Leroy C., Rancoita P.-G. Principles Of Radiation Interaction In Matter And Detection

Подождите немного. Документ загружается.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

290 Principles of Radiation Interaction in Matter and Detection

Fig. 4.1 1 MeV neutron equivalent ﬂuence (1 MeV n

/cm

/yr) in the ATLAS calorimetry system

and vicinity, integrated over a standard high luminosity year; Z (in cm) is the direction along the

beam axis and R (in cm) the radius around the Z-direction (from [Leroy and Rancoita (2007)]; see

also [ATLAS Collab. (1996)]).

Fig. 4.2 Yearly integrated dose (Gy/yr) for photons above 30 keV in the ATLAS calorimetry

system and vicinity for a standard high luminosity year; Z (in cm) is the direction along the

beam axis and R (in cm) the radius around the Z-direction (from [Leroy and Rancoita (2007)]; see

also [ATLAS Collab. (1996)]).

particles are absorbed. Then, one expects that charged hadrons (mainly protons,

pions) will dominate the radiation environment in regions close to the collision

point at least for small depth of traversed material (like for the LHC-ATLAS inner

detector partly equipped with silicon detectors). Particles backscattered from the

calorimeters and cascades leaking out the calorimeters produce neutrons, photons

and electrons through their interactions with surrounding material and equipment

located in the experimental hall. These particles also aﬀect the region around the

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

Radiation Environments and Damage in Silicon Semiconductors 291

collision point, although, neutrons in that region can be moderated by the pres-

ence of low-Z material [walls or low-Z planes of inner detectors with Z the atomic

numb er of the material (Sect. 3.1)]. While electromagnetic cascades are rapidly ab-

sorbed, neutrons will travel long distance losing their energy gradually. Therefore,

the neutron ﬂuence dominates the radiation environment at larger radii. Nuclear

capture of thermal neutrons frequently results into production of photons. Large

amounts of photons also result from excited-state decay of spallation products and

from fast neutron interactions with atomic nuclei. Photons, produced deep inside

the material, are rapidly absorbed. Only photons, produced close to the surface of

material, have a chance to produce electrons and p ositrons and are responsible for

a low-energy electron component in the environment.

As the showering process takes a relatively long time, this mixed ﬁeld of neu-

trons, photons and electrons constitutes a background with no time structure [AT-

LAS Collab. (1994b)]. The numerous scatterings of neutrons and photons before

their capture yield a uniform and isotropic ﬁeld of low-energy background particles.

The radiation level induced on detector material and equipment will depend

on the position of the detector with respect to the collision point. This location is

expressed in terms of the pseudorapidity, η, deﬁned in Eq. (3.61) and expressed as a

function of θ, which is the polar production angle with respect to the beam axis. For

experiments to be performed at LHC, the highest radiation levels are expected to oc-

cur at high pseudorapidity. As stressed above, the level of radiation also depends on

the amount of surrounding materials. In ATLAS example [ATLAS Collab. (1994b,

1996)], the forward calorimeter (FCAL) is covering the range 3.2 ≤ η ≤ 5. Due

to its location, the FCAL is subjected to radiation dose-rates up 10

Gy yr

−1

and

a neutron ﬂux (neutron kinetic energy > 100 keV) up to 1 × 10

−2

−1

. Repre-

senting a depth of 9.5 interaction lengths of material, the FCAL contributes to the

increase of the radiation level in the hadronic end-cap (HEC) calorimeter due to its

location (1.5 ≤ η ≤ 3.2) very close to FCAL. The HEC will be subjected to radia-

tion dose up to 0.3 kGy and a neutron ﬂuence of 0.3 × 10

−2

over 10 years of

operation at the highest luminosity (1 ×10

−2

−1

). It is standard to use 1 MeV

equivalent neutron ﬂuence to express neutral particle contribution to radiation. The

1 MeV equivalent neutron ﬂuence is the ﬂuence of 1 MeV neutrons producing the

same damage

∗

in a detector material as induced by an arbitrary ﬂuence with a

speciﬁc energy distribution [see Eqs. (4.79, 4.85), Sects. 4.1.3 and 4.2.1]. The di-

splacement damages induced by neutrons in semiconductors can be normalized by

displacement damages induced by 1 MeV neutrons through the hardness parame-

ter [see Eqs. (4.79, 4.84), Sects. 4.1.3 and 4.2.1]. The 1 MeV neutron ﬂuence and

yearly integrated dose (photons above 30 keV) in the ATLAS calorimetry system

∗

The amount of the induced displacement damage (i.e., the deposited damage energy, see

Sect. 4.2.1) can be deﬁned for any kind of incident particle. However, it cannot be assumed to

relate a damage equivalence until the correlation between measured device degradation and the

calculate NIEL deposition is determined (see discussion in Sects. 4.2.1–4.2.1.5 and, also, [Griﬃn,

Vehar, Cooper and King (2007)]).

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

292 Principles of Radiation Interaction in Matter and Detection

and vicinity are shown in Figs. 4.1–4.2, respectively, for a standard high luminosity

year [ATLAS Collab. (1996)]. As will be discussed later, one has also to consider

damages created by ionization processes, which are mainly inﬂicted to electronics

and may alter the good operation of semiconductor devices.

This part of the book addresses the study of semiconductor b ehavior in radiation

environment. However, it is interesting to look brieﬂy at other types of detectors and

look at the impact of accelerator generated radiation ﬁeld on their operation. Again,

the example of LHC will be used. Liquid argon is the active medium of FCAL,

electromagnetic and HEC calorimeters of the ATLAS experiment. The choice of

liquid argon, as active medium of these calorimeters, was largely motivated by the

radiation hardness of the technique, compared to scintillator, for instance. The

concern about activation of argon through production of

Ar (half-life of 1.82

hours) via the reaction

n +

Ar →

Ar + γ

↓

K + β

−

+ ¯ν

has been discarded by simulation and measurements performed at SARA [Belymam

et al. (1998)]. The short half-life of

Ar makes dose source negligible during access

to the detector for maintenance. Absorbers and structural materials, used in ATLAS

liquid argon calorimeters, are radiation hard and stable under irradiation. Under

irradiation, oxygen or oxygen-like impurities are released from the surface of mate-

rials and equipments immersed in liquid argon. This outgassing leads to pollution

of liquid argon and to p ossible reduction of collected ionization charge and, ﬁnally,

in turn to the reduction of the calorimeter signal. Then, the operation of the AT-

LAS liquid argon calorimeters requires limited pollution of liquid argon, below a

threshold of (1–2) ppm. This goal can be achieved if the components of the liquid

argon calorimeters are certiﬁed against release of polluting impurities under irradi-

ation. Therefore, the possible outgassing of components immersed in liquid argon

and exposed to high ﬂuences was investigated within the ATLAS Collaboration

for several years, in conditions similar to those encountered in the running experi-

ment during LHC operation. The tested materials and equipments were subjected

to accelerated irradiation, in order to receive the equivalent of a LHC ten-year dose

within a short period of time. Cold test facilities were built at SARA [Belymam

et al. (1998)], CERI

and Dubna [Leroy et al. (2000a)] for that purpose. In parti-

cular, in Dubna, the large beam geometrical acceptance of 800 cm

at that facility

allows the exposure of relatively large area inside a one liter cryostat ﬁlled with

liquid argon. The Dubna IBR-2 pulsed neutron reactor made possible the irradia-

tion of materials and equipments immersed in liquid argon at high neutron ﬂuences

(≈ 10

n cm

−2

−1

up to a total ﬂuence of 10

n cm

−2

for a standard measurement

perio d of 11 days) and γ-doses (≈ 10 kGy per day), simultaneously, as γ’s always

CERI, Neutron Irradiation Facility, CNRS, Orlans, France.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

Radiation Environments and Damage in Silicon Semiconductors 293

accompany neutrons in the reactor. Sintered tungsten “slugs” used as absorber in

FCAL have been shown to suﬀer no outgassing [Leroy et al. (2000b)]. The Dubna

cold facility also allowed the test of the characteristics integrity of electronic compo-

nents under the high irradiation level expected at LHC [Leroy et al. (2002c)]. Glues

used in calorimeter electronics were selected according to their radiation hardness

and mechanical hardness under irradiation [Leroy et al. (2000c)].

The use of scintillator at LHC is mainly constrained by two factors: i) the inter-

bunch crossing time of 25 ns which requires a fast scintillator decay time and ii)

the high radiation level which requires a radiation hard scintillator. A high density

(8.28 g/cm

) scintillating crystal, PbWO

, which fulﬁls these constraints is used in

CMS [CMS (1997)] for electromagnetic calorimetry. Its short (long) decay time is 5

(15) ns and it is radiation hard at the expense of relatively poor light yield

†

Semiconductor trackers, consisting of silicon pad or pixel detectors and silicon

microstrips detectors, are used in collider experiments [ATLAS Collab. (1994b)]. Si-

licon pad detectors are also used as active medium of preshower calorimeters [AT-

LAS Collab. (1994b); CMS (1998)]. The exposure of silicon detectors or devices

to the high level of radiation encountered in high luminosity accelerators lead to

degradation of their performances. After intensive investigations, solutions have

been found to improve the condition and operation of silicon detectors in such envi-

ronment. A brief review of these solutions is presented now, more will be discussed

in subsequent chapters of this book.

Various types of damage, surface or lattice defects like vacancies (point-like

defects) or damaged regions (clusters), are produced when incoming particles de-

posit their energy in silicon. The importance of the damage created by a speciﬁc

process depends on its relative cross section compared to those of other possible

processes. One can distinguish the ionization process, resulting from the interaction

of the incident particle with the atomic electrons (e.g., see Sect. 4.2.3), and non-

ionization processes, resulting from the interaction between incoming particles and

lattice nuclei (e.g., see Sect. 4.2.1).

4.1.1.1 Ionization Processes and Collider Environments

The eﬀect of damage to silicon devices due to ionization process originates from

ionizing energy-loss in surface layers. Even at large radii, radiations aﬀect the elec-

tronics components located inside and around the LHC detectors through ionization

in silicon oxide by photons, protons, pions, ions (total ionizing dose - TID). The

various aspects of electronics radiation hardness have been extensively discussed

in many reports (see for instance [Dentan (1999)] in the framework of the ATLAS

experiment). The relevant TID eﬀects for silicon devices employed in LHC environ-

ments are summarized in Sect. 4.2.3.2.

†

The light yield compared to NaI(Tl) is 0.01.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

294 Principles of Radiation Interaction in Matter and Detection

4.1.1.2 Non-Ionization Processes, NIEL Scaling Hypothesis and Collider

Environments

The non-ionization processes produce bulk defects or displacement damage (e.g., see

Sect. 4.2.1). The recoil nuclei are removed from their initial lattice position and

displaced into interstitial positions (interstitials). However, if the recoil energy of

the primary knock-on nucleus is higher than the threshold for atomic displace-

ment, this nucleus can produce displacement of another nucleus. Finally, as long as

the recoil energy is above displacement threshold, the cascading-process will con-

tinue and a chain of vacancies (V ) and interstitials (I) will be produced (e.g., see

Sects. 4.2.1 and 4.2.2). The bulk damage eﬀects pro duced by energetic particles have

been shown to be proportional to the displacement damage cross-section equivalent

to the non-ionizing energy-loss [NIEL, e.g., see Sect. 4.2.1 and Eqs. (4.86, 4.88)]. The

proportionality between NIEL-value and resulting damage eﬀects is referred to as

the NIEL scaling hypothesis. The NIEL value depends on the particle type and

energy (e.g., see Sect. 4.2.1.5). However, violations of the NIEL scaling hyp othesis

were observed. This scaling does not work for calculating N

eﬀ

in oxygenated sili-

con [Rose Collab. (2000)] and for low proton energies for standard silicon [Bechevet,

Glaser, Houdayer, Lebel, Leroy, Moll and Roy (2002)].

Subjected to increasing ﬂuence, the main eﬀects on the operation of silicon de-

tectors, from a practical point of view, are i) an increase with ﬂuence of the leakage

current, ii) an increase of the full-depletion bias voltage to be applied to the de-

tector and iii) a decrease of the charge collection eﬃciency (CCE). The damaged

regions created by radiation dose in the silicon bulk are acting as electrically active

defects with deep levels in the forbidden-band gap in silicon. The generation of these

additional traps decreases the carrier lifetime and increase the reverse current. For

ﬂuences of the order of 10

particles cm

−2

, i.e., in practice ﬂuences corresponding

to one year of operation, in many regions of the LHC, the reverse current can reach

values in the range of several tens of microamperes per cm

and even more. However,

the strong temperature dependence of the reverse current allows it to be minimized

by operating the detectors at moderate low temperature (in a range from -5

◦

C to

-10

◦

C, typically). Re-arrangement between these vacancies and interstitials or their

interactions with other defects, impurities and nuclei in the bulk (like oxygen) are

responsible of long term eﬀects, e.g., the evolution in time of electrical characteristic

of silicon detectors such as annealing and reverse annealing of leakage current and

full depletion bias voltage after irradiation. It is observed that the leakage current

anneals following irradiation by about 50 % over a period of about two weeks at

room temperature. From the measurement of the leakage currents of silicon detec-

tors made of diﬀerent starting materials irradiated with neutrons, up to ﬂuences

of 10

n cm

−2

, the leakage current damage parameter is independent of the initial

resistivity and impurity concentrations, once normalized to the sensitive volume

and to the 1 MeV-neutron equivalent hadron ﬂuence. This parameter is linked to

defect clusters which are not aﬀected by the material. The concentration of defects,

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

Radiation Environments and Damage in Silicon Semiconductors 295

that are responsible for the leakage current, are in the order of the impurity concen-

trations and, consequently, migrating interstitials and vacancies produced by the

irradiation are responsible for the increase of the leakage current. Annealing of the

leakage current is also material independent. The period of annealing is followed

by a fast increase of the leakage current. This reverse annealing eﬀect is largely

moderated, even suppressed, if the irradiated diode is kept at low temperature,

around -5

◦

C. Therefore, cooling of the detector to moderately low temperatures

is a necessity for the reduction of the leakage current so keeping a signal-to-noise

ratio required for physics exploitation during colliding beam periods. This cooling is

also necessary during machine and detector maintenance periods to avoid reverse-

annealing leading to the increase of the reverse leakage current and full depletion

bias voltage.

The full depletion bias voltage is the minimal bias voltage to be applied in

order to fully deplete the detector

. In non-irradiated diodes, the space-charge re-

sults from shallow dopants in the silicon. Irradiation brings changes in the doping

concentration (donors and/or acceptors) with a build-up of negative space-charge

in the depletion region due to creation of deep energy levels. The eﬀective doping

concentration being proportional to the full depletion voltage, the value of the full

depletion bias voltage is aﬀected and can reach very high and unpractical values

after several years of the detector operation in high radiation environment. This has

the possible consequence that a fraction of the detectors will have to be operated

without being fully depleted. Similarly to the reverse leakage current, an annealing

eﬀect is observed for the eﬀective doping concentration (N

eﬀ

) over a period of about

two weeks after irradiation and followed by a reverse annealing, i.e., a fast increase,

particularly after several months for diodes kept at ro om temperature. The amount

and rate of reverse annealing depends on the irradiation ﬂuence. For a typical silicon

detector 300 µm thick and 5 kΩcm resistivity irradiated up to 10

p cm

−2

, the full

depletion bias voltage saturates at nearly 350 V. It should be reminded that the

operating bias voltage should always be larger than the full depletion bias voltage

due to trapping eﬀects. This necessary over-bias depends on the exposure ﬂuence

and state of annealing.

The charge collection eﬃciency (CCE) is degraded down to (10–15) % of the

value measured for non-irradiated silicon detector, depending on the irradiation

ﬂuence. However, for temperatures below 100 K, temperature range unpractical for

LHC, 50 % of the charge is recovered in detectors irradiated up to 2 × 10

n cm

−2

(due to the so-called Lazarus eﬀect

‡‡

[Palmieri et al. (1998); Bell et al. (1999)]).

In the case of liquid argon, radiation hardness is improved by selection of materi-

als certiﬁed against outgassing. For scintillator, one uses a new type of crystal with

The silicon detectors are typically p

− n junctions (e.g., see Sects. 4.3.2 and 6.8).

‡‡

It was found out that, at temperatures below 130 K, largely damaged detectors, apparently,

exhibited a signiﬁcant recovery of operating conditions [Palmieri et al. (1998)]. The explanation of

this phenomenon, known as the Lazarus Eﬀect, is related to the dynamics of the induced defects

in the semiconductor bulk.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

296 Principles of Radiation Interaction in Matter and Detection

increased radiation hardness. The technique, called defect engineering, is applied

to improve the radiation hardness of silicon. It involves the deliberate addition

of impurities, oxygen in practice, to the silicon bulk material during the growth

process [Rose Collab. (2000)] in order to aﬀect the formation of electrically active

defect centers which, in turn, allow control of the macroscopic parameters of silicon

devices. The oxygen concentration achieved is [O]∼(3–5) ×10

−3

. It should be

noted that high oxygen concentration {[O]∼(4–20) ×10

−3

} can be found in

silicon crystals grown by the Czochralski (Cz) method. Cz growth technology was

generally characterized by low resistivity (≤ 900 Ωcm) not really suited for radia-

tion detector applications. Recently, higher resistivity Cz silicon (1.2 kΩcm) suitable

for radiation detector application became available, after developments in crystal

growth technology [Savolainen et al. (2002)]. An explanation of the behavior of

oxygen enriched silicon is found by interpreting the role played by interstitial oxy-

gen, which is believed to act as a sink of vacancies. Then, divacancy-oxygen

∗

center defects are identiﬁed as main resp onsible for the radiation-induced negative

space-charge. V

O is produced through the reaction:

V + O

→ V O, (4.1)

followed by

V O + V → V

O. (4.2)

The addition of oxygen in the bulk material suppresses V

O formation. This

happens through the interaction between interstitial oxygen with vacancies. An

increase of interstitial oxygen concentration enhances the ratio of the reactions

expressed by Eq. (4.1) and Eq. (4.2), reducing formation of V

Oxygenation makes no diﬀerence to leakage current. Introduction of high concen-

trations [at the level of ∼ (3–5)×10

−3

] of oxygen in silicon detectors has led to

improved radiation hardness, when the detectors were exposed to charged hadrons

(protons, pions). However, almost no improvement is observed for irradiations of

oxygenated silicon materials with neutrons of a few MeV. It should be observed that

the proton measurements have been initially done with 24 GeV/c protons at CERN-

PS, while neutron measurements at high ﬂuences have been performed at reactors

and CERN-PS for low energy neutrons [(1–2) MeV]. The absence of measurement of

high energy neutrons (≥ 20 MeV) is due to the extreme diﬃculty of achieving high

ﬂuences for neutrons of these energies during short periods of time. For instance,

neutrons of 14 MeV can be produced

†

by 38 MeV proton on a Be-target with ﬂuences

of 10

n cm

−2

−1

, comparable to those expected at LHC. However, no measure-

ment has been done with oxygenated silicon detectors at this neutron energy. Low

energy proton (≤ 10 MeV) irradiation of silicon, doped with high concentration of

oxygen, shows improved radiation hardness of these detectors already observed for

∗

A general presentation of the radiation-induced defects can be found in Sect. 4.2.2.

†

For instance at the facility at the Nuclear Physics Institute (Academy of Sciences of the Czech

Republic), CZ-25068

Reˇz near Prague.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

Radiation Environments and Damage in Silicon Semiconductors 297

24 GeV/c protons [Houdayer et al. (2003)]. The question remains if observed im-

provement is energy related, i.e., depends on the value of the proton and neutron

cross sections. Based on model, simulation [Huhtinen (2002)] and data [Bechevet,

Glaser, Houdayer, Lebel, Leroy, Moll and Roy (2002)], radiation hardness improve-

ment of oxygenated silicon is expected when it is exposed to neutrons of at least

20 MeV energy.

The diﬀerent behavior between protons and neutrons, at least at low particle

energy, is of no practical consequences at LHC, since the radiation ﬁeld in LHC

collision areas is largely dominated by charged hadrons, where silicon detectors are

operated, neutron acting at larger radii. Therefore the use of diﬀusion oxygenated

standard planar or ﬂoat zone (DOFZ) silicon will be beneﬁcial. In addition, for

proton irradiation, N

eﬀ

achieved at high ﬂuences is independent of the oxygenated

material initial resistivity, but for neutron irradiation, the N

eﬀ

increase at high

ﬂuence is reduced by the use of oxygenated material of low resistivity. Therefore,

using oxygenated silicon of low resistivity [(1–2) kΩcm] would improve their radia-

tion hardness when operated in the mixed ﬁeld of charged hadrons and neutrons at

LHC.

4.1.2 Space Radiation Environment

The continuous evolution of mission requirements and their electronic technologies

for payloads and spacecrafts (for example, the International Space Station

∗∗

shown

in Fig. 4.3), combined with the need to meet the space environment constraints, par-

ticularly radiation, constitute challenges for component engineers and designers. For

instance, the increased activities in space for communications, military services and

scientiﬁc research have required to take into account that the electronics, employed

for these purposes, contain advanced devices and electronic systems in VLSI (very

large scale integration) and ULSI (ultra large scale integration) technologies, which

might have radiation sensitivity.

The space radiation is not homogenously distributed in the Earth magnetosphere

and solar cavity (the so-called heliosphere) and varies with time. As a consequence,

Low- (LEOs), Mid- (MEOs), Geosynchronous/Geostationary-Earth Orbits (GEOs),

Geostationary Transfer Orbits (GTOs) and orbits for an Earth Observatory Satellite

(EOS) have diﬀerent requirements for radiation hardness (see, for instance, [Barth

(1997)] and references therein). These orbits are distinguished by their degrees of

inclination and distances of p erigee and ap ogee. The interplanetary missions are

aﬀected by the energetic particles of solar and galactic sources traveling through

the interplanetary space (Sect. 4.1.2.4), but are less exposed to radiation eﬀects of

the trapped particles in the Earth magnetosphere (Sect. 4.1.2.5). The ﬂuxes of these

∗∗

Thanks to NASA’s courtesy, we show an image (Fig. 4.3) at the ﬁnal stage of comple-

tion. The Reference Guide to the International Space Station is available at the web site:

http://www.nasa.gov/mission pages/station/news/ISS Reference Guide.html.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

298 Principles of Radiation Interaction in Matter and Detection

Fig. 4.3 Computer-generated artist’s rendering (Courtesy NASA) of the International Space Station (ISS) when completed. The ISS orbits Earth

at an altitude that ranges from 370 to 460 km.

January 9, 2009 10:21 World Scientiﬁc Book - 9.75in x 6.5in ws-bo ok975x65˙n˙2nd˙Ed

Radiation Environments and Damage in Silicon Semiconductors 299

particles are modulated by the solar wind

‡‡

, whose intensity and speed depends on

the solar activity.

The presence of this complex environment has drawn attention to the important

inﬂuence of space radiation on both the computation tools for modeling the space

radiation (e.g., see [Barth (1997)], Chapter 1 in [Claeys and Simoen (2002)], [Heyn-

derickx (2002)], Chapter 2 in [Holmes-Siedle and Adams (2002)], [Miroshnichenko

(2003)]) and the response to radiation of MOS and bipolar transistors made in

VLSI BiCMOS (Bipolar Complementary Metal-Oxide Semiconductor) technolo-

gies. These devices are, indeed, the essential part of any circuit used in that en-

vironment, thus, the knowledge of irradiation eﬀects on them is critical (e.g., see

[Baschirotto et al. (1995b, 1996); Fleetwood et al. (1994); Johnston, Swift and

Rax (1994); Baschirotto et al. (1997); Colder et al. (2001, 2002); Codegoni et al.

(2004b); Consolandi, D’Angelo, Fallica, Mangoni, Modica, Pensotti and Rancoita

(2006)] and references therein; see, also, Sects. 4.2.3.2 and Chapter on Displacement

Damage and Particle Interactions in Silicon Devices). For instance, bipolar junction

transistors (BJT) have important applications in analog or mixed-signal IC’s and

BiCMOS circuits because of their linearity and excellent matching characteristics,

for example. Many of the bipolar integrated circuits in space systems, including op-

erational ampliﬁers, comparators, voltage regulators, are used to accomplish analog

functions. Furthermore, MOS transistors [particularly those of the complementary

form (CMOS)] are of wide usage in high-performance and low-power electronics.

Once deﬁned the duration and the orbit of a satellite or a pay-load inside

the heliosphere or the Earth magnetosphere, the knowledge of fast charged par-

ticle ﬂuences and their time dependence makes possible to determine the expected

Frenkel-pairs concentration (FP ) due to the non-ionizing energy-loss processes (see

Sect. 4.2.1) and the absorbed dose mostly determined by collision energy-loss pro-

cesses (e.g., see the chapters on Electromagnetic Interaction of Radiation in Matter

and Nuclear Interactions in Matter, and [Ziegler, Biersack and Littmark (1985a);

Ziegler, J.F. and M.D. and Biersack (2008a)])

∗

. Both types of processes result from

the interactions of charged particles impinging on semiconductor devices (i.e., on

microelectronics).

This knowledge has to be complemented by investigations of possible latch up

and single event upset (e.g., see [Messenger and Ash (1997)]) for an eﬀective imple-

mentation of any VLSI circuit, whose design can, in turn, depend on the radiation

response (e.g., see Sects. 4.2.3.2 and 7.2). Obviously, a non-radiation hard technol-

ogy cannot be employed in a radiation environment: the space qualiﬁcation of VLSI

technologies with respect to TID is certainly needed for any design consideration

and may require the qualiﬁcation of their basic bipolar devices with regard to space-

radiation. As an example, when bipolar integrated circuits are exposed to radiation

‡‡

For instance, the reader can see [White (1970); Encrenaz, Bibring and Blanc (1991); Alurkar

(1997); Boella, Gervasi, Potenza, Rancoita and Usoskin (1998); Boella, Gervasi, Mariani, Rancoita

and Usoskin (2001); Grieder (2001); Lang (2001); Gosling (2006); Potgieter (2008)].

∗

The reader may also see [Leroy and Rancoita (2004)].