Scripta Materialia 162 (2019) 94–98

Contents lists available at ScienceDirect

Scripta Materialia

journal homepage: www.elsevier.com/locate/scriptamat

Regular article

Post irradiation examination of nanoprecipitate stability and α′

precipitation in an oxide dispersion strengthened Fe-12Cr-5Al alloy☆
Caleb P. Massey a,⁎, Philip D. Edmondson b, Kevin G. Field b, David T. Hoelzer b, Sebastien N. Dryepondt b,
Kurt A. Terrani b, Steven J. Zinkle a,b
a
Department of Nuclear Engineering, University of Tennessee, Knoxville, TN 37996, United States of America
b
Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States of America

a r t i c l e i n f o a b s t r a c t

Article history: Oxide dispersion strengthened (ODS) FeCrAl alloy (Fe-12Cr-5Al wt%) was neutron irradiated to 1.8
Received 21 July 2018 displacements per atom (dpa) at 215, 357, and 557 °C to investigate (Y,Al,O) nano-oxide stability and Cr-rich
Received in revised form 28 October 2018 α′ precipitate formation using atom probe tomography. The nano-oxide sizes remain unchanged after irradiation,
Accepted 28 October 2018
but the compositions are dependent on irradiation temperature due to competition between ballistic dissolution
Available online 7 November 2018
and precipitate reformation. Cr-rich clusters observed after 357 °C irradiation have lower Cr contents (61 at.%)
Keywords:
than equilibrium α′, with a lower number density than wrought FeCrAl alloys. This suggests a suppression of
Atom probe tomography α′ precipitation due to the high sink strength of ODS FeCrAl.
Oxide dispersion strengthened (ODS) alloy © 2018 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Precipitation
Neutron irradiation

The realization of advanced fission and fusion reactor concepts will separation kinetics [15,16]. In addition, compositional design regions
require the development of materials specifically designed to operate for FeCrAl alloys have been established to mitigate this phase separation
under high temperatures (300–1000 °C) and irradiation doses (N50 phenomenon after thermal ageing and irradiation [17,18]. However,
displacements per atom, or dpa) [1]. Furthermore, the recent desire few studies have examined the α′-precipitation in ODS variants of
for materials capable of reducing the impact of nuclear accident scenar- FeCr and FeCrAl alloys. A study on a Fe-14Cr ODS alloy MA957 neutron
ios has accelerated research into materials that can retain their high irradiated to 109 dpa at 412 °C indicated a high number density (9
temperature strength and oxidation resistance [2]. Oxide dispersion × 1023 m−3) of α′ precipitates within the matrix [19]. Furthermore,
strengthened (ODS) FeCrAl alloys are currently under investigation as thermal ageing experiments on the legacy ODS (Fe-20Cr-6Al) alloy
candidate materials for both fission and fusion reactor applications PM2000 developed by Plansee have indicated α′-α phase separation
[3,4]. These alloys attempt to combine the high temperature oxidation and the subsequent embrittlement of the alloy [20–22]. The α′-
resistance of FeCrAl [4,5] with the high strength and irradiation resis- induced embrittlement of these higher Cr (N14 wt%) ODS FeCr and
tance of conventional ODS FeCr alloys [1,6,7]. FeCrAl alloys has led to the development of low-Cr (≤12 wt%) ODS
One challenge in the development of high-Cr ferritic alloys is that at FeCrAl alloys to decrease α′-α phase separation within the material [4].
intermediate temperatures applicable to both fission and fusion reactor This work investigates the Cr-rich α′ precipitation in a low-Cr (12 wt
applications (300–450 °C), Fe-Cr phase separation kinetics favor the %) ODS FeCrAl alloy after neutron irradiation using atom probe tomog-
precipitation of deleterious Cr-rich α′ precipitates within the α-Fe ma- raphy, with a specific interest in comparisons with model FeCrAl alloys
trix. The precipitation of α′-Cr has been studied for both FeCr [8–11] of similar composition without oxide dispersions. The dependence of α′
and FeCrAl [12–14] alloys, with irradiation accelerating α′-α phase precipitate stability on irradiation temperature is also discussed. De-
tailed characterizations of the fine-scale (Y,Al,O) rich precipitate com-
positions, sizes, and number densities are reported to understand
☆ This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-
their post-irradiation stability as a function of irradiation temperature.
00OR22725 with the U.S. Department of Energy. The United States Government retains
and the publisher, by accepting the article for publication, acknowledges that the United The ODS FeCrAl alloy investigated in this work, designated 125YF,
States Government retains a non-exclusive, paid-up, irrevocable, worldwide license to was produced via the powder metallurgy approach by mechanically
publish or reproduce the published form of this manuscript, or allow others to do so, for alloying Fe-12Cr-5.6Al (wt%) gas atomized powder with nanocrystal-
United States Government purposes. The Department of Energy will provide public access line Y2O3 followed by degassing at 300 °C for 24 h and subsequent ex-
to these results of federally sponsored research in accordance with the DOE Public Access
Plan (http://energy.gov/downloads/doe-public-access-plan).
trusion at 950 °C. Details concerning the fabrication, microstructure,
⁎ Corresponding author. tensile properties, and oxidation resistance of the as-extruded 125YF
E-mail address: [email protected] (C.P. Massey). alloy are reported elsewhere [3]. Miniature SS-J3 sheet tensile

https://doi.org/10.1016/j.scriptamat.2018.10.047
1359-6462/© 2018 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
 C.P. Massey et al. / Scripta Materialia 162 (2019) 94–98 95

specimens were loaded into rabbit capsules and irradiated in the High (Y,Al,O)-rich precipitates were homogeneously distributed through-
Flux Isotope Reactor (HFIR) at Oak Ridge National Laboratory (ORNL) out the matrix in all samples examined. The average (Y,Al,O) precipitate
to 1.8 dpa at ~215, 357, and 557 °C with corresponding dose rates of radii were calculated using the atomic count method assuming spheri-
9.3–9.8 × 10−7 dpa/s [23]. The lower temperature irradiations targeted cal equivalent radii [9]. For all conditions, radii were highly consistent
light water reactor operating regimes while the higher temperature ir- and averaged 1.5 ± 0.6 nm (Table 2). The measured number densities
radiation focused on advanced fission/fusion reactor temperature were also high (4.6–7.4 × 1023 m−3) and comparable to 125Y series
regimes. ODS FeCrAl alloys previously developed at ORNL (1.4–2.5 × 1023 m−3)
Following irradiation, tensile testing was performed on the unirradi- [32]. This result indicates that the precipitate sizes and number densities
ated and irradiated 125YF specimens. Details surrounding the mechan- are stable at all irradiation temperatures 215–557 °C for the low dose
ical property evaluations of this alloy and subsequent microstructural (1.8 dpa) irradiation condition. In addition, the size distributions re-
investigations will be reported in a separate study. One half of the failed ported in this work have a dimensional scale comparable to those re-
tensile specimen for each condition was mounted and polished. Using a ported in Zr- and Hf-containing ODS FeCrAl alloys [33,34], which
focused ion beam (FIB) lift-out technique [24], atom probe tomography suggests that it is possible to refine the nanoprecipitate size for en-
(APT) specimens were prepared from the tensile specimen head using hanced sink strength using optimum processing conditions alone and
an FEI Quanta 3D dual-beam FIB at ORNL's Low Activation Materials De- without the need for significant additional alloying additions beyond
velopment and Analysis (LAMDA) Laboratory [25]. APT data collection the baseline ODS FeCrAl composition.
was performed at the Center for Advanced Energy Studies-Microscopy A previous APT study on neutron irradiated MA957 (109 dpa, 412
and Characterization Suite (CAES-MaCS) at Idaho National Laboratory °C) reported that α′ precipitates were anti-correlated with the (Y,Ti,O)
(INL) using a Cameca model 4000X HR Local Electrode Atom Probe nanoprecipitates within the ODS FeCr alloy and preferred to nucleate
(LEAP) with a 36% detection efficiency. The LEAP was operated in laser between precipitates [19]. Thus, to prevent trajectory aberrations from
mode with a pulse frequency of 200 kHz, a laser energy of 32 pJ, and a providing false-positive results for Cr-rich clustering behavior around
target detection rate of 0.005 atoms/pulse at 50 K. At least 3 APT speci- the (Y,Al,O)-rich precipitates, these precipitates were subtracted from
mens were acquired for each condition with at least 8 M ions collected the matrix prior to applying the cluster search algorithm on the Cr-
per specimen. Cameca's Integrated Visualization and Analysis Software rich clusters. Cr-rich clusters were only found in OD03 specimen. The
(IVAS) version 3.6.8 was used for both data reconstruction and analysis lack of observed α′ precipitates in the low temperature OD01 and high
[26]. The maximum separation method was implemented to identify (Y, temperature OD06 specimens is explained through a consideration of
Al,O)- and Cr-rich clusters in each control volume [27–29]. Since trajec- the phase diagram and the irradiation conditions. The higher tempera-
tory aberrations may create local Cr-enrichments in the vicinity of (Y,Al, ture (557 °C) of the OD06 specimen is well above the miscibility gap
O)-rich precipitates, the maximum separation was first applied to the for α′-α phase separation based on the Fe-Cr phase diagram, so it
(Y,Al,O)-rich precipitates. Following the identification of the (Y,Al,O)- would be expected that no α′ precipitates were noted for any OD06
rich nanoprecipitates, the maximum separation method was then ap- specimens. Based on the equilibrium phase diagram, α′ precipitates
plied to the APT specimen in the remaining control volume with (Y,Al, should form at the lower OD01 (215 °C) temperature. The lack of pre-
O)-rich precipitates omitted to identify Cr-rich precipitates. cipitation could be explained by a competition between the radiation
Table 1 compares the composition of as-fabricated control sample enhanced diffusion of Cr atoms and the ballistic dissolution of these pre-
reported elsewhere [3] with the bulk composition of the unirradiated cipitates at this temperature. It is also possible that α′-α phase separa-
and irradiated specimens calculated using the native peak decomposi- tion may still occur at 215 °C with higher doses (longer ageing times),
tion algorithm within the IVAS program (reported in at.%). Peak but future studies are needed to confirm this possibility.
deconvolution is important for accurate compositional corrections for Presented in Fig. 1 is a Cr-rich α′ precipitate (illustrated by the em-
both bulk and precipitate compositions due to the mass-to-charge bedded 2D contour plot) located near the {110} pole used for the sample
ratio peak overlap at 27 Da (27Al+1, 54Cr+2, 54Fe+2). The Y content in reconstruction and nearby (Y,Al,O) precipitates. The Cr concentration
all samples is underestimated in comparison to previous chemical anal- shown is for a cubic region of interest (ROI) slightly larger than the pre-
ysis results [3], which can partially be explained through the preferen- cipitate, which includes a substantial contribution from the matrix that
tial loss in Y content in previous APT studies [30]. However what is shifts the visualized concentration to values lower than the actual clus-
more interesting is the drastic underestimation of O content, which ter composition. The high number density of α′ clusters are illustrated in
has important consequences in compositional accuracy of the (Y,Al, magenta overlaid with the (Y,Al,O) nanoprecipitates in Fig. 2(a). A
O)-rich precipitates. Standard peak decomposition algorithms in IVAS closer look at one of these selected clusters shows a qualitative enrich-
assume natural isotopic abundances for each element. Although some ment in Cr as illustrated in the Cr-atom map (Fig. 2(b)). A significant in-
transmutation reactions may have occurred over the course of the neu- crease in the total atomic point density was observed across the
tron irradiation experiment, it has previously been shown that for precipitates (Fig. 2(c)) which suggests significant trajectory aberrations
model FeCrAl alloys irradiated up to 7 dpa in HFIR, the assumption of from the matrix overlapping with the identified precipitates [35,36]. To
natural isotopic abundance results in a negligible compositional error correct for matrix contributions to the cluster compositions, the percent
[13]. Since cluster compositions outputted by the maximum separation increase in point density across the clusters was used to subtract an
method are not corrected for peak overlaps, individual cluster composi- equivalent fraction of atoms of the matrix composition from the precip-
tions for both Cr- and (Y,Al,O)-rich precipitates were corrected using itates, resulting in an approximated aberration-corrected cluster com-
the methodology previously developed by Edmondson et al. for α′- position. Corrections for increased point density variations have
precipitates in neutron irradiated FeCrAl alloys [31]. previously been applied in a similar manner by Briggs for oxide-free

Table 1
Comparison of 125YF atom probe tomography specimen bulk compositions (in at.%) as a function of irradiation dose and temperature with previous chemical analysis results [3] using
standard peak decompositions.

125YF specimen Dose [dpa] Tirr Fe Cr Al Y O C N

[3] – – 78.09 11.79 9.21 0.11 0.63 0.09 0.08

Control (OD34) – – 78.00 11.82 9.23 0.06 0.54 0.18 0.01
OD01 1.8 215 °C 78.28 11.68 9.24 0.08 0.40 0.11 0.00
OD03 1.8 357 °C 77.59 11.65 9.05 0.07 0.33 0.08 0.01
OD06 1.8 557 °C 78.49 11.71 9.07 0.09 0.35 0.07 0.00
96 C.P. Massey et al. / Scripta Materialia 162 (2019) 94–98

Table 2
Trajectory aberration corrected pre- and post-irradiation precipitate characteristics for each 125YF specimen. Surrounding matrix compositions are also reported. Errors in parenthesis are
reported as one standard deviation from the mean. Compositions are in at.%.

Sample Type Rp [nm] Np × 1023 [m−3] Fe Cr Al Y O

OD34 Matrix n/a n/a 78.16 (0.18) 11.83 (0.05) 9.17 (0.12) 0.03 (0.01) 0.46 (0.06)
Y,Al,O 1.5 (0.6) 4.56 (0.65) 58.26 (10.37) 10.17 (3.58) 13.59 (3.56) 7.70 (7.57) 9.86 (5.00)
OD01 Matrix n/a n/a 78.42 (0.29) 11.67 (0.22) 9.23 (0.22) 0.03 (0.01) 0.30 (0.02)
Y,Al,O 1.5 (0.6) 4.56 (0.31) 59.40 (9.74) 12.75 (3.90) 8.06 (3.27) 8.97 (7.49) 10.11 (4.80)
OD03 Matrix n/a n/a 78.54 (0.08) 11.74 (0.16) 9.13 (0.21) 0.02 (0.01) 0.30 (0.02)
Y,Al,O 1.5 (0.6) 5.47 (0.36) 55.68 (10.68) 11.29 (3.52) 11.66 (3.69) 9.70 (7.89) 10.98 (5.35)
Cr-rich 1.6 (0.4) 0.78 (0.45) 30.43 (8.35) 60.72 (10.67) 5.37 (2.96) 0.58 (0.69) 2.48 (2.74)
OD06 Matrix n/a n/a 78.75 (0.02) 11.70 (0.01) 8.98 (0.01) 0.02 (0.01) 0.25 (0.01)
Y,Al,O 1.5 (0.6) 7.38 (0.22) 55.63 (12.55) 8.84 (3.32) 13.09 (3.81) 10.96 (9.56) 10.87 (5.99)

FeCrAl under similar irradiation conditions and temperatures [13]. This the margin of error for the same model FeCrAl alloy (1.6 ± 0.8 nm)
aberration-correction methodology was also extended to the popula- [13,31]. However, the α′ precipitate number density reported in this
tion of (Y,Al,O) precipitates in this work. work for the ODS FeCrAl alloy (0.78 × 1023 m−3) is an order of magni-
Comparing the measured α′ precipitates with model FeCrAl alloys of tude lower than for the compositionally similar non-ODS model FeCrAl
similar bulk composition shows significant similarities and differences, alloy (7 × 1023 m−3), which implies a sink-strength dependency on α′
particularly with respect to cluster composition, size, and number den- formation [31].
sity. Previous APT studies on Fe-12Cr-9Al (at.%) neutron irradiated to Rate theory approaches simulating the irradiation enhanced α′-α
7 dpa at a similar temperature (320 °C) have reported α′ precipitates phase separation have indicated that near strong sinks such as grain
containing 51–58 at.% Cr and 4.5–5 at.% Al, in comparison to 61 at.% boundaries the excess concentration of defects is lower and there is a
and 5 at.% reported in this work [13,31]. The lower Al content in the
clusters supports the conjecture that Al-partitioning has occurred
from the α′ precipitates to the surrounding matrix previously reported
in both model FeCrAl alloys and in thermally aged legacy ODS FeCrAl
alloy PM2000 [31,37]. The reported α′ radius in this work is also within

Fig. 2. (a) Reconstructed control volume for OD03 showing blue 1.5 at.% (Y,YO,AlO)
isoconcentration surfaces for nanoscale precipitates with overlaid magenta Cr-rich
Fig. 1. Y, YO, and AlO atom maps overlaid with black Fe matrix atoms in the OD03 precipitates, (b) a Cr atom map for a selected precipitate and (c) 2D density contour
irradiated specimen. Grey 1.5 at.% (Y,YO,AlO) isoconcentration surfaces highlight plots on the selected control volume. The reduced density (ρr) in the 2D contour plot
observed nanoprecipitates while an embedded 2D Cr atomic concentration contour plot indicates the ratio of atomic density across the precipitate in comparison divided by the
on a cubic region of interest reveals a Cr-rich precipitate presumed to be α′. lowest atomic point density in the surrounding matrix.
 C.P. Massey et al. / Scripta Materialia 162 (2019) 94–98 97

Fig. 3. (a) A representative (Y,Al,O) nano-oxide precipitate showing AlO, YO, and Y atom maps. Black dots indicate the Fe matrix atoms. Also shown are (b) averaged 1D concentration
profiles through 50 (Y,Al,O) precipitates within the control volume as a function of normalized distance across the precipitates. Error bars indicate 95% confidence intervals for
measured elemental concentrations.

denuded zone absent of precipitates near the boundary [16]. Thus, the work, the ratio of Cr/Fe within the (Y,Al,O) precipitates is slightly higher
precipitate sink strength associated with a high number density of co- than the Cr/Fe ratio of the bulk specimen, but averaged 1D concentra-
herent to semi-coherent nanoprecipitates may be sufficient to retard tion profiles through 50 precipitates indicated no presence of an interfa-
some α′ precipitation by locally lowering the defect concentrations cial shell enriched in Cr (Fig. 3). The total concentration profile length
near the interfaces. Enriched Cr within the (Y,Al,O) precipitates may was chosen as 3× the radius of gyration discretized into 100 bins. Al-
also be intrinsic to the precipitate nucleation process, whereby though the data presented in this work doesn't support the existence
displaced Fe and Cr from the local precipitate control volume may result of a Cr-rich shell within the smallest (Y,Al,O) precipitates, there does
in a Cr-rich shell surrounding these precipitates [38–40]. In the present seem to be a substantial amount of Fe and Cr still within the clusters

Fig. 4. (a) the Ternary Y-Al-O diagram comparing nano-oxide compositions with expected stoichiometric phases for specimen OD03. Point diameters are proportional to nano-oxide
diameters, while colors are mapped using the displayed hit map frequency distribution. Also shown is the (b) distribution of Y/Al ratio for unirradiated 125YF alloy OD34 in
comparison to samples irradiated to 1.8 dpa at 215 °C (OD01), 357 °C (OD03), and 557 °C (OD06).
98 C.P. Massey et al. / Scripta Materialia 162 (2019) 94–98

even after aberration correction. Some authors have compared the Fe programs. Neutron irradiation of the 125YF specimens at ORNL's HFIR
and Cr contributions within ODS ferritic alloy nanoprecipitates to user facility was sponsored by the Scientific User Facilities Division, Of-
what would be expected in small angle neutron scattering experiments fice of Basic Energy Sciences, DOE. The authors would like to thank
and have argued that nearly all of the reported Fe and Cr in the Jatuporn Burns and Dr. Yaqiao Wu for their assistance with instrument
nanoprecipitates are artifacts of the APT technique [36]. Thus, the Fe operation at the MaCS facility and the Irradiated Materials Examination
and Cr contents measured in these precipitates are likely attributed to and Testing facility and LAMDA laboratory staff for their continuing sup-
the imperfect methodology used for compositional correction without port of this research. This material is based upon work supported under
knowing the Al-partitioning behavior between the precipitates and a Department of Energy Nuclear Energy University Programs (NEUP)
the matrix. Graduate Fellowship.
A ternary composition diagram has been generated using 1815 iden-
tified nanoprecipitates in the OD03 APT specimens to compare the Y, Al, References
and O ratios in comparison to expected stoichiometric compositions
[1] S.J. Zinkle, L.L. Snead, Annu. Rev. Mater. Res. 44 (1) (2014) 241–267.
(Fig. 4(a)). Discretizing the ternary diagram into 50 bins, the number [2] B.A. Pint, K.A. Terrani, M.P. Brady, T. Cheng, J.R. Keiser, J. Nucl. Mater. 440 (1–3)
of clusters in each bin was quantified and a frequency hit map was gen- (2013) 420–427.
erated describing the probability of finding clusters in representative [3] S. Dryepondt, K.A. Unocic, D.T. Hoelzer, C.P. Massey, B.A. Pint, J. Nucl. Mater. 501
(2018) 59–71.
compositional regions. Clusters appear to be generally non- [4] B.A. Pint, S. Dryepondt, K.A. Unocic, D.T. Hoelzer, JOM 66 (12) (2014) 2458–2466.
stoichiometric with a broad range of compositions. Smaller clusters [5] K.A. Terrani, S.J. Zinkle, L.L. Snead, J. Nucl. Mater. 448 (1–3) (2014) 420–435.
are less rich in Al, while the average Y/Al ratio tends to favor what [6] S.J. Zinkle, J.L. Boutard, D.T. Hoelzer, A. Kimura, R. Lindau, G.R. Odette, M. Rieth, L.
Tan, H. Tanigawa, Nucl. Fusion 57 (9) (2017), 092005. .
would be expected for the YAG (Y3Al5O12) composition. The unirradi-
[7] G.R. Odette, Scr. Mater. 143 (2018) 142–148.
ated (OD34) and irradiated specimens (OD01, OD06) exhibited the [8] F. Bley, Acta Metall. Mater. 40 (7) (1992) 1505–1517.
same compositional spread. In all specimens, the O content is highly [9] M. Bachhav, G. Robert Odette, E.A. Marquis, Scr. Mater. 74 (2014) 48–51.
[10] D. Chen, A. Kimura, W. Han, J. Nucl. Mater. 455 (1–3) (2014) 436–439.
underestimated, which is explained through the larger spatial distribu-
[11] E.R. Reese, M. Bachhav, P. Wells, T. Yamamoto, G. Robert Odette, E.A. Marquis, J.
tion in O atoms [41] and other APT artifacts [42]. This is supported by the Nucl. Mater. 500 (2018) 192–198.
substantial fraction of O remaining in the matrix after subtracting the [12] K.G. Field, X. Hu, K.C. Littrell, Y. Yamamoto, L.L. Snead, J. Nucl. Mater. 465 (2015)
identified (Y,Al,O) precipitates (Table 2). It has also been shown that un- 746–755.
[13] S.A. Briggs, P.D. Edmondson, K.C. Littrell, Y. Yamamoto, R.H. Howard, C.R. Daily, K.A.
calibrated laser parameters results in differences in measured O content Terrani, K. Sridharan, K.G. Field, Acta Mater. 129 (2017) 217–228.
in materials studied in laser-mode [43]. [14] K.G. Field, K.C. Littrell, S.A. Briggs, Scr. Mater. 142 (2018) 41–45.
Although there exist similar compositional spreads for all (Y,Al,O)- [15] M.H. Mathon, Y. de Carlan, G. Geoffroy, X. Averty, A. Alamo, C.H. de Novion, J. Nucl.
Mater. 312 (2–3) (2003) 236–248.
rich precipitates, there does seem to be an effect of irradiation on the [16] F. Soisson, T. Jourdan, Acta Mater. 103 (2016) 870–881.
compositions of these precipitates. Consider the distribution of Y/Al ra- [17] S. Kobayashi, T. Takasugi, Scr. Mater. 63 (11) (2010) 1104–1107.
tios presented in Fig. 4(b). Before irradiation, the median Y/Al ratio is [18] J. Ejenstam, M. Thuvander, P. Olsson, F. Rave, P. Szakalos, J. Nucl. Mater. 457 (2015)
291–297.
0.41, which is similar to what would be expected for the YAG phase [19] N.A. Bailey, E. Stergar, M. Toloczko, P. Hosemann, J. Nucl. Mater. 459 (2015)
(0.6). After irradiation, the Y/Al ratio increases, but the increase in this 225–234.
ratio decreases with increasing temperature. This suggests that at [20] C. Capdevila, M.K. Miller, K.F. Russell, J. Chao, J.L. González-Carrasco, Mater. Sci. Eng.
A 490 (1–2) (2008) 277–288.
lower temperatures where ballistic dissolution dominates, a greater
[21] C. Capdevila, M.K. Miller, J. Chao, Acta Mater. 60 (12) (2012) 4673–4684.
change in precipitate composition exists, while at higher temperatures [22] Z. Száraz, G. Török, V. Kršjak, P. Hähner, J. Nucl. Mater. 435 (1–3) (2013) 56–62.
where atoms are more able to back-diffuse to the precipitates, the Y/ [23] M.N. Gussev, E. Cakmak, K.G. Field, J. Nucl. Mater. 504 (2018) 221–233.
[24] M.K. Miller, K.F. Russell, Ultramicroscopy 107 (9) (2007) 761–766.
Al ratio tends to drive toward the initial precipitate composition.
[25] C.M. Parish, N.K. Kumar, L.L. Snead, P.D. Edmondson, K.G. Field, C. Silva, A.M.
An examination of the atom probe tomography results presented in Williams, K. Linton, K.J. Leonard, Microsc. Microanal. 21 (S3) (2015) 1003–1004.
this work suggests that the (Y,Al,O) nanoprecipitate sizes within this [26] B. Gault, M.P. Moody, J.M. Cairney, S.P. Ringer, Atom probe microscopy, Springer,
ODS FeCrAl alloy are stable under neutron irradiation to a dose of 2012.
[27] E.A. Marquis, J.M. Hyde, Mater. Sci. Eng. R. Rep. 69 (4–5) (2010) 37–62.
1.8 dpa at temperatures relevant to both current and advanced reactor [28] L.T. Stephenson, M.P. Moody, P.V. Liddicoat, S.P. Ringer, Microsc. Microanal. 13 (6)
concept temperature regimes. This is supported by the consistency of (2007) 448–463.
the sizes and number densities for these precipitates in comparison to [29] D. Vaumousse, A. Cerezo, P. Warren, Ultramicroscopy 95 (2003) 215–221.
[30] C.A. Williams, G.D. Smith, E.A. Marquis, Ultramicroscopy 125 (2013) 10–17.
the control sample across the irradiation condition studied. However, [31] P.D. Edmondson, S.A. Briggs, Y. Yamamoto, R.H. Howard, K. Sridharan, K.A. Terrani,
there may be an effect of irradiation temperature on the compositions K.G. Field, Scr. Mater. 116 (2016) 112–116.
of these precipitates as ballistic dissolution and back-diffusion of solute [32] K.A. Unocic, B.A. Pint, D.T. Hoelzer, J. Mater. Sci. 51 (20) (2016) 9190–9206.
[33] P. Dou, A. Kimura, R. Kasada, T. Okuda, M. Inoue, S. Ukai, S. Ohnuki, T. Fujisawa, F.
atoms compete. In the sample irradiated at 357 °C, α′-α phase separa- Abe, J. Nucl. Mater. 444 (1–3) (2014) 441–453.
tion was noted and quantified, with compositions and sizes similar to [34] P. Dou, A. Kimura, R. Kasada, T. Okuda, M. Inoue, S. Ukai, S. Ohnuki, T. Fujisawa, F.
those previously investigated in neutron irradiated non-ODS FeCrAl al- Abe, S. Jiang, Z. Yang, J. Nucl. Mater. 485 (2017) 189–201.
[35] D. Blavette, F. Vurpillot, P. Pareige, A. Menand, Ultramicroscopy 89 (1–3) (2001)
loys of similar composition. A significant finding is that the number den- 145–153.
sity of α′-precipitates is lower by about an order of magnitude in 125YF [36] C. Hatzoglou, B. Radiguet, P. Pareige, J. Nucl. Mater. 492 (2017) 279–291.
than in model FeCrAl after irradiation at 357 °C. [37] C. Capdevila, M.K. Miller, K.F. Russell, J. Mater. Sci. 43 (11) (2008) 3889–3893.
[38] E.A. Marquis, Appl. Phys. Lett. 93 (18) (2008).
[39] L.L. Hsiung, M.J. Fluss, A. Kimura, Mater. Lett. 64 (16) (2010) 1782–1785.
Acknowledgments [40] L.L. Hsiung, M.J. Fluss, S.J. Tumey, B.W. Choi, Y. Serruys, F. Willaime, A. Kimura, Phys.
Rev. B 82 (18) (2010), 184103. .
Primary characterization efforts were funded by a Nuclear Science [41] C.A. Williams, E.A. Marquis, A. Cerezo, G.D.W. Smith, J. Nucl. Mater. 400 (1) (2010)
37–45.
User Facilities Rapid Turnaround Experiment (17-961) and were con- [42] B. Gault, D.W. Saxey, M.W. Ashton, S.B. Sinnott, A.N. Chiaramonti, M.P. Moody, D.K.
ducted at the Center for Advanced Energy Studies-Microscopy and Schreiber, New J. Phys. 18 (3) (2016).
Characterization Suite (CAES-MaCS) with the help of Jatuporn Burns [43] S. Pedrazzini, A.J. London, B. Gault, D. Saxey, S. Speller, C.R.M. Grovenor, M. Danaie,
M.P. Moody, P.D. Edmondson, P.A.J. Bagot, Microsc. Microanal. 23 (2) (2017)
and Dr. Yaqiao Wu. A portion of the irradiation campaign, including cap- 414–424.
sule design, capsule opening, radioactive specimen handling/prepara-
tion, and data analysis, was sponsored by the DOE-NE, Advanced Fuel
Campaign of the Nuclear Technology Research and Development