Content uploaded by Jiaqi Duan
Author content
All content in this area was uploaded by Jiaqi Duan on Sep 04, 2024
Content may be subject to copyright.
METALS & CORROSION
Effect of grain size on the irradiation response of grade
91 steel subjected to Fe ion irradiation at 300
°
C
Jiaqi Duan
1,2
, Haiming Wen
1,3,
*,LiHe
4
, Kumar Sridharan
4
, Andrew Hoffman
3
,
Maalavan Arivu
1
, Xiaoqing He
5,6
, Rinat Islamgaliev
7
, and Ruslan Valiev
7
1
Department of Materials Science and Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USA
2
Warwick Manufacturing Group, The University of Warwick, Coventry CV4 7AL, UK
3
Department of Nuclear Engineering and Radiation Science, Missouri University of Science and Technology, Rolla, MO 65409, USA
4
Departments of Engineering Physics and Materials Science & Engineering, University of Wisconsin, Madison, WI 53706, USA
5
Electron Microscopy Core Facilities, University of Missouri, Columbia, MO 65211, USA
6
Department of Mechanical and Aerospace Engineering, University of Missouri, Columbia, MO 65211, USA
7
Institute of Physics of Advanced Materials, Ufa State Aviation Technical University, Ufa 450008, Russia
Received: 2 May 2022
Accepted: 22 June 2022
Published online:
14 July 2022
ÓThe Author(s), under
exclusive licence to Springer
Science+Business Media, LLC,
part of Springer Nature 2022
ABSTRACT
Irradiation using Fe ion at 300 °C up to 100 dpa was carried out on three variants of
Grade 91 (G91) steel samples with different grain size ranges: fine-grained (FG,
with blocky grains of a few micrometers long and a few hundred nanometers
wide), ultrafine-grained (UFG, grain size of *400 nm) and nanocrystalline (NC,
lath grains of *200 nm long and *80 nm wide). Electron microscopy investi-
gations indicate that NC G91 exhibit higher resistance to irradiation-induced
defect formation than FG and UFG G91. In addition, nano-indentation studies
reveal that irradiation-induced hardening is significantly lower in NC G91 than
that in FG and UFG G91. Effective mitigation of irradiation damage was achieved
in NC G91 steel in the current irradiation condition.
Handling Editor: Naiqin Zhao.
Address correspondence to E-mail: wenha@mst.edu
https://doi.org/10.1007/s10853-022-07480-6
J Mater Sci (2022) 57:13767–13778
Metals & corrosion
GRAPHICAL ABSTRACT
surface
end
surface
end
surface
end
Fine-grained Ultrafine-grained Nanocrystalline
(a) (b) (c)
Introduction
Ferritic–martensitic (FM) steels such as Grade 91
(G91) are excellent candidates for fuel cladding and
structural materials for advanced fast reactors due to
their good thermo-mechanical properties at
400–550 °C and reduced void swelling compared to
austenitic stainless steels [1–5]. G91 steel has also
been suggested as potential structural material for
light water reactor (LWR) core internals. Structural
materials and fuel cladding in advanced fast reactors
will be subject to high irradiation doses. Life exten-
sion of LWRs and development of advanced fast
reactors requires steels with enhanced irradiation
tolerance and higher mechanical strength [4,6].
However, FM steels suffer profusely from pro-
nounced hardening and embrittlement when they are
irradiated at low temperatures [1,7]. Konstantinovic
´
MJc, et al. [7] reported the mechanical properties of
T91 subjected to 0.8–3.9 displacements per atom
(dpa) at 250–350 °C, and they found significant irra-
diation hardening and decrease of elongation in the
materials. At higher temperatures, softening was
found in a T91 steel at doses above 0.5 dpa, due to the
fast coalescence of dislocation loops [8]. To achieve
enhanced irradiation tolerance and higher strength,
currently used/considered steels may be processed
by advanced manufacturing techniques to improve
their performance through microstructural engineer-
ing at relatively low cost. Equal-channel angular
pressing (ECAP) and high-pressure torsion (HPT)
produce UFG and NC, respectively, metals and alloys
through application of severe plastic deformation.
The applications of ECAP and HPT to G91 steel for
grain refinement have been reported [9–11]. Nanos-
tructured (UFG and NC) steels possess dramatically
higher strength than their conventional FG (grain
diameter [1lm) counterparts, owing to significant
grain boundary (GB) strengthening. Meanwhile, GBs
can also significantly enhance irradiation tolerance of
13768 J Mater Sci (2022) 57:13767–13778
the materials by serving as sinks or recombination
centers for radiation-induced defects [12–14].
In-situ ion irradiation experiments provided the
direct evidence of GBs absorbing individual loops
and dislocation segments in nanocrystalline Ni when
the material was irradiated by Kr ions of 1 MeV at
room temperature [14]. In addition, molecular
dynamic (MD) simulations were used to determine
the mechanism of the interactions between the
defects and GBs at the atomic level. The MD simu-
lations of body-centered cubic Fe by Chen, et al. [15]
suggested that irradiation-induced defects can either
migrate via the bulk chain of defects from matrix to
GBs or via GB chain of defects until they are anni-
hilated. F/M T91 steel processed by ECAP was irra-
diated by Fe ions at 450 °Cupto*150 dpa, and the
UFG sample showed a lower defect density and
smaller swelling rate than the coarse-grained coun-
terpart [16].
Although the defect-GB interactions have been
well studied, only very limited investigations exist on
irradiation of nanostructured steels, and their per-
formance under irradiation at relevant reactor oper-
ating temperatures remains unclear. Notably,
previous reports on irradiation study of NC steels
have been rare. This work studied the irradiation
performance of G91 steel with different grain size
ranges down to the nanometer regime and enhanced
our understanding of irradiation effects in these
materials. The establishment of irradiation perfor-
mance of nanostructured F/M steels with appealing
properties will impact the life extension of current
reactors and the development of advanced reactors.
Experimental
A hot rolled G91 steel of the composition Fe–8.38Cr–
0.9Mo–0.2 V–0.06Nb–0.17Ni-0.43Mn-0.1C–0.03 N (in
wt.%) was normalized at 1050 °C for one hour and
then quenched in oil, followed by tempering at
800 °C for one hour and subsequent air cooling. After
tempering the G91steel samples were processed by
ECAP and HPT, respectively. The processing details
of ECAP and HPT and the resultant microstructures
were documented in our previous studies [9–11]. The
G91 after tempering consists of blocky grains of a few
micrometers long and a few hundred nanometers
wide. Grains were successfully refined to sub-mi-
crometer and nanometer range after ECAP and HPT,
respectively. Therefore, the as-tempered, as-ECAP,
and as-HPT G91 samples are designated as FG, UFG
and NC G91, respectively.
Samples of 5 9591 mm were cut from as tem-
pered, as-ECAP and as-HPT G91, respectively. Prior
to irradiation, samples were mechanically polished
with 0.02 lm colloidal silica in the final stage. Sam-
ples were irradiated by 3.7 MeV Fe2þions using a
NEC 1.7 MV tandem accelerator with a rastered beam
at the University of Wisconsin Ion Beam Laboratory.
Rastered beam was applied for irradiating a rela-
tively large area over multiple samples. Rastered
beam has been reported to suppress cavity swelling.
For ferritic and ferritic-martensitic steels peak void
swelling temperature of 450–480 °C was reported
under ion irradiations [17–19].
The ion energy was chosen for two reasons: (1).
implantation depth: the flat region of damage profile
(half of the dpa peak depth) is about 0.5 lm in depth.
Studies have indicated that the surface effect on
defects can extend to a depth of 0.2–0.3 lm[20,21].
The damage flat region is outside of the expected
surface effect depth. 2). The accelerating capability:
although the maximum accelerating voltage for Fe2þ
is 3 91.7 = 5.1 MV, 1.23 MV was operated for
achieving a stable ion beam current. An irradiation
temperature of 300 °C was selected considering that
the temperature regime important for LWR operation
is between 280 and 320 °C and that FM as nuclear
material could suffer from irradiation-induced
embrittlement and hardening at low temperatures.
However, at this temperature, swelling is not expec-
ted to be significant. Ion fluence was 9.73 910
16
ion=cm2, and the dose rate was 5:21012 ion/cm
2
/s.
Temperature was monitored continuously through-
out irradiation from two thermocouples attached to
the opposite corners of the sample irradiation stage.
The average temperature was measured as
300 ±5°C. Local dose and implanted Fe ion con-
centration in the samples were calculated using the
Kinchin-Pease model with full cascade in the Stop-
ping and Range of Ions in Matter (SRIM) software
[22–24], Fig. 1. A displacement threshold energy of
40 eV and a density of 8:47 1022 atoms/cm
3
for G91
was used in the calculation. The middle range dose
rate was about 2 103dpa/s. Studies have recom-
mended Kinchin-Pease model for comparison of ion
and neutron irradiation data while full cascade
J Mater Sci (2022) 57:13767–13778 13769
calculation is recommended with multicomponent
target materials [23].
Cross-sectional TEM samples were prepared by
focused ion beam (FIB) using a dual-beam FEI Helios
Nanolab. The voltage of the ion beam for early thin-
ning was 30 keV, and it was reduced to 5 keV as the
foil became thinner. For final thinning at thickness of
100–150 nm, a voltage of 5–2 keV was used, so that
the damage from FIB could be minimized. Three
regions in each sample were examined, as marked in
the damage & implantation-depth profile in Fig. 1,
i.e. low, middle and high displacement per atom
(dpa) regions. The TEM sample thicknesses were
measured by electron energy loss spectroscopy
(EELS) using a Gatan Quantum image filter on a FEI
Tecnai G2 FEG TEM operated with a 300 keV elec-
tron beam. Measurements were carried out in the
regions of interest. The log-ratio method was used to
calculate the thickness of the sample relative to the
inelastic mean free path (IMFP), and the measure-
ment errors were assumed ±20% [25,26]. The TEM
samples are wedge-shaped. The thickness variations
from the low to high dpa for the FG, UFG, and NC
G91 TEM samples are 104–135 nm, 128–159 nm and
140–155 nm, respectively. Scanning transmission
electron microscopy (STEM) observations were per-
formed in annular bright field mode under [110] or
[111] zone to capture the total loop and line disloca-
tion density. The semi-convergence angle was 24.5
mrad. The collection angle was 6–13 mrad. All
gvectors contributed to the images. The defects were
manually measured with ImageJ and only the black
spots [1 nm were considered. Under-focus obser-
vations were carried out in each specimen to analyze
the voids produced by ion irradiation. Carbon con-
tamination was not measured in this study.
Nano hardness was measured with a Hysitron TI-
950 triboindenter and a cube corner probe in the
displacement control mode. Ten indents were carried
out in the un-irradiated (unexposed to ion beam)
regions and in the irradiated regions, respectively, of
each sample at displacements of 100, 150 and 200 nm.
Results
Microstructure
The microstructures of FG, UFG and NC G91 steels
are presented in Fig. 2, with a dash line separating
the irradiated and un-irradiated regions in each
image. The microstructure of the FG G91 consists of
typical tempered martensitic blocky grains. Carbides
are observed along the lath boundaries. For the UFG
G91, the microstructure is characterized by equiaxed
grains with an average size of *400 nm. The NC
G91 exhibits lath grains elongated along the shear
direction. The average width and length of those lath
grains are 80 nm and 200 nm, respectively. In each of
the G91 steel samples, the grains in the irradiated and
un-irradiated regions have comparable sizes, indi-
cating that the current irradiation condition did not
lead to grain coarsening. However, grains in NC G91
(both in the irradiated and un-irradiated regions)
coarsened slightly during the irradiation experiment
due to the thermal effect [11]. FG and UFG G91 show
abundant irradiation-induced features, such as dis-
location loops and dislocation lines, in the irradiated
regions. For the NC G91, it is difficult to observe the
Figure 1 SRIM calculated damage and implanted ion
concentration profile in G91 steel. The depth for evaluation is
highlighted.
surface
end
surface
end
surface
end
(a) (b) (c)
Figure 2 Cross-section STEM micrographs showing the
microstructure of FG (a), UFG (b) and NC (c) G91 steel
samples subjected to Fe ion irradiation up to 100 dpa at 300 °C.
The irradiation affected zone is highlighted in each graph (from
surface to the end line).
13770 J Mater Sci (2022) 57:13767–13778
irradiation-induced defects at the magnification in
Fig. 2c.
Figure 3displays bright-field micrographs from
regions with different dpa in the FG, UFG and NC
G91 after irradiation at 300 °C. Images of the un-ir-
radiated regions are also provided in Fig. 3. The
corresponding dislocation loop size & density and
dislocation line density at these locations in each
sample are given in Fig. 4. For the FG G91, small
dislocation loops (black dots), as well as dislocation
lines, can be observed in the low dpa region near the
irradiation surface. The loop and line density are
4.10 910
22
m3and 1.34 910
14
m2, respectively,
and the average size of the dislocation loop is 4.3 nm.
In the middle dpa region, a higher density of dislo-
cation loops and lines are observed; some large loops
up to 10 nm in size coexist with small loops, indi-
cating that loop nucleation and growth occurred
simultaneously. It is interesting to note that disloca-
tion lines are often decorated by loops. Dislocation
loop density and line density further increased to
8.29 910
22
m3and 1.78 910
14
m2, respectively, in
the high dpa regions. Further increases in dislocation
loop density, dislocation line density and dislocation
loop size are observed in the regions with the highest
ion concentration. Note that arrays of dislocations
appear beyond the damage regions, which could
result from ion implantation. The dislocations were
(a1) (a2) (a3)
FG
Low dpa Middle dpa High dpa
<110> <110> <110>
Un-irradiated
(a4)
(c1) (c2) (c3)
NC
<110> <111> <111>
(b1) (b2) (b3)
U
FG
<110> <110> <110>
(b4)
(c4)
Figure 3 Cross-section bright
field STEM micrographs of
FG (a1-4), UFG (b1-4) and
NC (c1-4) G91 from low dpa,
middle dap, high dpa, and un-
irradiated regions. Scale bar,
20 nm. All g vectors
contributed to the image of the
irradiated regions; the zone
axis is indicated.
(a) (b) (c)
FG UFG NC FG UFG NC FG UFG NC
Low dpa Middle dpa High dpa
0
2
4
6
8
Dislocation loop size (nm)
FG UFG NC FG UFG NC FG UFG NC
Low dpa Middle dpa High dpa
0
2
4
6
8
10
12
Dislocation loop density (×10
22
m
-3
)
FG UFG NC FG UFG NC FG UFG NC FG UFG NC
Low dpa Middle dpa High dpa Un-irradiated
0
1
2
3
Dislocation line density (×10
14
m
-2
)
Figure 4 Measurements of the adislocation loop size, bdislocation loop density and cdislocation line density in the FG, UFG and NC
G91 steel samples from the low, middle and high dpa regions.
J Mater Sci (2022) 57:13767–13778 13771
formed to accommodate the surface stresses induced
by high doses of the implanted Fe [27]. For the UFG
G91, the density of loops and dislocation lines near
the surface is 4.50 910
22
m3and 1:4610
14
m2,
respectively. As the damage level increases, the loop
number density and line density increase. Our mea-
surements indicate comparable increase in irradia-
tion-induced defects in UFG and FG G91 when the
dpa increases. For example, as the dpa increases from
30 (low dpa) to 100 (high dpa), the dislocation loop
density and line density increase by 1.87 and 1.29
times, respectively, in UFG G91; in FG 91, the
increases in loop and line density are 2.02 and 1.33
times, respectively. NC G91 exhibits a smaller dislo-
cation loop density, i.e. 3.93 910
22
m3, and a higher
dislocation line density, i.e. 2.28 10
14
m2in the low
dpa regions near the surface, as compared to the FG
and UFG G91. Note that the UFG G91 appears to
contain a larger number of defects compared to FG
G91 in those scanning TEM (STEM) images; this is
because the TEM lamella of UFG G91 is thicker, as
confirmed by the EELS measurements. The higher
dislocation line density found in the low dpa regions
in NC G91 could be attributed to the higher strain
introduced during HPT processing. The increase in
irradiation damage in NC G91 was lower than those
in FG and UFG G91 as dpa increases. The dislocation
loop and line density only slightly increase to
4.23 910
22
m3and 2.45 10
14
m2, respectively, in
the high dpa regions. In addition, defects appear in
the unirradiated regions of the UFG and NC G91
specimens. We would like to emphasize the chal-
lenges of making FIB specimens, especially for the
HPT NC G91, due to the high residual strain.
Therefore, FIB contamination and damages were
more likely to be introduced to the NC G91 FIB
specimens. Meanwhile, severe plastic deformation
could also introduce defects before irradiation. Thus,
the defects observed in NC G91 are not necessarily
irradiation-induced. However, the NC G91 indeed
shows much less difference in defect density across
the depth than FG and UFG G91.
Void search was performed on the three variants of
irradiated G91 steel samples in under-focus bright
field. Some local voids with an average size of *
2.5 nm were found in larger grains of the FG G91 at a
depth of 0.45–0.75 um, as shown in Fig. 5. It is rea-
sonable that the voids only appear in the middle dpa
regions, as the high number of injected ions in the
high dpa regions and the high density of point defect
sinks near the surface could suppress the void for-
mation [28,29]. In contrast, no voids were identified
in UFG and NC G91.
Nano-indentation
Nano-indentations were carried out to monitor the
changes in mechanical property in the irradiated
regions. Figure 6shows the nano-hardness (average
and standard deviation in GPa) of the three variants
of G91 steel samples irradiated at 300 °C. Nano-
hardness of the un-irradiated counterparts (masked
during irradiation experiment) is also provided in
Fig. 6for comparison. For all the samples, the
apparent hardness decreases with the indent dis-
placement, which is the well-known indentation size
effect (ISE). Results indicate that irradiated FG and
UFG G91 exhibit notably higher hardness when
compared to their un-irradiated counterparts. The
irradiation-induced hardening, i.e. the hardness dif-
ference between the irradiated and un-irradiated
condition, in the FG G91 is 1.7 GPa, 1.6 GPa and 1.4
GPa at 100 nm, 150 nm and 200 nm indent dis-
placements, respectively. UFG G91 shows compara-
ble irradiation hardening to FG G91. In contrast, the
NC G91 exhibits negligible hardening after irradia-
tion, with only 0.2 GPa and 0.4 GPa increases mea-
sured when indent displacements were 150 nm and
200 nm, respectively.
Figure 5 Bright field TEM image (defocus -3 lm) of FG G91,
showing irradiation-induced voids in one grain.
13772 J Mater Sci (2022) 57:13767–13778
Discussion
The irradiation experiments using Fe ion at 300 °C
introduced damage up to 100 dpa to the three G91
steel variants of different grain size ranges. The
comparisons in the irradiation damage among the
three G91 steel variants were made, and the key
findings are as follows:
1. Overall, NC G91 shows lower densities of irradi-
ation-induced dislocation loops and lines when
compared to FG and UFG G91.
2. The irradiation defect density vs. dose relation-
ship is much less strong in NC G91 than in FG
and UFG G91.
3. FG and UFG G91 showed measurable irradiation-
induced hardening, whereas it was negligible in
NC G91. This is attributed to the lower irradiation
defect density in NC G91.
4. Voids were only identified in one grain in FG
G91, whereas no voids were found in UFG and
NC G91.
The results demonstrate that in current irradiation
condition, NC G91 exhibits higher irradiation resis-
tance than FG and UFG G91, which is discussed in
more detail as follows.
Irradiation resistance
During ion irradiation, high-energy ions travel
through the crystal and collide with lattice atoms,
which are displaced as interstitials, leaving vacancies
behind. The interstitials can travel and create more
collisions, causing cascades. The interstitials and
vacancies form various kinds of defect, e.g., disloca-
tion loops, dislocation lines and voids, etc. GBs can
help to mitigate the damage via promoting defect
recombination. GBs can act as neutral sinks that
remove the defects by absorbing them and recom-
bining interstitials-vacancies, and thus the irradiation
damage is mitigated. In-situ irradiation experiments
have provided the evidence for the efficacy of GBs
absorbing the defects in various metals [15,30,31].
The sink strength of the GBs is estimated as
k2
gb ¼60=d2, where dis the diameter of the grains and
kgb is the inverse of the average distance that a mobile
defect can move before being absorbed by a GB. The
effects of GBs on irradiation resistance have been
studied in different alloys subjected to various irra-
diation conditions. In the current case, the k2
gb for NC
(d=142 nm, the average area-equivalent circular
diameter) and FG G91 (d=610 nm) is 3:01015m2
and 1:61014m2respectively, indicating that the
sink strength of NC G91 is 19 times greater than that
of FG 91.
Dislocations were introduced during the severe
plastic deformation processing (ECAP or HPT) of the
G91 steel, which can also serve as defect sinks with a
strength of k2
dis ¼zdqd, where qdis the dislocation
density and zd*2–10 is the number of the atomic
sites on each of the crystallographic planes inter-
sected by the dislocation line [6,16]. Dislocation sink
strength was estimated for FG and NC with zd=6.
The dislocation densities in NC and FG G91 are
measured as 1.96 910
14
m2and 0.867 910
14
m2;
respectively (data from the un-irradiated regions,
Fig. 4c. The calculations indicate dislocation sink
strength is 6.6 910
14
m2higher in NC than in FG
G91, while GB sink strength is 28.4 910
14
m2higher
in NC than in FG G91. It is concluded that the higher
GB density in NC is the dominant reason for higher
Figure 6 Comparison in hardness between the irradiated and un-irradiated regions at various indent displacements: aFG; bUFU; and
cNC G91 irradiated with 3.7 MeV Fe ion at 300 °C.
J Mater Sci (2022) 57:13767–13778 13773
irradiation resistance, although dislocations could
add additional irradiation resistance to the NC G91.
Note that dislocations are biased sinks with slight
preference for interstitials over vacancies; however,
when dislocation density is high enough, they can
also provide many sink sites for vacancies, and thus
the void nucleation and growth rate would also be
suppressed [6].
It is noteworthy that FG and UFG G91 TEM
lamellas exhibit similar density of defects, although
the UFG G91 has a higher GB density. Possible
explanations are provided here. Firstly, the grain size
difference between FG and UFG G91 may not be
enough to cause evident difference in irradiation
resistance in the current irradiation conditions. Note
that during the processing of G91, oil quenching was
carried out to generate martensitic blocks, with the
width of a few hundred micrometers; therefore, the
G91 already exhibits fine structures even without
ECAP. Only marginal benefit of mitigating irradia-
tion damage was reported when Fe with ultra-fine
grain sizes was subjected to 10 keV helium ion irra-
diation at 573 K [32]. Secondly, the characters of the
GBs in the UFG G91 can affect the material’s irradi-
ation resistance. As reported [10], a high fraction of
the GBs in the UFG G91 is low-angle GBs, which
exhibit reduced sink strength than the high-angle
GBs [33,34]. Thus, finer grain sizes do not necessarily
lead to significant improvement in irradiation resis-
tance when a high fraction of the GBs are in low
misorientations. Thirdly, post-irradiation annihila-
tion may also be responsible for the negligible dif-
ference. After irradiation was completed, loops could
still move to GBs or recombine during cooling. EI-
Atwani et al. [30] reported significant less dislocation
loops in post-irradiation examination of a NC tung-
sten subjected to Kr ion irradiation to 2–10 dpa when
compared to in-situ measurements; this was true
even for room temperature irradiation. If indeed a
larger number of defects were present during irra-
diation in FG G91 as compared to UFG G91, more
defects would have the chance of being annihilated
after irradiation stopped. Therefore, the final differ-
ence between FG and UFG G91 in irradiation-in-
duced defects measured during the post-irradiation
examination would be reduced.
The current results demonstrate that NC G91 has a
higher radiation resistance than its UFG and FG
counterparts. However, conclusion here may not
necessarily be extended to extreme irradiation
conditions when the dose is up to hundreds of dpa.
Gigax et al. [34]. reported irradiation instability in
ECAP processed UFG T91 at ultra-high damage
levels, and reduced irradiation resistance of this
material as compared to its coarse-grained counter-
part at such high doses. As the dose increases,
structural changes in GBs occur, i.e., high angle
boundaries transformed to low angle boundaries,
leading to a decline in the effectiveness of reducing
irradiation damage. Also, voids become biased sinks
for vacancies at extreme high dose, and grain growth
is accelerated due to the biased absorption of inter-
stitials by GBs. The irradiation behavior of NC G91 at
ultra-high doses still needs to be investigated in the
future.
Irradiation-induced hardening
Irradiation hardening results from irradiation-in-
duced voids, precipitates, dislocation loops, disloca-
tion lines, etc. For the current G91, voids were
detected in one grain and only in irradiated FG G91;
in addition, the density of the precipitates (M
23
C
6
and
MX, where Mand Xdenote metallic elements and
carbon and/or nitrogen, respectively) [9,10] is com-
parable between the irradiated and un-irradiated
regions. Therefore, contributions from voids and
precipitates to the irradiation hardening are negligi-
ble, which is a reasonable assumption based on a
previous irradiation study on the G91 [35]. The bar-
rier hardening model is employed here to estimate
the hardening, Dry, from loops and dislocation lines:
Dry¼MaLGb ffiffiffiffiffiffiffiffiffiffi
qLdL
pþMaDGbðffiffiffiffiffiffiffi
qirr
D
qffiffiffiffiffiffiffiffiffiffiffi
qunirr
D
qÞð1Þ
where M¼2:7 is the Taylor factor for body-cen-
tered cubic materials, G¼75GPa is the shear modu-
lus of G91 steel at room temperature, b¼0:25nm is
the magnitude of the Burgers vector, aL,qLand dLare
the barrier strength, density and average size of the
dislocation loops, respectively, and aD,qunirr
Dand qirr
D
are the barrier strength of the dislocation lines and
their density before and after irradiation, respec-
tively. aLand aDhere are taken as 0.40 and 0.64,
respectively [35,36]. The qunirr
Dvalues are measured to
be 8.67 910
13
m2, 1.00 910
14
m2and 1.96 910
14
m2for the FG, UFG and NC G91, respectively. Based
on the obtained statistics of the defects, estimation of
the irradiation hardening using the empirical model
was performed. The calculation results are displayed
13774 J Mater Sci (2022) 57:13767–13778
in Table 1, indicating that: a) the FG and UFG overall
exhibit higher hardening than the NC counterpart; b)
as the dpa increases, the hardening increases in FG
and UFG while it remains almost constant in the NC
counterpart.
Nano-indentations were carried out to probe the
mechanical property changes after irradiation in G91
with different microstructures. Generally, indenta-
tion should be no more than 10%–20% of the thick-
ness of the interested region, so that the intrinsic
properties of the interested region can be obtained
and the interference from the underlying substrate
can be avoided. Therefore, 200 nm is chosen as the
reference depth for quantitative analyses of the
hardness measurements. The nanoindentation results
were first divided by 9.8 to convert to Vickers hard-
ness numbers; then they were converted to Vickers
hardness by multiplying a coefficient of 0.76, and
further converted to yield strength by multiplying a
coefficient of 3.06 [37–39]. The converted nano-in-
dentation results are also presented in Table 1for
comparison. The values of the calculated hardening
(e.g. from the low & middle dpa regions) via Eq. (1)
are comparable to the measured hardening values via
nano-indentation for FG and UFG G91. However, the
calculations yield much higher hardening values
than the measurement for NC G91. The discrepancy
may be explained by two reasons. First, artifacts
introduced during TEM sample preparation by FIB
could be counted as irradiation-induced defects. It is
difficult to entirely avoid any contamination and
damage from FIB during sample preparation,
although care was taken to minimize the damage
from FIB. Second, the severe plastic deformation
introduced defects before the irradiation [40,41].
Both could lead to an over-estimation of the irradia-
tion damage in HPT/NC G91.
In addition to GBs, the precipitate/matrix inter-
faces are also known as defect sinks that help to
decrease the density of irradiation-induced disloca-
tion loops [42–44]. G91 contains two types of pre-
cipitates, i.e., M23C6and MX. Although severe plastic
deformation can significantly refine the grain size, it
seems to have less impact on the size and fraction of
those precipitates [10,12]. Therefore, the improved
irradiation resistance observed in NC G91 is not
attributed to the precipitates.
Conclusions
Three G91 steel variants with different grain size
ranges (FG, UFG and NC) were irradiated by Fe ion
up to 100 dpa at 300 °C. The irradiation response of
each variant was accessed. None of the three G91
steel variants exhibited irradiation-induced grain
growth. Irradiation-induced voids were only found
in FG G91. As the dpa rose along the irradiation
depth, the density of the defects (dislocation loops
and lines) increased; however, the increase rate was
notably lower in NC G91 than in FG and UFG G91.
Meanwhile, the NC G91 showed significantly lower
irradiation hardening than FG and UFG G91. It is
concluded that GBs effectively mitigate the irradia-
tion damage when the microstructure is on the NC
scale in G91 steel under the current irradiation
condition.
Acknowledgements
This research was financially supported by the U.S.
Department of Energy, Office of Nuclear Energy
through the NEET-NSUF (Nuclear Energy Enabling
Technology—Nuclear Science User Facility) program
(award number DE-NE0008524), and through the
NSUF-RTE program (award number 18-1403). Partial
support for Haiming Wen and Andrew Hoffman
came from the U.S. Nuclear Regulatory Commission
(NRC) Faculty Development Program (award num-
ber NRC 31310018M0044). Ruslan Valiev gratefully
acknowledges the financial support from Russian
Foundation for Basic Research (Project 20-03-00614).
Table 1 Comparison of
measured hardening (from
nano-indentation) and
calculated hardening (via
Eq. 1) in three G91 steel
variants irradiated by Fe ion at
300 °C
Measured hardening Calculated hardening (MPa)
(MPa) Low dpa Middle dpa High dpa
FG 329 ±70 340 ±132 453 ±181 518 ±233
UFG 352 ±59 345 ±138 431 ±182 484 ±208
NC 94 ±35 283 ±107 282 ±119 302 ±119
J Mater Sci (2022) 57:13767–13778 13775
Declarations
Conflict of interest The authors declare that they
have no known competing financial interests or per-
sonal relationships that could have appeared to
influence the work reported in this paper.
References
[1] Henry J, Maloy SA (2017) 9 - Irradiation-resistant ferritic
and martensitic steels as core materials for Generation IV
nuclear reactors. In: Yvon P (ed) Structural Materials for
Generation IV Nuclear Reactors. Woodhead Publishing,
pp 329–355
[2] Bhattacharya A, Zinkle SJ (2020) 1.12 - Cavity Swelling in
Irradiated Materials. In: Konings RJM, Stoller RE (eds)
Comprehensive Nuclear Materials (Second Edition). Else-
vier, Oxford, pp 406–455
[3] Gaganidze E, Aktaa J (2013) Assessment of neutron irradi-
ation effects on RAFM steels. Fusion Eng Des 88:118–128.
https://doi.org/10.1016/j.fusengdes.2012.11.020
[4] Zinkle SJ, Was GS (2013) Materials challenges in nuclear
energy. Acta Mater 61:735–758. https://doi.org/10.1016/j.ac
tamat.2012.11.004
[5] Klueh RL, Harries DR (2001) High-chromium ferritic and
martensitic steels for nuclear applications, in, AsTM West
Conshohocken, PA,.
[6] Was GS (2016) Fundamentals of radiation materials science:
metals and alloys, springer.
[7] Konstantinovic
´MJ, Stergar E, Lambrecht M, Gavrilov S
(2016) Comparison of the mechanical properties of T91 steel
from the MEGAPIE, and TWIN-ASTIR irradiation pro-
grams. J Nucl Mater 468:228–231. https://doi.org/10.1016/j.
jnucmat.2015.07.038
[8] Yan H, Liu X, He L, Stubbins J (2021) Early-stage
microstructural evolution and phase stability in neutron-ir-
radiated ferritic-martensitic steel T91. J Nucl Mater
557:153207. https://doi.org/10.1016/j.jnucmat.2021.153207
[9] Duan J, Wen H, Zhou C, He X, Islamgaliev R, Valiev R
(2020) Annealing behavior in a high-pressure torsion-pro-
cessed Fe–9Cr steel. J Mater Sci 55:7958–7968. https://doi.
org/10.1007/s10853-020-04560-3
[10] Duan J, Wen H, Zhou C, He X, Islamgaliev R, Valiev R
(2019) Discontinuous grain growth in an equal-channel
angular pressing processed Fe-9Cr steel with a heteroge-
neous microstructure. Mater Characterization. https://doi.or
g/10.1016/j.matchar.2019.110004
[11] Duan J, Wen H, Zhou C, Islamgaliev R, Li X (2019) Evo-
lution of microstructure and texture during annealing in a
high-pressure torsion processed Fe-9Cr alloy. Materialia
6:100349. https://doi.org/10.1016/j.mtla.2019.100349
[12] Duan J, He L, Fu Z, Hoffman A, Sridharan K, Wen H (2021)
Microstructure, strength and irradiation response of an ultra-
fine grained FeNiCoCr multi-principal element alloy. J Alloy
Compd 851:156796. https://doi.org/10.1016/j.jallcom.2020.
156796
[13] Zhang X, Hattar K, Chen Y, Shao L, Li J, Sun C, Yu K, Li N,
Taheri ML, Wang H, Wang J, Nastasi M (2018) Radiation
damage in nanostructured materials. Prog Mater Sci
96:217–321. https://doi.org/10.1016/j.pmatsci.2018.03.002
[14] Sun C, Song M, Yu KY, Chen Y, Kirk M, Li M, Wang H,
Zhang X (2013) In situ evidence of defect cluster absorption
by grain boundaries in Kr ion irradiated nanocrystalline Ni.
Metall and Mater Trans A 44:1966–1974. https://doi.org/10.
1007/s11661-013-1635-9
[15] Chen D, Wang J, Chen T, Shao L (2013) Defect annihilation
at grain boundaries in alpha-Fe. Sci Rep 3:1450. https://doi.
org/10.1038/srep01450
[16] Song M, Wu YD, Chen D, Wang XM, Sun C, Yu KY, Chen
Y, Shao L, Yang Y, Hartwig KT, Zhang X (2014) Response
of equal channel angular extrusion processed ultrafine-
grained T91 steel subjected to high temperature heavy ion
irradiation. Acta Mater 74:285–295. https://doi.org/10.1016/
j.actamat.2014.04.034
[17] Toloczko MB, Garner F, Voyevodin V, Bryk V, Borodin O,
Mel’Nychenko V, Kalchenko A (2014) Ion-induced swelling
of ODS ferritic alloy MA957 tubing to 500 dpa. J Nucl
Mater 453:323–333. https://doi.org/10.1016/j.jnucmat.2014.
06.011
[18] Gigax J, Chen T, Kim H, Wang J, Price L, Aydogan E,
Maloy SA, Schreiber D, Toloczko M, Garner F (2016)
Radiation response of alloy T91 at damage levels up to 1000
peak dpa. J Nucl Mater 482:257–265. https://doi.org/10.10
16/j.jnucmat.2016.10.003
[19] Aydogan E, Chen T, Gigax J, Chen D, Wang X, Dzhumaev
P, Emelyanova O, Ganchenkova M, Kalin B, Leontiva-
Smirnova M (2017) Effect of self-ion irradiation on the
microstructural changes of alloy EK-181 in annealed and
severely deformed conditions. J Nucl Mater 487:96–104.
https://doi.org/10.1016/j.jnucmat.2017.02.006
[20] Was G, Jiao Z, Getto E, Sun K, Monterrosa A, Maloy S,
Anderoglu O, Sencer B, Hackett M (2014) Emulation of
reactor irradiation damage using ion beams. Scripta Mater
88:33–36. https://doi.org/10.1016/j.scriptamat.2014.06.003
[21] Stoller RE (2002) The effect of free surfaces on cascade
damage production in iron. J Nucl Mater 307:935–940.
https://doi.org/10.1016/S0022-3115(02)01096-6
[22] Weber WJ, Zhang Y (2019) Predicting damage production in
monoatomic and multi-elemental targets using stopping and
13776 J Mater Sci (2022) 57:13767–13778
range of ions in matter code: Challenges and recommenda-
tions. Curr Opin Solid State Mater Sci 23:100757. https://d
oi.org/10.1016/j.cossms.2019.06.001
[23] Stoller RE, Toloczko MB, Was GS, Certain AG, Dwaraknath
S, Garner FA (2013) On the use of SRIM for computing
radiation damage exposure. Nucl Instrum Methods Phys
Res, Sect B 310:75–80. https://doi.org/10.1016/j.nimb.2013.
05.008
[24] Ziegler JF, Ziegler MD, Biersack JP (2010) SRIM – The
stopping and range of ions in matter (2010). Nucl Instrum
Methods Phys Res, Sect B 268:1818–1823. https://doi.org/
10.1016/j.nimb.2010.02.091
[25] Lin Y-R, Bhattacharya A, Chen D, Kai J-J, Henry J, Zinkle
SJ (2021) Temperature-dependent cavity swelling in dual-
ion irradiated Fe and Fe-Cr ferritic alloys. Acta Mater
207:116660. https://doi.org/10.1016/j.actamat.2021.116660
[26] Egerton RF (2011) Electron energy-loss spectroscopy in the
electron microscope, Springer Science & Business Media.
[27] Didenko AN, Kozlov EV, Sharkeev YP, Tailashev AS,
Rjabchikov AI, Pranjavichus L, Augulis L (1993) Observa-
tion of deep dislocation structures and ‘‘long-range effect’’ in
ion-implanted a-Fe. Surf Coat Technol 56:97–104. https://d
oi.org/10.1016/0257-8972(93)90012-D
[28] Shao L, Wei CC, Gigax J, Aitkaliyeva A, Chen D, Sencer
BH, Garner FA (2014) Effect of defect imbalance on void
swelling distributions produced in pure iron irradiated with
3.5MeV self-ions. J Nucl Mater 453:176–181. https://doi.
org/10.1016/j.jnucmat.2014.06.002
[29] Sun C, Garner FA, Shao L, Zhang X, Maloy SA (2017)
Influence of injected interstitials on the void swelling in two
structural variants of 304L stainless steel induced by self-ion
irradiation at 500°C. Nucl Instrum Methods Phys Res, Sect
B 409:323–327. https://doi.org/10.1016/j.nimb.2017.03.070
[30] El-Atwani O, Esquivel E, Aydogan E, Martinez E, Baldwin
JK, Li M, Uberuaga BP, Maloy SA (2019) Unprecedented
irradiation resistance of nanocrystalline tungsten with
equiaxed nanocrystalline grains to dislocation loop accu-
mulation. Acta Mater 165:118–128. https://doi.org/10.1016/
j.actamat.2018.11.024
[31] El-Atwani O, Esquivel E, Efe M, Aydogan E, Wang YQ,
Martinez E, Maloy SA (2018) Loop and void damage during
heavy ion irradiation on nanocrystalline and coarse grained
tungsten: Microstructure, effect of dpa rate, temperature, and
grain size. Acta Mater 149:206–219. https://doi.org/10.1016/
j.actamat.2018.02.035
[32] El-Atwani O, Nathaniel JE, Leff AC, Hattar K, Taheri ML
(2017) Direct observation of sink-dependent defect evolution
in nanocrystalline iron under irradiation. Sci Rep 7:1836.
https://doi.org/10.1038/s41598-017-01744-x
[33] Han WZ, Demkowicz MJ, Fu EG, Wang YQ, Misra A
(2012) Effect of grain boundary character on sink efficiency.
Acta Mater 60:6341–6351. https://doi.org/10.1016/j.actamat.
2012.08.009
[34] Gigax JG, Kim H, Chen T, Garner FA, Shao L (2017)
Radiation instability of equal channel angular extruded T91
at ultra-high damage levels. Acta Mater 132:395–404. http
s://doi.org/10.1016/j.actamat.2017.04.038
[35] Liu X, Miao Y, Li M, Kirk MA, Maloy SA, Stubbins JF
(2017) Ion-irradiation-induced microstructural modifications
in ferritic/martensitic steel T91. J Nucl Mater 490:305–316.
https://doi.org/10.1016/j.jnucmat.2017.04.047
[36] Bergner F, Pareige C, Herna´ndez-Mayoral M, Malerba L,
Heintze C (2014) Application of a three-feature dispersed-
barrier hardening model to neutron-irradiated Fe–Cr model
alloys. J Nucl Mater 448:96–102. https://doi.org/10.1016/j.
jnucmat.2014.01.024
[37] Yabuuchi K, Kuribayashi Y, Nogami S, Kasada R, Hasegawa
A (2014) Evaluation of irradiation hardening of proton
irradiated stainless steels by nanoindentation. J Nucl Mater
446:142–147. https://doi.org/10.1016/j.jnucmat.2013.12.009
[38] Chen D, Murakami K, Dohi K, Nishida K, Li Z, Sekimura N
(2020) The effects of loop size on the unfaulting of Frank
loops in heavy ion irradiation. J Nucl Mater. https://doi.org/
10.1016/j.jnucmat.2019.151942
[39] Busby JT, Hash MC, Was GS (2005) The relationship
between hardness and yield stress in irradiated austenitic and
ferritic steels. J Nucl Mater 336:267–278. https://doi.org/10.
1016/j.jnucmat.2004.09.024
[40] Setman D, Kerber MB, Schafler E, Zehetbauer MJ (2010)
Activation enthalpies of deformation-induced lattice defects
in severe plastic deformation nanometals measured by dif-
ferential scanning calorimetry. Metall and Mater Trans A
41:810–815. https://doi.org/10.1007/s11661-009-0058-0
[41] Setman D, Schafler E, Korznikova E, Zehetbauer MJ (2008)
The presence and nature of vacancy type defects in
nanometals detained by severe plastic deformation. Mater
Sci Eng, A 493:116–122. https://doi.org/10.1016/j.msea.20
07.06.093
[42] Zhao MZ, Liu PP, Bai JW, Zhu YM, Wan FR, Ohnuki S,
Zhan Q (2014) In-situ observation of the effect of the pre-
cipitate/matrix interface on the evolution of dislocation
structures in CLAM steel during irradiation. Fusion Eng Des
89:2759–2765. https://doi.org/10.1016/j.fusengdes.2014.07.
022
[43] Getto E, Vancoevering G, Was GS (2017) The co-evolution
of microstructure features in self-ion irradiated HT9 at very
high damage levels. J Nucl Mater 484:193–208. https://doi.
org/10.1016/j.jnucmat.2016.12.006
J Mater Sci (2022) 57:13767–13778 13777
[44] Liu PP, Zhao MZ, Zhu YM, Bai JW, Wan FR, Zhan Q
(2013) Effects of carbide precipitate on the mechanical
properties and irradiation behavior of the low activation
martensitic steel. J Alloy Compd 579:599–605. https://doi.
org/10.1016/j.jallcom.2013.07.085
Publisher’s Note Springer Nature remains neutral with
regard to jurisdictional claims in published maps and
institutional affiliations.
13778 J Mater Sci (2022) 57:13767–13778
A preview of this full-text is provided by Springer Nature.
Content available from Journal of Materials Science
This content is subject to copyright. Terms and conditions apply.
