sign is re-inverted).

increasing the light intensity beyond this point doesn’t change the spacecharge much, because the e-h pairs created in the low field region recombine

Current vs. light intensity curve saturates

CCE doesn’t depend much on light intensity once it is high enough


6

CCE vs. bias voltage for standard operation, DC light (hole injection) and forward biasT=-30˚C Highest CCE at given voltage measured with forward bias!

Voltage (V)0 50 100 150 200 250 300

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1(a)

2A/cmµNo DC light, j = 4.4

2A/cmµDC light, j = 100

2A/cmµForward bias, j = 680

14Fluence 5x10

Voltage (V)0 50 100 150 200 250 300

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1(b)

2A/cmµNo DC light, j = 6.0



14Fluence 10x10

Voltage (V)0 50 100 150 200 250 300

CC

E

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

(c)

2A/cmµNo DC light, j = 10



14Fluence 40x10

Voltage (V)0 50 100 150 200 250 300

CC

E

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

(d)

2A/cmµNo DC light, j = 18



14Fluence 80x10


7

Current measurements at different temperatures and modes of operation

Temperature (K)100 120 140 160 180 200 220 240

Cu

rren

t (A

)

-910

-810

-710

-610

-510

-410(a) Forward Bias

DC holesNormal operation

2 n/cm14Fluence: 5x10

Uforward= 80 V Ureverse = 250 V

Temperature (K)100 120 140 160 180 200 220 240

Cu

rren

t (A

)

-910

-810

-710

-610

-510

-410

(b)



Temperature (K)100 120 140 160 180 200 220 240

Cu

rren

t (A

)

-910

-810

-710

-610

-510

-410

(c)



Temperature (K)100 120 140 160 180 200 220 240

Cu

rren

t (A

)

-910

-810

-710

-610

-510

-410

(d)




8

• CCE vs. temperature

Temperature (K)100 120 140 160 180 200 220 240

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(a)

Forward BiasDC holesStandard operationDC electrons

2n/cm14Fluence: 5x10

Uforward = 80 VUreverse = 250 V

Temperature (K)100 120 140 160 180 200 220 240

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(b)



Temperature (K)100 120 140 160 180 200 220 240

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(c)



Temperature (K)100 120 140 160 180 200 220 240

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

(d)




9

• CCE under forward bias doesn’t change much with temperature

• Hole injection:transition from beneficial to harmful with falling T at low fluence

• Standard mode or electron injection:improvement with falling T

both transitions occur between 180 and 220 K

These results agree with Space Charge Sign Inversion occurringin this temperature range.Temperature of transition agrees with temperatures ofSCSI and CCE transitions reported in the RD39 paper: E. Verbitskaya et al., NIM A514 (2003) 47-61.


10

Attempt of CCE calculation

• two dominant deep traps:

300x10-151x10-150.070.52donor

2x10-1515x10-1510.57acceptor

σh (cm2)σe (cm2)gt (cm-1)Et - Ev (eV)

• parameters of the traps chosen from parameterintervals of dominant electron and hole trap from the paper:G. Kramberger et al., “Determination of the effective electronand hole trap in neutron-irradiated silicon detector” NIM A516 (2004) 109


11

• effective space charge and trapping times were calculated using equations for defects in general stationary state:

)exp(,1/

1

TkEE

ncncncpc

Pb

itt

tipn

tinpt

−=

+

+

+=

−

χχχ

• contribution of the trap to the space charge is given by:

)(0 teqtt PkgeQ −Φ= k = 0,1 (donor, acceptor)

• effective trapping times for electrons and holes are obtained from:

eqhthhthhh

eqeteethee

gPv

gPv

Φ=

Φ−=

στ

στ

/1)1(/1


12

• Free carrier concentration was estimated from current measurements.In the case of standard and forward bias operation it was assumed (simplification!) that electrons and holes contribute equally to the current. Therefore concentration of electrons and holes was estimatedas:

evSIn⋅⋅

=2 hvS

Ip⋅⋅

=2

• In the case of DC light illumination the additional carrier densitywas estimated from measurement of current increase, taking intoaccount the surface of the illuminated area and the side of illumination.


13

Forward bias:

• electric field in forward bias approximated with E = Vbias/D in the whole detectorvolume

• at high fluences forward bias I-V dependence is almost ohmic and one can calculate resistance of the material as R = Vbias/I

)-1 Aµ1/I (0.05 0.1 0.15 0.2 0.25 0.3

U/I

(MO

hm

)

2

4

6

8

10

12

14

2 n/cm158x10

2 n/cm154x10

2 n/cm151x10

2 n/cm145x10

Forward Bias T = -30 C

Resistance vs. 1/I in forward bias


14

Resistance at high fluence very high: 100 MΩcm at 8x1014 n/cm2

also in reverse bias there should be significant voltage dropand therefore electric field in the undepleted region because of current flow: E = ρ ·j, (ρ = resistivity, j = current density)

“undepleted” bulk should contribute to CCE

• in reverse bias: Ubias = Ujunction + I·R, (R – resistance, I – current)

• in the junction w=√(2εε0Ujunction/e0Neff) el. field is Ej ~ Neff·x

• total field E = Ej + j ·ρ inside the junction and E = j ·ρ elsewhere


15

CCE vs bias voltage at T = -30CFull markers: measurements, empty markers: calculation

Voltage0 50 100 150 200 250 300

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

, T = -30 C2 n/cm15 = 1x10Φ

Voltage0 50 100 150 200 250 300

CC

E

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

, T = -30 C2 n/cm15 = 4x10Φ

Voltage0 50 100 150 200 250 300

CC

E

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

, T = -30 C2 n/cm15 = 8x10ΦVoltage

0 50 100 150 200 250 300

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

DC holesForward bias

Normal operation

, T = -30 C2 n/cm14 = 5x10Φ


16

Conclusions:

• DC hole injection and forward bias operation greatly improve CCEat moderate voltages (up to 250 V) already at moderate cooling (-30°C)

• at given voltage, highest CCE achieved by operating under forward bias

• if operated at low temperature (below ~ 200 K) the problem ofhigh current in forward bias is much smaller while CCE stays constant

• increase of CCE seems to be caused by increase of sensitive depth of the detector and to lesser extent by change of trapping times

• trapping is the main limitation for using standard Si detectors at high fluences Φeq > 1015 n/cm2


17

Examples of signals:

• averages of 5000 events:

Time (ns)0 2 4 6 8 10 12 14 16 18

Ind

uce

d C

urr

ent

(arb

)

-0.03

-0.02

-0.01

0

0.01

0.02

0.03Reverse Bias 250 V, DC hole injection

Forward Bias 120 V

Co, T = -302 n/cm15 10⋅ = 1 Φ


18

• spectra of integrated pulses were fitted with convolutionof Landau and Gaussian distribution. The most probable value of the Landau distribution returned by the fit was taken as the measure of collectedcharge.

• collected charge measured with unirradiated diode was used as normalization to calculate CCE

Examples of fitted spectra:

Charge (arb)-0.1 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3

Nb

. of

entr

ies

0

50

100

150

200

250

300

350 = 0Φ

2 n/cm15 10⋅ = 1 Φ

, DC holes2 n/cm15 10⋅ = 1 Φ

CoBias = 250 V, T = -30


19

CCE doesn’t depend much on light intensity once it is high enough

Light Power (arb.)0 50 100 150 200 250 300 350 400 450 500

A)

µD

etec

tor

curr

ent

incr

ease

(

0

2

4

6

8

10

12

14Not Irradiated

2 n/cm14 = 5x10Φ2 n/cm15 = 1x10Φ2 n/cm15 = 4x10Φ2 n/cm15 = 8x10Φ

CoVoltage = 250 V, T = -30

Measurements of CCE with DC light illumination were done with high light intensity (in “saturated” regime)


20

• calculate CCE vs. fluence at -30°Cbest agreement for DC light injection and forward bias if in bothconstant field E = V/D assumed in the whole detector at all fluencesso that CCE is a function of trapping only

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Measurements (markers):

Calculations (lines):

DC hole injection

Full depletion assumed

(a)

Bias = 250 V T = -30 C

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

CC

E0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Measured

Calculated

(b)

Forward bias = 80 V, T = -30 C

Reverse bias +DC light Forward bias

Measured and calculated CCE


21

Ohmic voltage drop (I·R) and voltage drop in the junction (Vj = Vbias - I·R)

DC hole injection: Standard operation:

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

Vo

ltae

Dro

p (

V)

0

50

100

150

200

250DC hole injection

Voltage drop in the pn junction

Ohmic voltage drop

(a)

DC hole injection, Bias = 250 V T = -30 C

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

Vo

ltae

Dro

p (

V)

0

50

100

150

200

250

Standard operation

Voltage drop in the pn junction

Ohmic voltage drop

(b)

Standard operation, Bias = 250 V T = -30 C


22

Different calculations of CCE compared to measured CCE:

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

CC

E

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

DC hole injection

Measurements

Calculations (lines + empty markers): Full depletion assumed

Ohmic Voltage dropStandard

(b)

DC hole injection, Bias = 250 V T = -30 C

)2 n/cm14Fluence (100 10 20 30 40 50 60 70 80 90 100

CC

E

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

Standard operation

Measurements

Calculations (lines + empty markers):

Ohmic Voltage drop

Standard

(c)

Standard operation, Bias = 250 V T = -30 C

Disagreement smaller if ohmic voltage drop is taken into account.

Documents

CCE of heavily irradiated Si diodes operated with ... · • Measurements of CCE with DC light illumination were done with high light intensity (in “saturated” mode) ¾once the