Comprehensive nuclear materials 4 02 radiation damage in austenitic steels

4.02

Radiation Damage in Austenitic Steels

F. A. Garner
Radiation Effects Consulting, Richland, WA, USA

ß 2012 Elsevier Ltd. All rights reserved.

4.02.1
4.02.2
4.02.2.1
4.02.2.2
4.02.3
4.02.4
4.02.5
4.02.6
4.02.7
4.02.8
4.02.8.1
4.02.8.2
4.02.8.3
4.02.8.3.1
4.02.8.3.2
4.02.8.3.3
4.02.8.3.4
4.02.8.3.5
4.02.9
4.02.9.1
4.02.9.2
4.02.9.3

4.02.9.4
4.02.9.5
4.02.9.5.1
4.02.9.5.2
4.02.9.5.3
4.02.9.6
4.02.9.7
4.02.9.8
4.02.10
References

Introduction
Basic Damage Processes
Atomic Displacements
Transmutation
Differences in Neutron Spectra
Transmutation Issues for Stainless Steels
Evolution of Radiation-Induced Microchemistry and Microstructure
A Cross-Over Issue Involving Radiation-Induced Microstructural Evolution and
Transmutation
Radiation-Induced Changes in Mechanical Properties
Radiation-Induced Changes in Dimension
Precipitation-Related Strains
Void Swelling and Bubble Swelling
Parametric Dependencies of Void Swelling
Stress state
Elemental composition
Alloy starting state
Irradiation temperature
Influence of dpa rate on swelling

Irradiation Creep
Introduction
Stages of Irradiation Creep
Examples of Creep Behavior
Creep Disappearance
Recent Revisions in Understanding of Irradiation Creep
Dependence of irradiation creep on dpa rate
Dependence of creep and creep relaxation on neutron spectra
Dependence of creep modulus on hydrostatic stress
Stress Relaxation by Irradiation Creep
Stress Rupture
Fatigue
Conclusions

Abbreviations
ATR
BN-350
BN-600
BOR-60

Advanced Test Reactor in Idaho Falls,
Idaho
Russian acronym for Fast Neutron at 350
MW in Actau, Kazakhstan
Russian acronym for Fast Neutron at 600
MW in Zarechney, Russia
Russian acronym for Fast Experimental
Reactor at 60 MW in Dimitrovgrad,
Russia

BR-2
BR-10
BWR
CAGR
CANDU
DFR

34
35
35
37
37
40
44
49
50
61
62
65
67
67
68
69
69
70
74
74
78
79
79

83
83
84
85
86
88
89
90
91

Belgium Research Reactor-II in Mol,
Belgium
Russian acronym for Fast Reactor at 10
MW in Obninsk, Russia
Boiling water reactor
Commercial Advanced Gas Reactor
Registered trademark for Canadian
Deuterium Uranium Reactor
Dounreay Fast Reactor in Dounreay,
Scotland

33


34

Radiation Damage in Austenitic Steels

DMTR
EBR-II

FFTF
HFIR
HFR
IASCC
IGSCC
JMTR
NRU
ORNL
ORR
PWR
T/F
VVER

Dounreay Materials Test Reactor in
Dounreay, Scotland
Experimental Breeder Reactor-II in
Idaho Falls, Idaho
Fast Flux Test Facility, fast reactor in
Richland, WA
High Flux Isotope Reactor at Oak Ridge
National Laboratory
High Flux Reactor in Petten, Netherlands
Irradiation-assisted stress corrosion
cracking
Intergranular stress corrosion cracking
Japan Material Testing Reactor in Oarai,
Japan
National Research Universal Reactor in
Chalk River, Canada
Oak Ridge National Laboratory:

Oak Ridge Research Reactor in Oak
Ridge, Tennessee
Pressurized water reactor
Thermal-to-fast neutron ratio
Russian acronym for water-cooled,
water moderated energetic reactor

4.02.1 Introduction
Austenitic stainless steels are widely used as structural components in nuclear service in addition
to being employed in many other nonnuclear
engineering and technological applications. The
description of these steels and their as-fabricated
properties is covered in Chapter 2.09, Properties
of Austenitic Steels for Nuclear Reactor Applications. This chapter describes the evolution of both
microstructure and macroscopic property changes
that occur when these steels are subjected not only
to prolonged strenuous environments but also to the
punishing effects of radiation. While various nuclear
environments involve mixtures of charged particles,
high-energy photons and neutrons, it is the latter
that usually exerts the strongest influence on the
evolution of structural steels and thereby determines
the lifetime and continued functionality of structural
components.
To describe the response of austenitic stainless
steels in all neutron environments is a challenging
assignment, especially given the wide range of
neutron spectra characteristic of various neutron
devices. This review of neutron-induced changes in
properties and dimensions of austenitic stainless

steels in all spectral environments has therefore
been compiled from a series of other, more focused
reviews directed toward particular reactor types1–8
and then augmented with material from a recently
published textbook9 and journal articles. It should be
noted, however, that many of the behavioral characteristics of iron-based stainless steels following
neutron irradiation are also observed in nickelbased alloys. Whenever appropriate, the similarities
between the two face-centered-cubic alloy systems
will be highlighted. A more comprehensive treatment of radiation effects in nickel-base alloys is
provided in Chapter 4.04, Radiation Effects in
Nickel-Based Alloys.
This review is confined to the effects of neutron
exposure only on the response of irradiated steels and
does not address the influence of charged particle
irradiation. While most of the phenomena induced
by neutrons and charged particles are identical, there
are additional processes occurring in charged particle studies that can strongly influence the results.
Examples of processes characteristic of charged particle simulations are the injected interstitial effect,10,11
strong surface effects,12,13 dose gradients,14,15 and
atypical stress states.16,17 Chapter 1.07, Radiation
Damage Using Ion Beams addresses the use of
charged particles for irradiation.
Austenitic stainless steels used as fuel cladding or
structural components in various reactor types must
often withstand an exceptionally strenuous and challenging environment, even in the absence of neutron
irradiation. Depending on the particular reactor type,
the inlet temperature during reactor operation can
range from $50 to $370  C. The maximum temperature can range from as high as 650 to 700  C for
structural components in some reactor types, although

most nonfueled stainless steel components reach
maximum temperatures in the range of 400–550  C.
During operation, the steel must also withstand the
corrosive action of fission products on some surfaces
and flowing coolant on other surfaces. The coolant
especially may be corrosive to the steel under
operating conditions. Some of these environmental
phenomena are synergized or enhanced by the effect
of neutron irradiation.
Dependent on the nature of the component and
the length of its exposure, there may also be significant levels of stress acting on the component.
Stress not only influences cracking and corrosion
(see Chapter 5.08, Irradiation Assisted Stress Corrosion Cracking) but can also impact the dimensional stability of stainless steel, primarily due to


Radiation Damage in Austenitic Steels

thermal creep and irradiation creep, and also from
the influence of stress on precipitation, phase stability, and void growth, some of which will be discussed
later. However, it will be shown that neutron irradiation can strongly affect both the microstructure and
microchemistry of stainless steels and high-nickel
alloys, with strong consequences on physical properties, mechanical properties, dimensional stability, and
structural integrity.
Stainless steels are currently being used or have
been used as structural materials in a variety of
nuclear environments, most particularly in sodiumcooled fast reactors, water-cooled and water-moderated
test reactors, water-cooled and water-moderated
power reactors, with the latter subdivided into light
water and heavy water types. Additionally, there are
reactor types involving the use of other coolants

(helium, lithium, NaK, lead, lead–bismuth eutectic,
mercury, molten salt, organic liquids, etc.) and other
moderators such as graphite or beryllium.
The preceding reactor types are based on the
fission of uranium and/or plutonium, producing
neutron energy distributions peaking at $2 MeV
prior to moderation and leakage effects that produce
the operating spectrum. However, there are more
energetic sources of neutrons in fusion-derived
spectra, with the source peaking at $14 MeV and
especially from spallation events occurring at energies of hundreds of MeV, although most spallation
spectra are mixtures of high-energy protons and
neutrons. It is important to note that in each of
these various reactors, there are not only significant
differences in neutron flux-spectra but also significant differences in neutron fluence experienced by
structural components. These differences in fluence
arise not only from differences in neutron flux
characteristic of the different reactor types but also
the location of the steel relative to the core. For
instance, boiling water reactors and pressurized
water reactors have similar in-core spectra, but
stainless steels in boiling water reactors are located
much farther from the core, resulting in a factor of
reduction of $20 in both neutron dose rate and
accumulated dose compared to steels in pressurized
water reactors.

4.02.2 Basic Damage Processes
4.02.2.1

Atomic Displacements

What are the nature and origins of neutron-induced
phenomena in metals? The major underlying driving

35

force arises primarily from neutron collisions with
atoms in a crystalline metal matrix. When exposed to
displacive irradiation by energetic neutrons, the atoms
in a metal experience a transfer of energy, which if
larger than several tens of eV, can lead to displacement
of the atom from its crystalline position. The displacements can be in the form of single displacements
resulting from a low-energy neutron collision with a
single atom or a glancing collision with a higher energy
neutron. More frequently, however, the ‘primary
knock-on’ collision involves a larger energy transfer
and there occurs a localized ‘cascade’ of defects that
result from subsequent atom-to-atom collisions.
There are several other contributions to displacement of atoms from their lattice site, but these are
usually of second-order importance. The first of
these processes involve production of energetic electrons produced by high-energy photons via the
photoelectric effect, Compton Effect, or pair production.18 These electrons can then cause atomic displacements, but at a much lower efficiency than that
associated with neutron-scattering events. The second type of process involves neutron absorption by
an atom, its subsequent transmutation or excitation,
followed by gamma emission. The emission-induced
recoil of the resulting isotope often is sufficient to
displace one or several atoms. In general, however,
such recoils add a maximum of only several percent
to the displacement process and only then in highly

thermalized neutron spectra.4 One very significant
exception to this generalization involving nickel will
be presented later.
For structural components of various types of
nuclear reactors, it is the convention to express the
accumulated damage exposure in terms of the calculated number of times, on the average, that each atom
has been displaced from its lattice site. Thus, 10 dpa
(displacements per atom) means that each atom
has been displaced an average of 10 times. Doses
in the order of 100–200 dpa can be accumulated
over the lifetimes of some reactor components in
various high-flux reactor types. The dpa concept is
very useful in that it divorces the damage process
from the details of the neutron spectrum, allowing
comparison of data generated in various spectra,
providing that the damage mechanism arises primarily from displacements and not from transmutation.
The use of the dpa concept also relieves researchers from the use of relatively artificial and sometimes
confusing threshold energies frequently used to
describe the damage-causing portion of the neutron
spectrum. Neutrons with ‘energies greater than


36

Radiation Damage in Austenitic Steels

X MeV,’ where X is most frequently 0.0, 0.1, 0.5, or
1.0 MeV, have been used for different reactor concepts at different times in history. The threshold
energy of 0.1 MeV is currently the most widely
used value and is most applicable to fast reactors

where large fractions of the spectra lay below 0.5
and 1.0 MeV. Many older studies employed the total
neutron flux (E > 0.0) but this is the least useful
threshold for most correlation efforts. Caution
should be exercised when compiling data from
many older studies where the neutron flux was not
adequately identified in terms of the threshold
energy employed.
There are rough conversion factors for ‘displacement effectiveness’ for 300 series austenitic steels that
are useful for estimating dpa from >0.1 MeV fluences
for both in-core or near-core spectra in most fission
spectra. Examples are $7 dpa per 1022 n cmÀ2 (E >
0.1) for most in-core light water spectra with lower
in-core values of $5 dpa per 1022 n cmÀ2 (E > 0.1) for
metal fueled fast reactors and $4 dpa per 1022 n cmÀ2
(E > 0.1) for oxide-fueled fast reactors.4 Such conversion factors should not be trusted within more
than (10–15%), primarily due to spatial variations
across the core resulting from neutron leakage. For
fast reactor spectra, E > 1.0 conversion factors are
completely unreliable.
When E > 1.0 fluxes are employed in light water
reactor studies, the conversion factor increases
from $7 dpa per 1022 n cmÀ2 (E > 0.1) to $14 dpa
per 1022 n cmÀ2 (E > 1.0). In Russia, a threshold
energy of >0.5 MeV is popular for light water

reactors with $9 dpa per 1022 n cmÀ2 (E > 0.5). All
of these conversion factors assume that within several
percent pure iron is a good surrogate for 300 series
alloys. Note that other metals such as Cu, Al, W, etc.

will have different conversion values arising from
different displacement threshold energies and sometimes different displacement contributions.
A standard procedure for calculating dpa has been
published,19 although other definitions of dpa were
used prior to international acceptance of the ‘NRT
model’ where the letters represent the first letter
of the three author’s last name (see Garner1 for
details on earlier models). Caution must be exercised
when compiling doses from older studies where
displacement doses were calculated using other models (Kinchin-Pease, Half-Nelson, French dpa, etc.)
sometimes without clearly identifying the model
employed. Conversion factors between the NRT
model and various older models of dpa are provided
in Garner,1 but all models agree within $23%.
While sometimes controversial with respect to
how far the dpa concept can be stretched to cover
the full range of spectral differences for neutron and
especially for charged particle environments, it appears
that the dpa concept is very efficient to stretch over
light water, heavy water, fusion, and spallation spectra,
providing that all energy deposition and displacement
processes are included. Note in Figure 1 how well the
dpa concept collapses the data on neutron-induced
strengthening of stainless steel into one response
function for three very different spectra (light water
fission, pure D–T fusion and ‘beam-stop’ spallation).20

300
LASREF, 40 ЊC
RTNS-II, 90 ЊC

OWR, 90 ЊC

250

Yield stress change (MPa)

Yield stress change (MPa)

300

200

150

100

50

0

1017

1018

1019

Neutron fluence, E > 0.1 MeV

1020

250

LASREF, 40 ЊC
RTNS-II, 90 ЊC
OWR, 90 ЊC

200

150

100

50

0

10-3

10-2

dpa

Figure 1 Radiation-induced yield stress changes of 316 stainless steel versus (left) neutron fluence (n cmÀ2 E > 0.1 MeV),
and (right) displacements per atom. Reproduced from Heinisch, H. L.; Hamilton, M. L.; Sommer, W. F.; Ferguson, P.
J. Nucl. Mater. 1992, 191–194, 1177, as modified by Greenwood, L. R. J. Nucl. Mater. 1994, 216, 29–44.


Radiation Damage in Austenitic Steels

4.02.2.2

Transmutation

It is important to note that material modification by
radiation arises from two primary spectral-related
processes . In addition to the neutron-induced displacement of atoms there can be a chemical and/or
isotopic alteration of the steel via transmutation.
With the exception of helium production, transmutation in general has been ignored as being a significant
contributor to property changes of stainless steels and
nickel-base alloys. In this chapter, transmutation is
shown to be sometimes much more important than
previously assumed.
Both the displacement and transmutation processes are sensitive to the details of the neutron
flux-spectra, and under some conditions each can
synergistically and strongly impact the properties of
the steel during irradiation. In addition to the brief
summary presented below on flux-spectra issues relevant to stainless steels, the reader is referred to
various papers on transmutation and its consequences
in different reactor spectra.5–8,18,21–23
Transmutation may be subdivided into four categories of transmutants. Three of these are relevant
to fission-derived or fusion-derived spectra, and the
fourth is associated with spallation-derived spectra.
The first three are solid transmutants, gaseous transmutants, and ‘isotope shifts,’ the latter involving production of other isotopes of the same element. While
the latter does not change the chemical composition
of stainless steels, it is an underappreciated effect that
is particularly relevant to nickel-containing alloys
such as stainless steels and nickel-base alloys when
irradiated in highly thermalized neutron spectra.
Whereas the first three categories arise from discrete nuclear reactions to produce discrete isotopes of
specific elements, the spallation-induced transmutation arising in accelerator-driven devices involves a

continuous distribution of every conceivable fragment
of the spalled atom, producing every element below
that of the target atom across a wide range of isotopes
for each element. While individual solid transmutants
in spallation spectra are usually produced at levels
that do not change the alloy composition significantly,
the very wide range of elements produced allows the
possibility that deleterious impurities not normally
found in the original steel may impact its continued
viability. This possibility has not received sufficient
attention and should be examined further if spallation
devices continue to be developed.
Another consequence of spallation-relevant transmutation is that the induced radioactivity per unit

37

mass is correspondingly much higher than that produced per dpa in other spectra. The majority of the
spalled fragments and their daughters/granddaughters are radioactive with relatively short half-lives,
leading to materials that are often much more difficult to examine than materials irradiated in fission
spectra.
Most importantly, there is a very strong production of hydrogen and helium in spallation spectra at
levels that are one or two orders of magnitude greater
than produced in most fission or fusion spectra.5,6,21
While there is a tendency to view displacement
and transmutation processes as separate processes,
it will be shown later that under some circumstances the two processes are strongly linked and
therefore inseparable in their action to change alloy
behavior.

4.02.3 Differences in Neutron

Spectra
There are significant differences in neutron spectra
for water-cooled, sodium-cooled, and other types of
fission-based reactors. It should be noted that there is
a conventional but slightly misleading practice to
differentiate between ‘fast’ and ‘thermal’ reactors.
Thermal reactors have a significant portion of their
spectra composed of thermal neutrons. Thermalized
neutrons have suffered enough collisions with the
moderator material that they are in thermal equilibrium with the vibrations of the surrounding atoms.
Efficient thermalization requires low-Z materials
such as H, D, and C in the form of water, graphite,
or hydrocarbons. At room temperature the mean
energy of thermalized neutrons is 0.023 eV.
The designation ‘fast’ reactor, as compared to
‘thermal’ reactor, refers to the portion of the neutron
spectrum used to control the kinetics of ascent to full
power for each type of reactor. As shown later, this
practice incorrectly implies to many that fast reactors
have ‘harder’ neutron spectra than do ‘softer’ thermal
reactors. Actually, the opposite is true.
Examples of typical flux-spectral differences in
fission-based reactors are shown in Figures 2–5.
The local spectrum at any position is determined
primarily by the fuel (U, Pu) and fuel type (metal,
oxide, carbide, etc.), the coolant identity and density,
the local balance of fuel/coolant/metal as well as the
proximity to control rods, water traps, or core boundaries. Additionally, it is possible to modify the neutron
spectra in a given irradiation capsule by including in it

38

Radiation Damage in Austenitic Steels

1.E + 16

1015
HFIR

Flux/lethargy

Flux per unit lethargy

HFIR-PTP
1.E + 14

1014
ORR

HFIR-RB*
1.E + 12
ATR-ITV
1.E + 10

1013

EBRII
FFTF

1.E + 08
1.E - 9

EBR II
1012

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100
Neutron energy (MeV)

101 102

Figure 2 Difference in neutron flux-spectra of two
water-cooled test reactors (high-flux HFIR and lower-flux
ORR) and one high-flux sodium-cooled fast reactor (EBR-II).

1014
T/F ~0.15

1.E - 7

1.E - 5
1.E - 3
1.E - 1
Neutron energy (MeV)

1.E + 1

Figure 4 Comparison of flux-spectra in various test
reactors. Note that FFTF is softer in spectrum compared to
EBR-II due to the use of oxide fuel rather than metal fuel.

Neither fast reactor has measurable fluxes of thermal
neutrons. In the PTP position of HFIR a water trap strongly
contributes to a high thermal-to-fast ratio, while in the RB*
(removable beryllium) position the predominance of Be over
water reduces the thermal population. In the ATR position
where the ITV assembly was located, the use of strong
absorber sleeves strongly depressed the thermal flux.

Flux per unit lethargy

1013
Baffle bolt

Top of
bolt head

12

10

1011
Upper core plate
10

10

109

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 100
Neutron energy (MeV)

101 102

Figure 3 Typical neutron flux-spectra of internal
components of a pressurized water reactor, having a
thermal-to-fast neutron ratio smaller by factors of 10–20
than that of typical light water test reactors. Reproduced
from Garner, F. A.; Greenwood, L. R. In 11th International
Conference on Environmental Degradation of Materials
in Nuclear Power Systems – Water Reactors; 2003;
pp 887–909.

or enclosing it with a moderator or absorber. Metal
hydrides are used in fast reactors to soften the spectrum, while in mixed-spectrum reactors the thermalto-fast ratio can be strongly reduced by incorporating
elements such as B, Hf, Gd, and Eu.
The most pronounced influence on neutron
spectra in fission reactors arises from the choices of
coolant and moderator, which are often the same
material (e.g., water). Moving from heavy liquid metals
such as lead or lead–bismuth to lighter metals such as
sodium leads to less energetic or ‘softer’ spectra.
Use of light water for cooling serves as a
much more effective moderator. Counterintuitively,

however, this leads to both more energetic and less
energetic spectra at the same time, producing a twopeaked ‘fast’ and ‘thermal’ distribution separated by a
wide energy gulf at lower fluxes.
Such two-peaked spectra are frequently called
‘mixed spectra.’ The ratio of the thermal and fast
neutron fluxes in and near such reactors can vary

significantly with position and also with time.4 Using
heavy water, we obtain a somewhat less efficient
moderator that does not absorb neutrons as easily as
light water, but one that produces an even more pronounced two-peak spectral distribution where the
thermal-to-fast neutron ratio can be very large.
These spectral differences lead to strong variations between various reactors in the neutron’s ability
to displace atoms and to cause transmutation. Depending on the reactor size and its construction details
there can also be significant variations in neutron
spectra and ‘displacement effectiveness’ within a
given reactor and its environs, especially where
more energetic neutrons can leak out of the core.
Examples of these variations of displacement effectiveness for fast reactors are shown in Figures 6
and 7. Compared to fission-derived spectra, there
are even larger spectral differences in various fusion
or spallation neutron devices.
The reader should note the emphasis placed here
on flux-spectra rather than simply spectra. If we focus
only on light water-cooled reactors for example,
there are in general three regimes of neutron flux of
relevance to this review. First, there are the relatively
low fluxes typical of many experimental reactors that


Radiation Damage in Austenitic Steels

1016

Thermal flux, clean core
Thermal flux, 21-day core
Total nonthermal flux

>0.111 MeV
>0.821 MeV

5
2
1015
5
2

1013

0

4

8

12

H2O

Permanent
beryllium

Removable
beryllium

Control region

Outer fuel annulus

H2O annulus

2

Inner fuel annulus

5

H2O outer annulus

1014
300-g Pu target

Neutronflux (neutrons per cm2 s–1)

39

16 20 24 28 32 36 40 44
Radial distance from core center (cm)

48

52

56

60

Figure 5 Variation in fast and thermal fluxes in HFIR as a function of radial position at mid-core at 85 MW, also showing

change in thermal population with burn-up (Source: ORNL website).

can produce doses of 10 dpa or less over a decade.
Second, there are moderate flux reactors that are
used to produce power that can introduce doses as
high as 60–100 dpa maximum over a 30–40 year lifetime and finally, some high-flux thermal reactors that
can produce 10–15 dpa yearÀ1 in stainless steels.
Most importantly, fast reactors also operate in
the high-flux regime, producing 10–40 dpa yearÀ1.
Therefore, the largest amount of published highdpa data on stainless steels has been generated in
fast reactors. Some phenomena observed at high
exposure, such as void swelling, have been found to
be exceptionally sensitive to the dpa rate, while
others are less sensitive (change in yield strength)
or essentially insensitive (irradiation creep). These
sensitivities will be covered in later sections.
For light water-cooled reactors, the various flux
regimes need not necessarily involve large differences in neutron spectra, but only in flux. However,
the very large dpa rates characteristic of fast reactors
are associated with a significant difference in spectrum. This difference is a direct consequence of the
fact that fast reactors were originally designed to
breed the fissionable isotope 239Pu from the relatively
nonfissile isotope 238U, which comprises 99.3% of
natural uranium.
In order to maximize the breeding of 239Pu, it
is necessary to minimize the unproductive capture
of neutrons by elements other than uranium. One

5.5
Row 2

5.0
dpa
1022 (E > 0.1)
Row 4

4.5

4.0
-20

-10

0

10

20

Axial position (cm)
Figure 6 Displacement effectiveness values of dpa per
1022 n cmÀ2 (E > 0.1 MeV) across the small core (30 cm
tall and $30 cm diameter) of the EBR-II fast reactor,
showing effects of neutron leakage to soften the spectrum
near the core axial boundaries. Near core center (Row 2)
the spectrum and displacement effectiveness are
dictated primarily by the use of metal fuel, producing a
maximum of $5.2 dpa per 1022 n cmÀ2 (E > 0.1 MeV).
In mid-core Row 4 the radial leakage is just becoming
significant.

strategy used to accomplish this goal is to avoid
thermalization of the reactor neutrons, which requires
that no low atomic weight materials such as H2O, D2O,
Be, or graphite be used as coolants or as moderators.
For this purpose, sodium is an excellent coolant with
a moderate atomic weight. The use of sodium results


40

Radiation Damage in Austenitic Steels

6.0

BC

1

2

FFTF core
3
4

5.0

5

Above core

6
7
8

FFTF cycles 2 and 3

dpa
1022 (E > 0.1)

FFTF cycle 10

4.0

3.0
−100

−75

−50

−25

0

25

50

75

100

125

150

Distance from core midplane (cm)
Figure 7 Values of dpa per 1022 n cmÀ2 (E > 0.1 MeV) across the much larger core of FFTF for two different fuel/experiment
loadings, showing a lesser effect of neutron leakage in larger cores. Note, however, that the in-core values are less than
the in-core values of EBR-II, reflecting the softer spectra arising from the use of oxide fuel. Far from the core the
displacement effectiveness values are lower, determined primarily by the absence of fuel and the balance of sodium
and steel.

in a neutron spectrum that is nominally single-peaked
rather than the typical double-peaked (thermal and
fast) neutron spectrum found in light water or heavy
water reactors. The single-peaked fast reactor spectrum is significantly less energetic or softer, however,
than that found in the fast peak of light water reactors.
Depending on the fuel type (metal vs. oxide) the
mean energy of fast reactor spectra varies from $0.8
to $0.5–0.4 MeV while light water-cooled reactors
have a fast neutron peak near $1.2 MeV.
One consequence of attaining successful breeding
conditions is that the spectrum-averaged crosssection for fission is reduced by a factor of 300–400
relative to that found in light water spectra. To reach
a power density comparable to that of a light water
power-producing reactor, the fast reactor utilizes two
concurrent strategies: increases in fissile enrichment
to levels in the order of 20% or more, and most
importantly, an increase in neutron flux by one or

two orders of magnitude.
Thus, for a given power density, the fast reactor
will subject its structural materials to the punishing
effects of neutron bombardment at a rate that is
several orders of magnitude greater than that in
light water reactors. At the same time, however, the
softer ‘fast’ spectrum without thermalized neutrons
leads to a significant reduction in transmutation
compared to typical light water spectra, at least for
stainless steels and nickel-base steels.

4.02.4 Transmutation Issues for
Stainless Steels
For most, but not all fission-derived spectra, stainless
steels are relatively immune to transmutation, especially when compared to other elements such as
aluminum, copper, silver, gold, vanadium, tungsten,
and rhenium,5,21,24–27 each of which can rapidly
become two or three component alloys via transmutation arising from thermal or epithermal neutrons.
Whereas the properties of these metals are particularly sensitive to formation of solid transmutation
products, stainless steels in general do not change
their composition by significant amounts compared
to preexisting levels of impurities, but significant
amounts of helium and hydrogen can be produced
in fission-derived spectra, however.
In stainless steels the primary transmutant
changes that arise in various fission and fusion reactor
spectra involve the loss of manganese to form iron,
loss of chromium to form vanadium, conversion of
boron to lithium and helium, and formation of helium
and hydrogen gas.4,28 While each of these changes in

solid or gaseous elements are produced at relatively
small concentrations, they can impact the evolution
of alloy properties and behavior.
For instance, vanadium is not a starting component of most 300 series stainless steels, but when
included it participates in the formation of carbide


Radiation Damage in Austenitic Steels

with the major alloy components. This type of reaction occurs only above high neutron threshold energies (>6 MeV). Figure 8 shows that nickel is the
major contributor to helium production by (n, a)
reactions,36 and thus the helium generation rate
scales almost directly with nickel content for a large
number of commercial steels.
A similar behavior occurs for production of
hydrogen by transmutation via high-energy neutrons,
where nickel is also the major source of hydrogen
compared to other elements in the steel.4,7 In this
case, the threshold energy is around 1 MeV with
58
Ni being the major contributor.
This generality concerning nickel as the major
source of He and H is preserved in more energetic
fusion-derived spectra, although the He/dpa and
H/dpa generation rates in fusion spectra are much
larger than those of fast reactor spectra. When
moving to very energetic spallation-derived neutron
and proton spectra, however, the observation that
nickel accounts for most of the helium and hydrogen
is no longer correct. Iron, nickel, chromium, cobalt,

and copper produce essentially the same amounts of
helium and hydrogen for energies above $100 MeV
as shown in Figure 9.6
Another very important helium-generation process also involves nickel. Helium is produced
via the two-step 58Ni(n, g)59Ni(n, a)56Fe reaction
sequence.37,38 This sequence operates very strongly
in mixed-spectrum reactors. 59Ni is not a naturally
occurring isotope and is produced from 58Ni. Thus,
this helium contribution involves a delay relative to

0.14
Cross-section (barns)

precipitates that change the distribution and chemical activity of carbon in the alloy matrix. Carbon
plays a number of important roles in the evolution
of microstructure1 and especially in grain boundary
composition. The latter consideration is very important in determining the grain boundary cracking
behavior, designated irradiation-assisted stress corrosion cracking (IASCC), especially with respect to the
sensitization process.29
The strong loss of manganese in highly thermalized neutron spectra has been suggested to degrade
the stability of insoluble MnS precipitates that tie up
S, Cl, and F, all of which are elements implicated
in grain boundary cracking.30 Late-term radiationinduced release of these impurities to grain boundaries may participate in cracking, but this possibility
has not yet been conclusively demonstrated.
In some high-manganese alloys such as XM-19
manganese serves to enhance the solubility of
nitrogen which serves as a very efficient matrix
strengthener. In highly thermalized spectra the loss
of manganese via transmutation has been proposed to
possibly lead to a decrease in the strength of the alloy

and perhaps to induce a release of nitrogen from
solution to form bubbles.31
The overwhelming majority of published transmutation studies for stainless steels and high-nickel
alloys steels have addressed the effects of He/dpa
ratio on mechanical properties and dimensional
instabilities. Much less attention has been paid to
the effect of H/dpa ratio based on the long-standing
perception that hydrogen is very mobile in metals
and therefore is not easily retained in steels at
reactor-relevant temperatures. As presented later,
this perception is now known to be incorrect, especially for water-cooled reactors.
The focus of most published studies concerned
the much higher helium generation rates anticipated
in fusion spectra ($3–10 appm He/dpa) compared to
the lower rates found in fast reactors ($0.1–0.3 appm
He/dpa).32 It was later realized that in some highly
thermalized test reactors, such as HFIR, very large
generation rates could be reached ($100 appm
He/dpa), and even in pressurized water reactors
the rate could be very high ($15 appm He/dpa).33
In heavy water reactors the rate can be much larger,
especially in out-of-core regions.34,35
While some helium arises from (n, a) reactions
with thermal and epithermal neutrons interacting
with the small amounts of boron found in most
stainless steels, the major contribution comes initially
from high-energy threshold-type (n, a) reactions

41

0.12

Ni

0.10
0.08
0.06

Cr
Ti

0.04

Fe

0.02
1

2

4
6
Energy (MeV)

8 10

20

Figure 8 Cross-sections for (n, a) reactions as a function
of neutron energy for common elements used in stainless

steels. Reproduced from Mansur, L. K.; Grossbeck, M. L.
J. Nucl. Mater. 1988, 155–157, 130–147. Nickel dominates
the production of helium at higher neutron energies.


42

Radiation Damage in Austenitic Steels

2500
Inconel
304L
316L
9Cr–1Mo
Fe
Co
Ni
Cu

1500
1000

1.6

500
0

0

5

10

15

Ratio to initial value

He (appm)

2000

60

Ni
Natural nickel
58Ni 67.85%
60Ni 26.2%

1.2

61Ni
58

Ni

0.8

62Ni
64Ni

6.1%
total

dpa

Figure 9 Measured amount of helium in alloys and pure
metals that were irradiated by a mixed spectrum of high
energy neutrons and protons produced by 800 MeV proton
irradiation of tungsten rods. There is some significant
uncertainty in the dpa assignment for Inconel 718 at the
highest dose. Otherwise the He/dpa ratio appears to be
independent of composition. Reproduced from Garner, F. A.;
Oliver, B. M.; Greenwood, L. R.; James, M. R.; Ferguson, P. D.;
Maloy, S. A.; Sommer, W. F. J. Nucl. Mater. 2001, 296,
66–82.

that of single-step threshold (n, a) reactions. Since
both steps of the sequence involve cross-sections that
increase with decreasing energy and the second step
exhibits a resonance at 203 eV, the generation rate
per dpa in fast reactors increases near the core
boundaries and out-of-core areas.
It is in thermalized neutron spectra characteristic
of light and heavy water-cooled reactors, however,
where the 59Ni(n, a) reaction can produce He/dpa
generation rates that are significantly larger than
those characteristic of fusion-derived spectra.
Nickel has five naturally occurring stable isotopes
with 58Ni comprising 67.8% natural abundance, 60Ni
comprising 26.2%, and $6.1% total of 61Ni, 62Ni, and

64
Ni. There is no natural 59Ni or 63Ni at the beginning
of radiation. During irradiation in a highly thermalized neutron spectrum, all nickel isotopes are
strongly transmuted, primarily to the next higher
isotopic number of nickel. 59Ni has a half-life of
76 000 years and is progressively transmuted to 60Ni
while 58Ni is continuously reduced in concentration.
Therefore, the 59Ni concentration rises to a peak at a
thermal neutron fluence of 4 Â 1022 n cmÀ2 where the
59/58 ratio peaks at $0.04 and then declines, as shown
in Figure 10.
This transmutation sequence in nickel is an example of the isotopic shift category of transmutation
defined earlier. For other elements used to make stainless steels, there are no consequences to such a shift
since the total amount of the element is unchanged

0.4

0.0
1021

59

Ni

1022
1023
Thermal fluence (n cm-2)

1024

Figure 10 Transmutation-induced evolution of three
nickel isotopes during irradiation in thermalized
neutron spectra. Reproduced from Garner, F. A.;
Greenwood, L. R. In 11th International Conference on
Environmental Degradation of Materials in Nuclear Power
Systems – Water Reactors; 2003; pp 887–909.
Reproduced from Garner, F. A.; Griffiths, M.; Greenwood,
L. R.; Gilbert, E. R. In Proceedings of the 14th
International Conference on Environmental Degradation
of Materials in Nuclear Power Systems – Water
Reactors; American Nuclear Society, 2010;
pp 1344–1354.

and isotope shifts induce no significant consequences.
However, in the case of nickel there is an intimate
linkage between the displacement and transmutation
processes that arises from the isotope shift.
The recoil of the 59Ni upon emission of the
gamma ray produces only about five displacements
per event, and usually is not a significant addition to
the displacement dose. However, the isotope 59Ni
undergoes three strong reactions with thermal and
resonance ($0.2 keV) neutrons, two of which are
exceptionally exothermic and can significantly add
to the dpa level. These reactions, in order of
highest-to-lowest thermal cross-section, are (n, g) to
produce 60Ni, followed by (n, a) and (n, p) to produce
helium and hydrogen, respectively.
Even at relatively low thermal-to-fast neutron
ratios, the reaction sequence can produce significant

amounts of helium. For example, He/dpa ratios in
the order of $3–8 appm dpaÀ1 can be experienced
along the length of a 316 stainless baffle bolt in
the baffle-former assembly of a pressurized water


Radiation Damage in Austenitic Steels

43

100
Pure nickel
in HFIR-PTP
Percentage increase

80

60
56Fe

40

340 keV
1701 dpa

4

He 4.8 MeV
62 dpa

20

0

20

40

60

80

100

120

140

160

Displacements (dpa) neglecting 59Ni (n, a) 56Fe reaction
Figure 11 Increase in dpa arising from the effect of 59Ni to produce helium when pure nickel is irradiated in the HFIR test
reactor in the peripheral target position (PTP) where the thermal-to-fast ratio is 2.0. Reproduced from Garner, F. A.;
Greenwood, L. R. In 11th International Conference on Environmental Degradation of Materials in Nuclear Power Systems –
Water Reactors; 2003; pp 887–909. The rate of dpa acceleration will be increased $3% further if the 59Ni(n, p) and (n, g)
reactions are taken into account. Reproduced from Garner, F. A.; Griffiths, M.; Greenwood, L. R.; Gilbert, E. R. In Proceedings
of the 14th International Conference on Environmental Degradation of Materials in Nuclear Power Systems – Water
Reactors; American Nuclear Society, 2010; pp 1344–1354.

reactor4,33,39 while comparable rates in fast reactors

are in the order of 0.1–0.2 appm dpaÀ1. In thermalized spectra the latter two reactions can quickly
overwhelm the gas production produced by nickel
at high neutron energies.
As mentioned previously, the thermal neutron
reactions of 59Ni are quite exothermic in nature and
release large amounts of energy, thereby causing
increases in the rate of atomic displacements, and concomitant increases in nuclear heating rates. Nuclear
heating by elastic collisions with high-energy neutrons is usually too small to be of much significance.
The 59Ni(n, a) reaction releases 5.1 MeV, producing a 4.8 MeV alpha particle which loses most of its
energy by electronic losses, depositing significant
thermal energy but producing only $62 atomic displacements per each event. However, the recoiling
56
Fe carries 340 keV, which is very large compared to
most primary knock-on energies, and produces an
astounding $1701 displacements per event.
The thermal (n, p) reaction of 59Ni produces
about one proton per six helium atoms, reflecting
the difference in thermal neutron cross-sections of
2.0 and 12.3 barns, and is somewhat less energetic
(1.85 MeV), producing a total of $222 displacements
per event.7,40 In addition, approximately five displaced atoms are created by each emission-induced
recoil of 60Ni. This reaction occurs at six times higher

rate compared to the 59Ni(n, a) reaction, resulting
from a thermal neutron cross-section of 77.7 barns. In
effect, the dpa rate increases during irradiation due to
the three 59Ni reactions even though the neutron
flux-spectrum may not change.
The major point here is that use of standardized
computer codes to calculate dpa does not track

shifts in isotopic distribution and therefore will
underpredict the dpa level when 59Ni production is
an important consideration.
A strong example of this time-dependent increase
in dpa rate in highly thermalized light water spectra
is shown for pure nickel in Figure 11 for a thermalto-fast ratio of 2.0. Note that the calculated increase
in this figure addresses only the 59Ni(n, a) reaction.
Additional increases occur as a result of the 59Ni(n, p)
and 59Ni(n, g) reactions, resulting in almost doubling
of dpa by the three 59Ni reactions before a calculated
dose of $40 dpa is attained.
Recently, however, an even stronger example of
the linkage of the 59Ni transmutation effect and the
displacement process has been observed.34,35 In-core
thermal-to-fast ratios in heavy water-moderated
reactors such as CANDUs are in the order of $10,
but far from the core the ratio can be near $1000.
Compression-loaded springs constructed of highnickel alloy X-750 were examined after 18.5 years of
operation far from the core and were found to be


44

Radiation Damage in Austenitic Steels

completely relaxed. Calculating the 59Ni contribution, it was deduced that full relaxation occurred
in $3–4 years rather than the 650–700 years one
would predict based on dpa calculated without taking
into account the 59Ni contribution.
Therefore, in this case 59Ni contributed $95% of

the dpa. Additionally, 1100 appm of helium was calculated to have been produced at the mid-section of
the spring in $3 years, with $20 000 appm helium
having been produced when the spring was examined
after 18.5 years of exposure.
There is another consequence of the 59Ni sequence
that causes the temperature to increase during irradiation. At the peak 59Ni level reached at 4 Â 1022 n cmÀ2,
the nuclear heating rates from the energetic (n, a) and
(n, p) reactions are 0.377 and 0.023 W gÀ1 of nickel,
significantly larger than the neutron heating level of
$0.03 W gÀ1 of natural nickel. Thus, an increase in
nuclear heating of $0.4 W gÀ1 of nickel must be
added to the gamma heating rate at the peak 59Ni
level. Fractions of the peak heating rates that are proportional to the current 59Ni level should be added at
nonpeak conditions. Depending on the nickel level of
the steel and the level of gamma heating, which is the
primary cause of temperature increases in the interior
of thick plates, this additional heating contribution
may or may not be significant.
Gamma heating is also a strong function of the
thermal-to-fast (T/F) neutron ratio and the neutron
flux, being $54 W gÀ1 in the center of the HFIR test
reactor where the T/F ratio is $2.0. In pressurized
water reactors at the austenitic near-core internals,
however, the T/F ratios are lower by a factor of 2–10,
depending on location, and the gamma heating rates
in the baffle-former assembly are $1–3 W gÀ1. In this
case, an additional 0.4 W gÀ1 of nuclear heating can
be a significant but time-dependent addition to total
heating, especially for high-nickel alloys.
It should be noted that thermal neutron populations

can vary during an irradiation campaign with consequences not only on 59Ni production but also on gamma
heating levels. In PWRs boric acid is added to the water
as a burnable poison at the beginning of each cycle. As
the 10B burns out the thermal neutron population
increases, leading to an increase in gamma heating and
transmutation.3,4 Over successive cycles there is a sawtooth variation of gamma heating rate in the baffleformer assembly and therefore in DT, with the latter
reaching values as large as Æ20  C in the worst case.
Additionally, another concern may arise in that
small radiation-induced nickel-rich phases such as
g0 , Ni-phosphides, and G-phase may become less
stable. This concern arises due to cascade-induced

dissolution as the 56Fe from the 59Ni(n, a) reaction
recoils within the precipitates, thereby altering the
phase evolution in thermalized neutron spectra compared to nonthermalized spectra typical of fast reactors. These precipitates are known to form as a direct
result of irradiation and contribute to hardening,
swelling, and irradiation creep processes.1 The size
of these precipitates at PWR-relevant temperatures
(290–400  C) is often comparable to or smaller than
the $80 nm range of the recoiling 56Fe atom.
Finally, another significant source of helium can
arise from the implantation of energetic helium
resulting from collisions with neutrons into the surface layers of helium gas-pressurized or gas-cooled
components, often involving hundreds and often
thousands of appm of injected helium. In gas-cooled
reactors helium injection has been investigated as a
possible degradation mechanism of alloy surfaces.41
In fast reactor fuel cladding helium was found to
be injected into the inner surface, coming from two
major sources, ternary fission events (two heavy fission fragments plus an alpha particle) in the fuel and

from helium recoiling from the pins’ helium cover
gas as a result of collisions with neutrons.42
The injection rates from these two sources of
injected helium are slowly reduced during irradiation, however, as heavy fission gases build up in the
space between the fuel pellet and the cladding.
These gases slow down the energetic helium atoms,
reducing their energy sufficiently to prevent most of
them from reaching the cladding. Helium injection
at high levels was also found on the inner surface of
helium-pressurized creep tubes.42 Although helium
injection tends to saturate in fuel pin cladding with
increasing dose, it does not saturate in pressurized
tubes due to the lack of increasing fission gases to
reduce the range of helium knock-ons in the gas
phase.
Some studies have cited this early source of helium
as contributing to the embrittlement of fuel pin cladding and its poor performance during transient heating
tests,43 although more recent studies have linked the
major mechanism to delayed grain boundary attack by
the fission products cesium and tellurium.44,45

4.02.5 Evolution of RadiationInduced Microchemistry and
Microstructure
When metals are subjected to displacive irradiation,
especially at elevated temperatures, an intricate
and coordinated coevolution of microstructure and


Radiation Damage in Austenitic Steels

microchemistry commences that is dependent primarily on the alloy starting state, the dpa rate, and
the temperature, and secondarily dependent on variables such as He/dpa rate and applied or internally
generated stresses.
In general, the starting microstructure and microchemistry of the alloy determine only the path taken
to the radiation-defined quasi-equilibrium state,
and not the final state itself. If an alloy experiences
enough displacements, it effectively forgets its starting state and arrives at a destination determined only
by irradiation temperature and dpa rate. This quasiequilibrium or dynamic-equilibrium state consists of
microstructural components existing at relatively
fixed densities and size distributions, but individual
dislocations, loops, precipitates, or cavities at any one
moment may be growing, shrinking, or even disappearing by shrinkage or annihilation.
The displacement process produces two types
of crystalline point defects, vacant crystalline positions (vacancies) and displaced atoms in interstitial
crystalline positions (interstitials). These two defect
types are both mobile, but move with different diffusional modes and at vastly different velocities,
with interstitials diffusing much faster than vacancies. Therefore it is obvious that all diffusion-driven
processes will be strongly affected by radiation.
Both defect types have the ability to recombine
with the opposite type (annihilation) or to form
agglomerations of various types and geometries.
These agglomerations and their subsequent evolution
alter both the microstructure and elemental distribution of the alloy.
It is important to note that interstitial agglomerations are constrained to be two-dimensional, while
vacancies can agglomerate in both two-dimensional
and three-dimensional forms. This dimensional disparity is the root cause of the void swelling phenomenon covered in a later section.
The developing ensemble of various defect
agglomerations with increasing dose induces significant time-dependent and dose-dependent changes
in physical and mechanical properties, as well as
resulting in significant dimensional distortion. Most

importantly, under high displacement rates stainless
steels and other alloys are driven far from equilibrium
conditions as defined in phase diagrams, affecting not
only phase stability but also all physical, mechanical,
and distortion processes that involve phase changes in
their initiation or evolution.
During irradiation, the phase evolution can be
significantly altered, both in its kinetics and in the
identity and balance of phases that form.46,47 Phases

45

can be altered in their composition from that found in
the absence of irradiation, and new phases can form
that are not found on the equilibrium phase diagram
of a given class of steels. In 300 series stainless steels
these new or altered phases have been classified as
radiation-induced phases, radiation-modified phases,
and radiation-enhanced phases.48–51 These classifications are equally applicable to phases formed in other
classes of steel.
Radiation-induced alterations of microstructure
and microchemistry occur because new driving forces
arise that do not occur in purely thermal environments. The first of these new driving forces is the
presence of very large supersaturations of point
defects, especially at relatively low irradiation temperatures (250–550  C). Not only are vacancies
present in uncharacteristically high levels, thereby
accelerating normal vacancy-related diffusional processes, but interstitials are also abundant. Solutes that
can bind with either type of point defect tend to flow
down any microstructurally induced gradient of that
defect, providing a new mechanism of solute segregation referred to as solute drag.52 This mechanism

has been proposed to be particularly important for
binding of smaller solute atoms such as P and Si, and
sometimes Ni, with interstitials.
A second new driving force is the inverse
Kirkendall effect 53 whereby differences in elemental
diffusivity via vacancy exchange lead to segregation
of the slowest diffusing species at the bottom of sinkinduced vacancy gradients. This mechanism is particularly effective in segregating nickel in austenitic
Fe–Cr–Ni alloys at all sinks which absorb vacancies,
leading to nickel-rich shells or atmospheres on
grain boundaries and other preexisting or radiationproduced microstructural sinks. This type of segregation arises because the elemental diffusivities of
Fe–Cr–Ni alloys are significantly different, with
DCr > DFe > DNi at all nickel levels.54–57
A third new driving force results from the action of
the other two driving forces when operating on
microstructural sinks that are produced only in irradiation environments. These are Frank interstitial
loops, helium bubbles, and voids that may have developed from helium bubbles. Precipitates are often
observed to form and to co-evolve on the surface of
such radiation-induced sinks. Examples of typical
radiation-induced microstructures in stainless steels
are shown in Figures 12–15. These microstructural
sinks have been implicated as participating in the
evolutionary path taken by the precipitates and
thereby influencing the microchemical evolution of
the matrix.1,58–60


46

Radiation Damage in Austenitic Steels

(a)

CW 316 SS, thimble tube
70 dpa, 315 ºC

50 nm

(b)

CW 316 SS, thimble tube
33 dpa, 290 ºC

50 nm
(c)

CW 316 SS, thimble tube
33 dpa, 290 ºC

50 nm
Figure 12 Frank loops observed in a 316 stainless flux thimble from a PWR power reactor (a) 70 dpa, 315  C and (b) 33 dpa,
290  C imaged edge-on on one set of the four (111) planes using the dark-field relrod technique. Reproduced from
Edwards, D. J.; Garner, F. A.; Bruemmer, S. M.; Efsing, P. G. J. Nucl. Mater. 2009, 384, 249–255. The image in (c) is from
Frank loops that are slightly inclined to the beam direction imaged using a relrod in the diffraction pattern.

G-phase

50 nm
Figure 13 Electron micrograph of radiation-induced
voids in annealed ‘PCA’ stainless steel irradiated in the
ORR water-cooled test reactor at 500  C to 11 dpa.

The largest voids have radiation-induced G-phase
particles attached to them that are rich in Ni, Si, and Ti.
Reproduced from Maziasz, P. J. J. Nucl. Mater. 1989,
169, 95–115.

Minor solute elements such as Si and P have much
higher diffusivities than those of Fe, Ni, and Cr and
also participate in the segregation process. Additionally, these elements increase the diffusivities of the
major elements Fe, Ni, and Cr.54
When the solute drag mechanism, operating
between interstitials and smaller size Si and P atoms,
combines with nickel segregation via the inverse
Kirkendall mechanism, phases that are rich in nickel,
silicon, or phosphorus often form (g0 , G-phase and
Ni2P for example), although in 300 series stainless
steels these phases do not form thermally. Other
phases that are normally stable in the absence of
radiation (carbides, intermetallics) can be forced during irradiation to become enriched in these elements.1
The removal of nickel, silicon, and phosphorus
from the matrix by radiation-induced precipitation
exerts a large effect on the effective vacancy diffusivity.57,61 On a per atom basis, phosphorus has been


Radiation Damage in Austenitic Steels

50 nm
Figure 14 Void swelling ($1%) and M23C6 carbide
precipitation produced in annealed 304 stainless steel after
irradiation in the reflector region of the sodium-cooled
EBR-II fast reactor at 380  C to 21.7 dpa at a dpa rate of

0.84 Â 10À7 dpa sÀ1. Reproduced from Garner, F. A.;
Edwards, D. J.; Bruemmer, S. M.; et al. In Proceedings,
Fontevraud 5, Contribution of Materials Investigation to the
Resolution of Problems Encountered in Pressurized Water
Reactors; 2002; paper #22. Dislocations and dislocation
loops are present but are not in contrast.

Figure 15 Reverse contrast image showing void and line
dislocation microstructure in Fe–10Cr–30Mn model alloy
irradiated in FFTF fast reactor to 15 dpa at 520  C. Average
void sizes are $ 40 nm. Reproduced from Brager, H. R.;
Garner, F. A.; Gelles, D. S.; Hamilton, M. L. J. Nucl. Mater.
1985, 133–134, 907–911. Frank loops have unfaulted to
produce a line dislocation network whose segments end
either on void surfaces or on upper and lower surfaces of
the thin microscopy specimen. The voids are coated with
ferrite phase due to Mn depletion from their surfaces via the
Inverse Kirkendall effect.

shown to exert an even larger effect on the effective
vacancy diffusivity57 and its removal into Ni2P and
other precipitates has a strong influence on matrix

47

diffusion. Silicon is the next most effective element
on a per atom basis. As the effective vacancy diffusion
coefficient falls with decreasing matrix levels of Ni,
Si, and P, conditions for void nucleation become
more favorable.

The radiation-induced evolution of diffusional
properties has been strongly implicated in determining the transient duration before void swelling accelerates.1 This evolution often does not necessarily
proceed by only one path but occurs in several interactive stages. Some phases such as nickel phosphides
and TiC, especially when precipitated on a very
fine scale, are thought to be beneficial in resisting
the evolution of nickel silicide type phases.59,62,63
It has been shown, however, that continued radiationinduced segregation eventually overwhelms these
phases by removing critical elements such as Ni and
Si from solution, causing their dissolution and
replacement with nickel-rich and silicon-rich phases
that coincide with accelerated swelling.63–65
In high-nickel alloys that normally form the g0 and
00
g ordered phases, irradiation-induced segregation
processes do not significantly change the identity or
composition of the phases, but can strongly change
their distribution, dissolving the original distribution
but plating these phases out on voids, dislocations,
and grain boundaries, with the latter often leading to
severe grain boundary embrittlement.66,67
The original dislocation microstructure quickly
responds to mobile displacement-generated point
defects, increasing their mobility and leading to
reductions in dislocation density and distribution
in the cold-worked steels most frequently used for
fuel cladding and structural components.1 These
dislocations are quickly replaced by new microstructural components, often at very high densities,
with two-dimensional interstitial Frank loops first
dominating the microstructure, then generating new
line dislocations via unfaulting and interaction of

loops. In well-annealed alloys there are very few preexisting dislocations but the same radiation-induced
loop and dislocation processes occur, eventually
reaching the same quasi-equilibrium microstructure
reached by cold-worked alloys.
At lower temperatures found in water-cooled
test reactors especially, the microstructural features
appear to be three-dimensional vacancy clusters
or stacking fault tetrahedra and two-dimensional
vacancy or interstitial platelets, which are probably
also small dislocation loops. These ‘defect clusters’ at
temperatures below $300 C are usually too small to
be easily resolved via conventional transmission


48

Radiation Damage in Austenitic Steels

Figure 16 (top) Spiral distortion of 316-clad fuel pins induced by swelling and irradiation creep in an FFTF fuel assembly
where the wire wrap swells less than the cladding. Reproduced from Makenas, B. J.; Chastain, S. A.; Gneiting, B. C.
In Proceedings of LMR: A Decade of LMR Progress and Promise; ANS: La Grange Park, IL, 1990; pp 176–183; (middle)
Swelling-induced changes in length of fuel pins of the same assembly in response to gradients in dose rate, temperature, and
production lot variations as observed at the top of the fuel pin bundle. Reproduced from Makenas, B. J.; Chastain, S. A.;
Gneiting, B. C. In Proceedings of LMR: A Decade of LMR Progress and Promise; ANS: La Grange Park, IL, 1990; pp 176–183;
(bottom) swelling-induced distortion of a BN-600 fuel assembly and an individual pin where the wire swells more than
the cladding. Reproduced from Astashov, S. E.; Kozmanov, E. A.; Ogorodov, A. N.; Roslyakov, V. F.; Chuev, V. V.;
Sheinkman, A. G. In Studies of the Structural Materials in the Core Components of Fast Sodium Reactors; Russian Academy
of Science: Urals Branch, Ekaterinburg, 1984; pp 48–84, in Russian.

electron microscopy and are often characterized as

either ‘black dots’ or ‘black spots.’ These dots are
generally thought to be very small Frank interstitial
loops.
The cluster and dislocation loop evolution is frequently concurrent with or followed by the loss
or redistribution of preexisting precipitates. Most
importantly, new radiation-stabilized precipitates at
high density often appear with crystal structure and
composition that are not found on an equilibrium
phase diagram for austenitic steels.
As a consequence of these various processes the
microstructure at higher doses often develops very
high densities of crystallographically faceted, vacuumfilled ‘cavities’ called voids, thought to nucleate on
helium clusters formed by transmutation, although
residual gases in the steel often help nucleate voids
at lower concentrations. Voids have frequently been
observed in charged particle irradiations where no
helium was introduced.

The void phenomenon is not a volumeconservative process and the metal begins to ‘swell’
as the microscopic voids in aggregate contribute
to macroscopic changes in dimension, sometimes
increasing the metal volume by levels of many tens
of percent.
Concurrently, the dislocation microstructure
responds to the local stress state, moving mass via a
volume-conservative process designated irradiation
creep. In general, irradiation creep is not a directly
damaging process but it can lead to component
failures resulting from distortion that causes local
blockage of coolant flow or strong postirradiation

withdrawal forces. Both swelling and irradiation
creep are interrelated and are interactive processes
that can produce significant distortions in component
dimensions. Figure 16 shows some pronounced
examples of such distortion.68,69
Eventually, the microstructural/microchemical
ensemble approaches a quasi-equilibrium condition


Radiation Damage in Austenitic Steels

CW 316 SS, thimble tube
70 dpa, 330 ЊC

49

CW 316 SS, thimble tube
70 dpa, 330 ЊC

Bubbles on grain boundary

Matrix bubbles
1.6 ϫ 1023 m−3

20 nm

−256 nm UF

20 nm

Figure 17 High densities of nanocavities observed using highly under-focus conditions in a PWR flux thimble tube
constructed from cold-worked 316 stainless steel. Reproduced from Edwards, D. J.; Garner, F. A.; Bruemmer, S. M.;
Efsing, P. G. J. Nucl. Mater. 2009, 384, 249–255. The irradiation conditions were $70 dpa and 330  C, producing $600 appm
He and 2500 appm H. Note the high density of cavities on the grain boundary.

or ‘saturation’ state, usually at less than 10 dpa for
mechanical properties but at higher doses for swelling.
As a consequence, the mechanical properties tend to
stabilize at levels depending primarily on temperature and to a lesser extent on dpa rate. The two major
deformation processes, swelling and irradiation
creep, do not saturate but reach steady-state deformation rates when quasi-equilibrium microstructures
are attained. This coupling of saturation microstructure with steady-state behavior has been characterized as ‘persistence.’70
Interestingly, the saturation states of each property
change are almost always independent of the starting
thermal–mechanical state of the material.1,70,71 If
irradiation continues long enough, the memory of
the starting microstructural state and the associated
mechanical properties is almost completely lost. The
only deformation-induced microstructural component
that succeeds in resisting this erasure process is that of
preexisting, deformation-induced twin boundaries.
If this quasi-equilibrium is maintained to higher
neutron exposure no further change occurs in the
steel’s mechanical properties. However, some slowly
developing second-order processes are nonsaturable
and are often nonlinear. Eventually, these processes
force the system to jump toward a new quasiequilibrium. These new states usually arise from
either the microstructural or microchemical evolution, with voids dominating the former and the
latter involving continued segregation, continued
transmutation, or a combination of these factors.70–72

A number of such late-stage changes in quasiequilibrium state are discussed later in this paper.

4.02.6 A Cross-Over Issue Involving
Radiation-Induced Microstructural
Evolution and Transmutation
Recently, it been discovered that significant levels of
hydrogen can be stored in bubbles and voids in both
stainless steels and pure nickel when the hydrogen is
cogenerated with helium, especially in light water
spectra where there are also environmental sources
of hydrogen.73–75 It was shown in these studies that
this phenomenon is a direct result of the 59Ni nuclear
reactions. Previously, it was a long-standing perception that such storage could not occur at reactorrelevant temperatures.
The retained hydrogen levels are in significant
excess of the levels predicted by Sievert’s Law and
appear to be increasing with both cavity volume
and neutron fluence. Since these gases are known to
assist in nucleation and stabilization of cavities, it is
expected that the nonlinear 59Ni reactions discussed
earlier may lead to a rapidly developing, nonlinear,
cavity-dominated microstructure in stainless steels
irradiated at temperatures characteristic of pressurized water reactors.
Figure 17 presents such a microstructure observed
in a PWR flux thimble tube (cold-worked 316 stainless steel) at $70 dpa and 330  C.76 There is a very
high density (>1017 cmÀ3) of nanocavities with diameters <3 nm in both the alloy matrix and especially
on grain boundaries. The measured concentrations of
600 appm He and 2500 appm H in this specimen are
thought to reside primarily within the cavities. Most
importantly, these cavities are essentially invisible

50

Radiation Damage in Austenitic Steels

4.02.7 Radiation-Induced Changes
in Mechanical Properties

under well-focused imaging conditions and can only
be imaged using very large levels of under-focus.
This implies that previous studies on similar materials may have overlooked such cavity-dominated
structures.
When this specimen and near-identical specimens
were subjected to slow strain rate testing after irradiation, the fracture surface was indicative of $100%
intergranular stress corrosion cracking (IGSCC),
with lower doses and gas levels producing proportionally less IGSCC.77 As hydrogen is known to be a
contributor to grain boundary cracking, it appears
plausible that hydrogen storage may accelerate the
cracking process and that higher exposure will lead to
an increasing susceptibility to cracking. This issue
may therefore become increasingly important as
PWRs previously licensed for 40 years are being
considered for life extension to 60 and possibly
80 years.

Long before the onset of significant phase evolution
or void swelling is observed, the first manifestation of
the radiation-induced microstructural/microchemical
evolution appears in changes of the mechanical properties. As shown in Figure 18 the stress–strain diagrams of stainless steels begin to change significantly
even at very low dpa levels. The strength of the alloy

increases, the elongation decreases, and there is a
progressive decrease in work-hardening. This behavior is dependent somewhat on test temperature but is
not very sensitive to neutron spectrum.
Movement of dislocations in metals during
deformation following irradiation is impeded by the
microstructural components produced by radiation
(dislocations, dislocation loops, voids, bubbles, precipitates) and therefore the strength of annealed steel
1000

1000

316 SS tested at 288 ЊC

316 SS tested at RT
800

600
Unirradiated
0.0001 dpa
0.001 dpa
0.01 dpa
0.1 dpa
0.78 dpa

400
200
0
0.0

(a)

Eng. stress (MPa)

Eng. stress (MPa)

800

0.2

0.4
0.6
Eng. strain

400
Unirradiated
0.01 dpa
0.1 dpa

200
0
0.0

1.0

0.8

600

0.2

0.4
0.6
Eng. strain

(b)

0.8

1.0

1000

Eng. stress (MPa)

800

EC316LN tested at RT
10.7 dpa
3.64 dpa
0.86 dpa
0.45 dpa

600

2.53 dpa
1.36 dpa

0.0 dpa

400

200
0
0.0

(c)

0.2

0.4
0.6
Eng. strain

0.8

1.0

Figure 18 Engineering stress–strain curves for irradiated austenitic stainless steels: (a) annealed 316 SS irradiated in HFIR
mixed spectrum reactor at 60–100  C and tested at 25  C, (b) annealed 316 SS irradiated in HFIR at 350  C and tested at
288  C, and (c) annealed EC316LN irradiated in the LANSCE spallation neutron and proton spectrum at 60–100  C and tested
at 25  C. Reproduced from Kim, J. W.; Byun, T. S. J. Nucl. Mater. 2010, 396, 10–19.


Radiation Damage in Austenitic Steels

increases. The strength increase usually saturates at
relatively low exposure levels (<10 dpa) as shown
in Figure 19, reflecting a similar saturation of microstructural densities. Since the concentration of
most radiation-induced microstructural components
decreases with increasing temperature above $300  C,
one would expect that the saturation strength would

also decrease with increasing temperature, as is shown
in Figures 20–23.

1000
316LN
316

Yield strength (MPa)

800
304
600
316

304, 304L

400

316, 316LN
PCA

316
200
Unirradiated
304, 316, PCA alloys
0
0.0001

0.001

0.01
0.1
1
Neutron dose (dpa)

10

100

Figure 19 Strengthening of various annealed 300 series
stainless steels versus dpa in various water-cooled reactors
at relatively low temperatures (280–330  C). Reproduced
from Pawel, J. P.; Ioka, I.; Rowcliffe, A. F.; Grossbeck, M. L.;
Jitsukawa, S. In Effects of Radiation on Materials: 18th
International Symposium; ASTM STP 1325; 1999;
pp 671–688. At these temperatures strengthening saturates
at $10 dpa.

Frank loops

Yield strength (MPa)

1400

Cavities, precipitates

1200
1000

Defect clusters

800
600
400
200
0
0

Unirradiated
100

200
300
Temperature ( ЊC)

400

500

Figure 20 Radiation-induced strengthening of annealed
300 series steels versus irradiation temperature and the
microstructural components causing the strengthening.
Note the peak strengthening at 300  C followed by a decline
at higher temperatures. Reproduced from Pawel, J. P.;
Ioka, I.; Rowcliffe, A. F.; Grossbeck, M. L.; Jitsukawa, S.
In Effects of Radiation on Materials: 18th International
Symposium; ASTM STP 1325; 1999; pp 671–688.

51

Irradiation of cold-worked steels also leads to
strengthening at lower temperatures but softening
can occur at higher temperatures if the saturation
strength level at a given temperature is below the
starting strength, as seen in Figure 21. Most importantly, both annealed and cold-worked steels converge to the same saturation level when irradiated
at the same dpa rate and temperature as seen in
Figure 22.78
Similar convergence behavior has been observed
in the evolution of microhardness.79 Note also that
radiation-induced changes in strength are roughly
independent of composition within the annealed
300 series stainless steels, especially at lower irradiation temperatures, as shown by Figure 19.
Such convergence behavior has been observed
many times, but there are exceptions; for example,
cold-worked steels converge in their notch tensile
strength, but not to the level reached by annealed
steels.80 Such behavior is usually observed in steels that
twin heavily during deformation and were irradiated at
low temperatures that resist recrystallization. Twin
boundaries are not easily erased by displacements,
so their hardening contribution persists.
Concurrent with an increase in radiation-induced
hardening is a loss of ductility,81–83 as shown in
Figures 23 and 24.
The concept of saturation or persistence of
mechanical properties, especially with respect to temperature, applies to the most recent irradiation
temperature, as demonstrated by comparing isothermal and nonisothermal histories. In Figure 25 the
mechanical properties of three model alloys are seen
to converge during isothermal irradiation without
being affected by composition, He/dpa ratio, and

mechanical starting state.84 In Figure 26, however,
an early detour in temperature led to differences
from isothermal behavior, but these differences disappeared when the intended isothermal temperature
was reestablished.84
Previous saturation states are soon forgotten, usually by $5 dpa, but only if the hardening components
are easily erased and replaced at the new temperature. If hardening arises primarily from dislocation
loops and dislocations, this condition is easily met.
If the primary hardening arises from a fine density
of voids and especially bubbles produced at lower
temperatures, then the microstructural memory cannot be easily erased, even at much higher temperatures. An example is shown in Figure 27 where a series
of Fe–Cr–Ni ternary austenitic alloys were irradiated
at 400 and 500  C in ORR at high He/dpa ratios


52

Radiation Damage in Austenitic Steels

1100
1000
900

371 ЊC

Yield strength (MPa)

800

427 ЊC

700
483 ЊC
600
500
538 ЊC

400

593 ЊC
300

649 ЊC
704 ЊC

200

760 ЊC

100
0

816 ЊC
0

1

2

3

4

Neutron fluence

5

6

(n cm–2)

7

8

9

10 ϫ 1022

(E > 0.1 MeV)

Figure 21 Evolution of yield strength in 20% cold-worked 316 stainless steel irradiated in EBR-II over a wide range of
temperatures. Reproduced from Garner, F. A.; Hamilton, M. L.; Panayotou, N. F.; Johnson, G. D. J. Nucl. Mater. 1981, 103
and 104, 803–808.

600
650 ЊC

20% Cold-worked

400

200

Annealed
0
538 ЊC
20% Cold-worked

Yield strength (MPa)

600
400
200

Annealed

0
427 ЊC
800
600

20% Cold-worked

400

Annealed

200
0
0

1

2

3

4

5

6
22

7

8

9

10

–2

Neutron fluence (10 n cm )
Figure 22 Influence of temperature and neutron exposure
on evolution of yield strength in both annealed and 20%
cold-worked AISI 316 irradiated in EBR-II, showing that the
saturation strength level is independent of starting
condition, converging at doses of 5–15 dpa. Reproduced
from Garner, F. A.; Hamilton, M. L.; Panayotou, N. F.;

Johnson, G. D. J. Nucl. Mater. 1981, 103 and 104,
803–808.

(27–58 appm dpaÀ1) and 395 and 450  C in EBR-II at
very low He/dpa ratios (0.7–1.2 appm dpaÀ1).85
Note that there are very significant differences in
hardening observed between the two experiments
and that the differences arose primarily from a very
large difference in cavity density, a difference that
was too large to be explained in terms of helium
content alone. It was later shown that the ORR
experiment suffered a very large number (237 over
2 years) of unrecognized negative temperature setbacks of 1–2 h, with decreases varying from 50 to
500  C.86 Even though the total dpa accumulated
during these setbacks was only $1% of the total
dose, the frequent bloom of high densities of small
Frank loops at lower temperatures provided a very
large periodic increase in nucleation sites for helium
bubbles on the new Frank loops that significantly
strengthened the matrix. The loops could subsequently dissolve but the bubbles could not.
In addition to temperature, the most prominent
irradiation variable is the dpa rate and it is known
that the microstructural densities, especially Frank
loops and voids, are known to increase in concentration as the dpa rate increases. Various radiation-stable
phases such as w0 are also known to be flux-sensitive,
while other phases such as carbides and intermetallics
are more time-sensitive.1
Thus, it is not surprising that some sensitivity to
dpa rate might be observed in strength properties, as

Radiation Damage in Austenitic Steels

53

Test temperature = Irradiation temperature
1200

40

Yield strength (MPa)

800

400 ЊC

450–500 ЊC

600

525–565 ЊC

400

Uniform elongation (%)

215–245 ЊC

1000

20
525–565 ЊC

10
5

2

400 ЊC

1
585–625 ЊC

200
0

1

2

3

215–245 ЊC

0.5

5 ´ 1026

4

0

1

2

3

5 ´ 1026

4

Neutron fluence (n m-2) E > 0.1 MeV

Neutron fluence (n m-2) E > 0.1 MeV

Figure 23 Neutron-induced changes in tensile properties of annealed 1.4988 stainless steel irradiated in the DFR fast
reactor. Reproduced from Ehrlich, K. J. Nucl. Mater. 1985, 133–134, 119–126. Ductility declines as strength increases.

50
20% CW 316
25% CW PCA

40
Uniform elongation (%)

Yield stress (MPa)

1200

800

400

0

SA 316L
SA PCA

SA 316L
SA PCA
20% CW 316
25% CW PCA

30

20

10

0
0

2

4

6
dpa

8

10

0

400
800
Yield stress (MPa)

1200

Figure 24 Strengthening and ductility loss observed in two stainless steels irradiated in the HFIR, HFR, and R2 mixed
spectrum reactors at 250  C at He/dpa ratios ranging from 10 to 35 appm dpaÀ1. Note that both annealed and cold-worked
(CW) steels quickly converge to the same elongation levels, while convergence of strength is not developing as quickly.
Reproduced from Elen, J. D.; Fenici, P. J. Nucl. Mater. 1992, 191–194, 766–770.

suggested by the behavior shown in Figure 28 where
both the transient rate of strength rise and saturation
strength appear to increase with increasing dpa rate.
Unfortunately, this figure does not represent a single
variable comparison, and by itself is not sufficiently
convincing evidence of flux sensitivity. The data
shown in Figure 29 is much closer to a single variable
comparison, indicating that the transient rise may or
not be somewhat flux-sensitive, depending on the
details of the microstructural evolution of each alloy.
The authors of this study used microscopy to confirm
the microstructural origins of the observed differences of behavior as a function of dpa rate.

More recently, Chatani and coworkers showed
that at relatively low irradiation temperatures characteristic of boiling water reactors, the radiationinduced increments in strength of 304 stainless steel
increased by the 1/4 power of the increase in dpa
rate.87 It was demonstrated that the black-spot microstructure dominated the strengthening. It was also
shown that the concentration of black spots varied
with the square root of the flux as expected, and it is
known that hardening varies with the square root of
the loop density, thereby producing a fourth-root
dependence. Thus, in the absence of any significant
microchemical or phase stability contributions, it


54

Radiation Damage in Austenitic Steels

1000

Yield strength (MPa)

365 ЊC
800

600
Cold-worked
Annealed
400
25 Ni + 0.04 P

25 Ni

200

45 Ni
With 59Ni
Without

~0.5 and ~15
appm He per dpa

0

Total elongation (%)

40

25 Ni + 0.04 P

25 Ni

45 Ni

30
Annealed
20
Cold-worked

10

0

0

10

20

0

10

20

0

10

20

30

dpa
Figure 25 Influence of starting state, composition of isotopically doped alloys and He/dpa ratio on changes in mechanical
properties produced during isothermal irradiation at 365  C in FFTF. Reproduced from Garner, F. A.; Hamilton, M. L.;
Greenwood, L. R.; Stubbins, J. F.; Oliver, B. M. In Proceedings of 16th ASTM International Symposium on Effects of Radiation
on Materials; ASTM STP 1175; 1992, pp 921–939.

appears that radiation-induced strengthening is
affected by dpa rate but not very strongly.
The loss of ductility proceeds in several stages,
first involving convergence of the yield and ultimate

strengths as shown in Figures 29 and 30, such that
a loss of work-hardening occurs and very little
uniform elongation is attained. As the irradiation proceeds, there is a progressive tendency toward flow
localization followed by necking. As seen in Figure 31
the failure surface shows this evolution with increasing dose.
The flat faces observed at highest exposure in
Figure 31 are often referred to as ‘channel fracture’
but they are not cleavage faces. They are the result
of intense flow localization, resulting from the first
moving dislocations clearing a path of radiationproduced obstacles, especially Frank loops, and thereby
softening the alloy along that path. It is not possible to
remove the voids by channeling but the distorted

voids provide a microstructural record of the flow
localization as shown in Figure 32. Linkage of the
elongated voids is thought to contribute to the failure.
Such a failure surface might best be characterized
as ‘quasi-embrittlement’, which is a suppression of
uniform deformation, differentiating it from true
embrittlement, which involves the complete suppression of the steel’s ability for plastic deformation.
This distinction is made because under some conditions quasi-embrittlement can evolve into true
embrittlement.
The tendency toward quasi-embrittlement grows
with increasing swelling but the alloy is actually
softening with increasing swelling rather than hardening. As shown in Figure 33 brittle fracture
(defined as strength reduction with zero plasticity)
of a Fe–18Cr–10Ni–Ti stainless wrapper in BOR-60
at 72 dpa maximum was observed at positions where
peak swelling occurs.88 Some decrease of strength is

Radiation Damage in Austenitic Steels

55

800

Yield strength (MPa)

495 ЊC
Original
series

600

400

45 Ni

25 Ni + 0.04 P

25 Ni
200

Isothermal
repeat series

With 59Ni
Without

~0.5 and ~5.0
appm He per dpa

Total elongation (%)

0

40

25 Ni

30

Isothermal
repeat
series

45 Ni

25 Ni + 0.04 P

Original
series

20

10

0

0

20

40

0

40

20

0

20

40

60

dpa
Figure 26 Comparison of isothermal and nonisothermal behavior on convergence behavior. The original target
temperature of 495  C was maintained for some time but thereafter there was a large, relatively brief over-temperature event,
followed by a prolonged and significant under-temperature event. Reproduced from Garner, F. A.; Hamilton, M. L.;
Greenwood, L. R.; Stubbins, J. F.; Oliver, B. M. In Proceedings of 16th ASTM International Symposium on Effects of Radiation
on Materials; ASTM STP 1175; 1992, pp 921–939. When the target temperature was reestablished in the second and third
irradiation segments the mechanical properties returned to the isothermal destination.

observed with increasing irradiation temperature, but
the primary strength reduction for specimens tested at

the irradiation temperature arises from the magnitude
of swelling. Testing at temperatures below the irradiation temperature (e.g., 20  C) demonstrates the same
dependence on swelling and irradiation temperature,
but the strength and plasticity values are higher. As
expected, the strengths for tests conducted at 800  C
are uniformly much lower than that observed at lower
temperatures, but there is an absence of any relationship between strength and swelling at this temperature.
As shown in Figure 34 failure surfaces at high
swelling levels exhibit transgranular cup-cone morphology where failure proceeded by micropore coalescence arising from stress concentration between
deforming voids.88 Similar fracture morphology has
been observed in studies on other stainless steels.1

Although voids and bubbles initially serve to
harden the microstructure,78 large swelling levels
allow previously second-order void effects to become
dominant.1,88,89 One of these second-order effects is
the strong decrease of elastic moduli at high swelling
levels. All three elastic moduli decrease initially
at $2% per each percent of void swelling.90–93
At >10% swelling this leads to significant reduction
in strength.
As a consequence, the slope of the elastic region
(Young’s modulus) of the stress–strain curve decreases,
and more importantly, the barrier strengths of all sinks
decrease as the shear modulus likewise decreases.
Therefore, the yield and ultimate strengths decrease
with increasing swelling, even though the elongation
strongly decreases. Similar behavior has also been
observed in pure copper.94

Radiation Damage in Austenitic Steels

25

Average DYS (MPa)

800

Ni
35

1024

45

7
Cr 15
20

700
600
MFE-4

500
400
300

2.3

2.1

400 ЊC

2.4

1023

1.9
9.4

5.5

5.6

3.4

1022
23 nm

1021

1020

AD-1

100
400

500

600

Temperature (ЊC)

2.9

500 ЊC

23

395 ЊC

35

200

0
300

MFE-4 experiment in ORR

1.9 nm

Cavity density (m-3)

56

1019

25

AD-1 experiment
in EBR-II

20

43

30
Nickel (wt%)

450 ЊC 40

40

50

Figure 27 Comparison of hardening of Fe-YCr-XNi ternary alloys observed in the MFE-4 experiment in ORR at $13 dpa
and the AD-1 experiment in EBR-II at $10 dpa. Reproduced from Hamilton, M. L.; Okada, A.; Garner, F. A. J. Nucl. Mater.
1991, 179–181, 558–562; Garner, F. A.; Sekimura, N.; Grossbeck, M. L.; et al. J. Nucl. Mater. 1993, 205, 206–218. Higher
levels of hardening in ORR arise from a refinement and elevation of cavity density arising from frequent negative temperature
excursions at high He/dpa rates. Mean cavity sizes are shown next to each data point.

800

0.2% proof stress (MPa)

Phénix
Rapsodie

600

400

200

0

10

20
30
Fluence (dpa)

40

50

Figure 28 Differences in strength change exhibited by
annealed 316 stainless steel after irradiation at 390  C in
the PHENIX and RAPSODIE fast reactors. Dupouy, J. M.;
Erler, J.; Huillery, R. In Proceedings International
Conference on Radiation Effects in Breeder Reactor
Structural Materials, Scottsdale; The Metallurgical Society
of AIME: New York, 1977, pp 83–93. Phe´nix operated at a
displacement rate that was $three times higher than that of
RAPSODIE.

The nature of the void-related failure changes from

quasi-embrittlement to true embrittlement for tests at
or near room temperature, demonstrating another
example of a late-term second-order process growing
to first-order importance at higher swelling levels.

Hamilton and coworkers observed that above $10%
swelling the previously established saturation strength
level of 316 stainless steel suddenly increased very
strongly in room temperature tensile tests.95 Similar
results were observed in Russian steels.96,97 As shown
in Figures 35 and 36 the failure surfaces in such tests
had rotated from the expected 45 (relative to the
stress axis) to 90 as swelling approached 10%, indicating complete brittle failure, as also indicated by the
fully transgranular nature of the failure surface. Concurrently, the ductility vanished and the tearing modulus plunged to zero, indicating no resistance to crack
propagation. Once a crack has initiated it then propagates completely and instantly through the specimen.
Neustroev and coworkers observed such failures in
Russian steels that are subject to greater amounts of
precipitation and determined that the critical microstructural condition was not defined solely by the level
of swelling, but by the obstacle-to-obstacle distance of
the void-precipitate ensemble, indicating that stress
concentration between obstacles was one contributing
factor.96 However, it was the progressive segregation
of nickel to increasing amounts of void surface and
the concurrent rejection of chromium from the surfaces that precipitated the rather abrupt change in
failure behavior.1,95 This late-term void-induced microchemical evolution induces a martensite instability in
the matrix, as evidenced by the failure surface being
completely coated with alpha-martensite.95


Radiation Damage in Austenitic Steels

57

1200

700

370 ЊC

AISI 304

500

Ultimate

300
Yield
100
0.1

E,
MeV

odf,
dpa s–1

Ti,
ЊC

0.75

0.53
0.29
0.29
0.19
0.17

7.9 ´ 10-7
3.9 ´ 10-7
1.8 ´ 10-7
1.5 ´ 10-7
0.8 ´ 10-7
0.6 ´ 10-7

392
376
373
426
371
371

1
10
Exposure (dpa)

(a)

Strength (MPa)

Strength (MPa)

1000

100

800

600
Yield strength
400

200

0

Strength (MPa)

700

AISI 316

300
Yield

100
0.1
(b)

1

E,

MeV

odf,
dpa s–1

0.76
0.63
0.38
0.35
0.21
0.17

8.4 ´ 10-7
5.1 ´ 10-7
2.3 ´ 10-7
1.9 ´ 10-7
1.0 ´ 10-7
0.6 ´ 10-7

10
Exposure (dpa)

0

6
2
4
8
Neutron fluence (E > 0.1 MeV, n cm–2)

10 ´ 1022

Figure 30 Convergence of ultimate and yield strengths of
annealed 304 stainless steel irradiated in EBR-II and tested
at 370  C. Reproduced from Holmes, J. J.; Straalsund, J. L.
In Proceedings of International Conference: Radiation
Effects in Breeder Reactor Structural Materials; 1977;
pp 53–63.

Ultimate

500

Ultimate strength

Ti,
ºC
399
378
374
424
372
371

100

Figure 29 Strength changes observed in annealed 304
and 316 stainless steels irradiated in EBR-II at 371–426  C
and tested at 385  C. Reproduced from Brager, H. R.
Blackburn, L. D.; Greenslade, D. L. J. Nucl. Mater. 1984,

122–123, 332–337. Microscopy showed that the
dependence of microstructure on displacement rate was
consistent with the macroscopic behavior exhibited by each
alloy. In AISI 316, the flux dependence of precipitation
canceled the opposite dependence of other microstructural
components.

The abrupt jump in strength just before failure
observed by Hamilton and coworkers is the result
of a stress-induced blossoming of a high density
of small, thin, epsilon-martensite platelets, as seen
in Figure 37. These platelets are essentially stacking
faults that form under stress as a result of the influence of both falling nickel level and low deformation temperature to decrease the stacking fault
energy of the matrix.1 When encountered by the
advancing crack tip, the epsilon-martensite is converted to alpha-martensite in the strain field ahead of
the crack, providing a very brittle path for further
cracking.
The correlation between void swelling and both
quasi-embrittlement and true embrittlement is
observed not only in slow tensile tests (Figures 36,
38, and 39) but also in Charpy impact tests as shown

in Figure 39. Figures 40–44 present examples of
swelling-induced failures in components experiencing a wide range of physical insults. The example
of Porollo et al. in Figure 44 (top) is particularly
noteworthy in that it results from significant swelling
at $335  C, a temperature earlier thought not to
produce significant amounts of swelling.
If there are no physical insults experienced by
the component during irradiation, the continued

segregation of nickel to void surfaces and the concurrent rejection of chromium can lead to strong
changes in composition in the matrix during irradiation, pushing the matrix toward ferrite rather than
martensite at higher temperatures, especially for steels
with nickel content of <10%. In some observations
voids encased in austenite shells have been observed
to exist in a pure ferrite matrix.98,99 To date, however,
no significant component failure has been reported
to result from this particular late-term instability.
Finally, there appears to be another late-term
phase instability developing at lower irradiation temperatures that involves martensite but does not
appear to be due to void swelling. Gusev et al. have
shown that for irradiation temperatures below
$350  C a growing tendency for stress-induced martensite formation is occurring in Russian austenitic
steels at doses in the range of 25–55 dpa when tested
at room temperature.100–102 Surprisingly, this instability results in a restoration of engineering ductility
to preirradiation levels. However, the ductility is