
N. Croitoru
1,2 , P. D’Angelo 1
, M.De Marchi
1
, P.G. Rancoita
1
, M.Rattaggi
1
, and A.Seidman
1,2
.
1 INFN – Istituto Nazionale di Fisica Nucleare, sezione di Milano – Via Celoria, 16, 20133 Milano, Italy.
2 Department of Physical Electronics, Tel-Aviv University, Ramat Aviv, 69 978, Israel.
S UMMARY
In recent years, intensive investigations of radiation hardness of BJT and MOS devices, integrated in BiCMOS technology, were done [1-6], in order to analyze the performance changes of the individual devices and to find better design strategies. In this paper, measurements of the radiation effects on the npn transistors (HF2CMOS process) are presented. The base current (I
B
), collector current (I
C forward gain (
) and
) in function of polarization, before and after irradiation with neutrons,

-rays and Carbon ions, were measured and analyzed. A correlation between the variations of

F

F
and the concentration of defects produced in the silicon bulk was found.
The I
B
, I
C
and

F
variations of npn transistors were measured for the fist time after 12 C-ions irradiation. As seen in neutrons and

-rays irradiated samples, I
B
increases with increasing ions fluences. Moreover, an increase of I
C observed for V
BE
< 0.5V, not seen on neutrons and

-rays irradiated samples. Also npn transistors with emitter area equal 40 times the minimum size were irradiated with ions. They show the same variations, at the same V
BE minimum size transistors. is
12 C
, seen for
The investigated devices were manufactured by ST-
Microelectronics, using an industrial standard high speed technology, called HF2CMOS. We analyzed minimum size and 40 times larger npn-transistors, in order to check also effects due to layout.
Devices were irradiated with neutrons,

-rays and Carbon ions. The neutron irradiation was made at the Triga reactor
RC:1 of the National Organization of Alternative Energy
Using this large set of data, we tried to correlate the variation of current gain to the atomic damage produced in
silicon. Taking from [7] the cross sections for Frenkel pair
formation for neutrons and

-rays interactions and using the simulation program TRIM for ions, the number of expected
Frenkel pairs per micron generated by the radiation was evaluated. This way, the fluence/dose of radiation can be expressed in term of the expected concentration of Frenkel pairs, regardless of the kind of particle which causes the damage. The variation of 1/

F
, in function of the number of
Frenkel-pairs/
 m, for collector current I
C
=1mA, is shown in
Fig. 1a. Different symbols were used in the figure, to distinguish between the different sorts of radiation, and the
1.0 x 10 13 n/cm 2 - 1.0 x 10 15 n/cm 2 . The irradiation with 1
Mrad and 5 Mrad doses of

-rays was done at the Calliope irradiated at the same fluence. In Fig. 1b, a different scale is facility of the National Organization of Alternative Energy chosen, to better show the region corresponding to low number of Frenkel-pairs/
 m. Two linear fits were performed:
(ENEA) at Casaccia, Rome, using a 60 Co source with activity the first (dotted line) using only the data corresponding to the of 300 kRad/h. The Carbon ions irradiation was made at the neutron irradiated devices, the second (dashed line) using all
Grand Accelerateur National d’Ions Lourds (GANIL), with
95MeV/a beam of fluences of the analyzed samples range from 5.3 x 10 up to 1.0 x 10
The npn devices were characterized by using a modular
DC source/monitor (Hewlett-Packard HP4142B), controlled by a work station HP9000/C160. The forward voltage applied to the emitter-base junction, V
0.2V < V
10 -12
BE
< 1.1V, which allow to measure the value of

A < I
C
13
12 C (ionized +6), at room temperature. The
C/cm 2 .
BE
10 C/cm 2
F
for
< 5 x 10 -3 A, the base and the collector were grounded. The measurements were done using the same setup, before and after the irradiation. All the measurements were performed at room temperature (T

25°C). the available data. The two fits agree within the errors. The variation of 1/

F
is proportional to the number of Frenkel pairs, regardless of the radiation which produced the damage. and
At fixed I

C
, the value of

(1/

F
F-irr
are the values of

) = 1/

F-irr
-1/

F0
(where

F0
F
before and after the irradiation respectively) is proportional at the increase of base current, I
B mainly due to the increase to generation current of the reverse
, was in the range
, obtained from investigations on the reverse current of diodes between the reverse current and the radiation fluence, for each kind of radiation, separately [8, 9]. This increase was attributed to generation processes, due to deep levels exposed to neutral and charged radiation, show a linearity introduced in the silicon gap by radiation damage and attributed to Di-Vacancy, Vacancy-Phosphorus and Vacancy-
As previously reported [1,3], a variation of I
B
, caused by the trapping of the minority carriers in generationrecombination centers introduced by irradiation, was observed for neutron and

-rays irradiated devices. Since I
C affected by irradiation, a reduction of

F
The variations of

F
(=I
C
/ I
B
is slightly
) is measured. are larger for low values of I
C
(or V
BE
), in correspondence to a large increase of I
B
.
Oxygen complexes, created by re-arrangement of primary defects (Frenkel pairs) inside the bulk [10-12]. Consequently, the concentration of the Frenkel pairs is responsible for the increase of the reverse current of base-collector junction and the variation of I
B
depends linearly on it.

The results presented in this work show that the same linear dependence is maintained for neutron,

-rays and 12 Cions irradiation. Moreover, if the variation of

F
can be understood in term of expected Frenkel pairs in silicon, we can assume that only generation processes in the bulk affect the behavior of irradiated npn transistors, at least for high collector currents. a) b)
Fig.1. a)

(1/

F
), measured for I
C
= 1mA, for neutrons,

-rays and 12 C irradiated devices in function of expected number of
Frenkel pairs per micron. The dotted line is a linear fit of

(1/

F
) measured on neutron irradiated devices, the dashed line is the fit of all the data.
b) Zoom of the plot (a) to better see the region corresponding to low Frenkel-pairs/
 m.

R
EFERENCES
[1] A.Baschirotto et al., “Radiation Damage Investigation for the Design of a Hardened Fast Bipolar Monolithic
Charge Sensitive Preamplifier”, Nucl. Instr. and Meth.
A 362 (1995) 466.
[2] A.Baschirotto et al., “ Radiation Damage Study of a
Fast Bipolar Monolithic Charge Sensitive
Preamplifier”, Nucl. Instr. and Meth. B 114 (1996)
327.
[3] A.Baschirotto et al., “Effects of Neutron Fluences up to 10 16 n/cm 2 and Gamma Doses up to 5Mrad on
Monolithic Fast Preamplifiers”, Nucl. Instr. and Meth.
B 122 (1997) 73.
[4] S.Caruso et al., “RHBIP1 Technology Evaluation to
Total Dose, Low Dose Rate and Neutrons, for LHC
Experiments and Space Applications”, presented to
RADECS97
[5] D.M.Fleetwood et al., “Physical Mechanisms
Contributing to Enhanced Bipolar Gain Degradation at
Low Dose Rates”, IEEE Trans. Nucl. Sci. NS-41(6)
(1994) 1871.
[6] A.H.Johnston et al., “Total Dose Effects in
Conventional Bipolar Transistors and Linear
Integrated Circuits”, IEEE Trans. Nucl. Sci. NS-41(6)
(1994) 2427.
[7] G.Dearnaley, ''Semiconductor counter for nuclear radiation" , E. & F.N. Spon LTD, London, 1966.
C.Caso et al., “Review of Particle Physics”, The
European Physical Journal C3 (1998) 1.
[8] E.Borchi et al., “Leakage current, annealing and deep defect production studies in neutron irradiated n-type
Si-detectors”, Nucl. Instr. and Meth. A 301 (1991),
215
[9] H.Ziock et al., IEEE Trans. Nucl. Sci. NS-38 (1991)
269.
[10] E.Borchi et al., “DLTS Measurements of Majority
Carriers Traps in Neutron Irradiated n-type Silicon
Detectors”, Nucl. Instr. and Meth. A 279 (1989) 277.
[11] M.Bosetti et al., “DLTS Measurements of Energetic
Levels, Generated in Silicon Detectors, Nucl. Instr. and Meth. A 361 (1995) 461.
[12] P.Mangiagalli et al., “Swift Heavy Ion Radiation
Damage on High Resistivity Silicon”, Nucl. Phys. B
(Proc. Suppl.) 61B (1998) 464.