Scripta Materialia 146 (2018) 213–217

Contents lists available at ScienceDirect

Scripta Materialia

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

Regular article

Dose rate dependence of Cr precipitation in an ion-irradiated

Fe\\18Cr alloy
Elaina R. Reese a, Nathan Almirall b, Takuya Yamamoto b, Scott Tumey c,
G. Robert Odette b, Emmanuelle A. Marquis a,⁎
a
Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109, USA
b
Materials Department, University of California, Santa Barbara, CA 93106, USA
c
Lawrence Livermore National Laboratory, Livermore, CA, 94550, USA

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

Article history: Precipitation of α′ in Fe\\Cr alloys under neutron irradiation is thermodynamically driven while being accelerated
Received 22 August 2017 by radiation-enhanced diffusion. However, similar alloys under ion irradiation at high dose rates (N 10−4 dpa/s)
Received in revised form 19 November 2017 fail to exhibit α′ precipitation. Here, the microstructure of an Fe\\18Cr alloy under ion or neutron-irradiation at
Accepted 21 November 2017
300 °C at dose rates from ~ 10−7 to 10−4 dpa/s was analyzed by atom probe tomography. The steady-state
Available online 6 December 2017
composition content of the clusters depends on the ion irradiation dose and dose rate, confirming the contribution
Keywords:
of ballistic mixing in diluting the Cr concentration in non-equilibrium α′ precipitates.
Fe-Cr © 2017 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Ion irradiation
Precipitation
Dose rate
Atom probe tomography

Ferritic/martensitic (F/M) alloys, with high Cr content, are leading
candidates for nuclear reactor components because of their corrosion
resistance and favorable mechanical properties [1]. However, the evolu-
tion of F/M alloy microstructures and the corresponding degradation of
their mechanical properties under neutron irradiation are a concern.
One issue is the hardening and embrittlement caused by the precipita-
tion of the α′ phase observed in both model and commercial Fe\\Cr
alloys under varying neutron irradiation conditions [2–8].
Prior observations of α′ precipitation in neutron-irradiated Fe\\Cr
alloys above 300 °C, are consistent with a thermodynamically driven
process that is kinetically accelerated by radiation-enhanced diffusion
(RED) [3,4,8–12]. Small angle neutron scattering (SANS) and atom
probe tomography (APT) studies of neutron-irradiated and thermally
aged model Fe\\Cr alloys found near equilibrium α′ compositions,
from 80 to 95 at.% Cr for temperature from 320 to 500 °C [6,12–15].
However, a number of APT studies on binary Fe\\Cr alloys also reported
significantly lower than equilibrium Cr concentrations, down to
b50 at.% Cr, generally for smaller α′ precipitates (r b 1 to 1.5 nm) [5,
16]. These values were rationalized in terms of technique artifacts [12,
17] including data analysis procedures [16], alloying effects [7], interface
energy effects on the local non-equilibrium thermodynamics [13], and
perhaps some contribution of ballistic mixing [12].
⁎ Corresponding author.
E-mail address: emarq@umich.edu (E.A. Marquis).
While the α′ phase is observed in ferritic alloys containing N 9 at.% Cr
and neutron-irradiated at 300 °C, self-ion irradiations of similar alloys in
the temperature range of 300 to 500 °C and up to 400 dpa yielded no
clustering at high dose rates (10−3 to 10− 2 dpa/s) [18,19]. At lower
dose rates (10−5 to 10−4 dpa/s), limited clustering [20], in addition to
dilution and dissolution of precipitates or spinodal structures [21,22]
were observed. In contrast, electron irradiations at ~300 °C at interme-
diate dose rate of 4 × 10−5 dpa/s produced distinct α′ precipitates with
the near equilibrium Cr concentration (96 at.% Cr) [23,24]. Proton
irradiation of a commercial HT9 alloy at 400 °C to 7 dpa at 10−5 dpa/s,
yielded distinct α′ precipitates, but with a low non-equilibrium Cr
concentration (48 at.%) [25], that cannot be fully explained by the
technique and analysis limitations described above. Instead, the modifi-
cation or suppression of α′ precipitation need to be analyzed in light of
differences among the neutron, ion, electron, and proton irradiation
conditions.
Ballistic dissolution plays a less significant role in proton irradiations
and much less in electron irradiations than in neutron irradiations. These
differences are in part related to the primary knock-on energies that are
on average 35 keV, 5 keV, 200 eV and 60 eV in Fe for 1 MeV neutrons,
self-ions, protons and electrons, respectively [26]. Another key differ-
ence is the much higher dose rates at which ion irradiations are typically
conducted. Therefore, the most important mechanism is the increase in
interstitial and vacancy recombination with dose rate, since RED of Cr
that leads to accelerated α′ precipitation depends on the defects that
escape recombination or annihilation at irradiation-induced sinks. It is

https://doi.org/10.1016/j.scriptamat.2017.11.040
1359-6462/© 2017 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
214 E.R. Reese et al. / Scripta Materialia 146 (2018) 213–217
well established that higher dose rates delay precipitation to higher
doses [27–31]; however, at sufficiently high dose, the precipitation of
Cr would still saturate as the α matrix and α′ precipitates approach
their respective phase boundaries. Another dose rate-dependent
mechanism is related to the rate of ballistic mixing versus the rate of
self-healing recovery that drives precipitates towards their equilibrium
composition by thermal back diffusion of Cr. There is insufficient time
for full recovery at high dose rates, while at low dose rates, self-healing
is dominant. This mechanism qualitatively rationalizes the differences
in α′ precipitation between proton [25], ion [20], and electron [23]
irradiations performed at comparable dose rates.
To test the hypothesis that α′ precipitation is affected by dose rate,
primarily by the ballistic mixing mechanism, a series of Fe self-ion irra-
diations were performed, spanning three orders of magnitude in dose
rate. The resulting spatial distributions of Cr were quantified using
APT. The ion irradiation results were also compared to lower dose rate
neutron irradiations at 320 °C.
The Fe-18 at.% Cr alloy in this study has been included in several
irradiation experiments dating back to the 1980s. Details of the solution
annealing heat treatment used to produce a low dislocation density are
summarized in [32]. Note that this alloy contained b0.1 at.% Si, V, and P
and very few and isolated Si and P atoms clusters were observed (see
Supplementary Material), but are not the focus of this work. The
measured C contents was below 0.008 at.% with no noticeable contam-
ination during irradiation. Two irradiations were performed at the Cen-
ter for Accelerator Mass Spectrometry (CAMS) at Lawrence Livermore
National Laboratory using a defocused beam of 70 MeV Fe2 + ions at
290 ± 5 °C. The same nominal doses (3 dpa at the damage peak) were
reached at peak dose rates designed to be a factor of 10 apart. The
dose profiles (Fig. 1) were estimated using SRIM calculations performed
using the Kinchin-Pease model with a displacement energy of 40 eV
[33]. The large penetration depth (~6 μm) of the 70 MeV ions allows ex-
amination of the effects of a correspondingly large range of dose and
dose rates at various depths. To extend the range of dose, an additional
300 ± 5 °C ion irradiation was performed at the Michigan Ion Beam Lab-
oratory (MIBL) using a defocused beam of 5 MeV Fe2+ ions to a nominal
dose of 6.0 dpa at a depth of 0.5 μm and a dose rate of 3 × 10−5 dpa/s.
Note the dose rates are based on estimates at the FIB liftout sample
location. The MIBL experimental procedure is described elsewhere, e.g
[34]. In addition, previously reported neutron irradiation data at
320 °C to ≈ 1.8 dpa [6] and 6.5 dpa [12] at a rate of 3 × 10− 7 dpa/s
was compared to the ion irradiation results.
APT specimens were prepared by following standard liftout and
milling procedures using a FEI Helios 650 Nanolab scanning electron mi-
croscope (SEM) and focused ion beam (FIB) instrument. The apexes of
the APT needles were positioned at various specified distances from
the sample surfaces. The same depths were used for the two CAMS
Fig. 1. Damage and dose rate profiles for the two CAMS irradiations.
samples to compare the irradiation microstructures at the same dose.
For the MIBL irradiation, APT specimens were extracted at depths be-
tween 0.5 and 0.6 μm to mitigate possible surface and/or end-of-range
effects. Data collection was performed using a Cameca LEAP 4000XHR
instrument operated in voltage pulsing mode using a pulse fraction of
20%, a pulse repetition rate of 200 kHz, data collection rate of 3 atoms
per 1000 pulses, and specimen temperature of 50 K. The CAMECA Inte-
grated Visualization & Analysis Software (IVAS) version 3.6.10 was used
for data reconstruction. The values of the image compression factor and
field factor were adjusted to the appropriate d-spacing between planes
in visible and identifiable poles, and uniform depth and radial atomic
densities [35]. The values of the image compression factor and field fac-
tor ranged between 1.2 and 1.5 and between 3.8 and 4.6, respectively.
Cr-rich regions, hereafter referred to as Cr clusters, were visible in all
of the irradiation conditions, with characteristics that varied with dose
and dose rates. Fig. 2 qualitatively highlights the differences in the
spatial distribution of Cr atoms that are colored based on the local Cr
concentration measured in spheres of 1 nm diameter. For a specified
dose rate of ~10−5 dpa/s, the extent of Cr clustering visually increased
with dose. Similarly, for a given dose (1.5 to 2 dpa), Cr clustering
decreased with increasing dose rates. Among the ion-irradiated micro-
structures, the most significant clustering was observed for the doses
of 2.4 and 6.0 dpa and dose rate of 3 × 10− 5 dpa/s. In this case, the
largest clusters reached a nominal core concentration of 55 ± 6 at. %
Cr, that was estimated using a proximity histogram with a 28 at.%
Cr isoconcentration surface and considering only the largest clusters
with N1 nm radii. While this relatively low concentration value
agrees with both prior ion- and neutron-irradiated data, e.g. [5,16,
19,22], it deviates significantly from the 320 °C neutron-irradiated
data at 6.5 dpa and 3 × 10− 7 dpa/s, where the composition of the
larger Cr-rich clusters reached 84.2 ± 1.1 at. % [12].
The qualitative trends are clear. However, in the absence of well
formed α′ precipitates, composition and size quantification of dilute
clusters is difficult to determine from APT data. Thus, the amplitude of
the Cr fluctuations and a size metric of the Cr-rich clusters for each irra-
diation condition were assessed using the radial distribution functions
of the Cr atoms [36]. The curves were fitted using the equation
suggested in [37]:
   
C Cr ðr Þ A2 2πr r
o ¼ cos exp − þ1
C Cr 2C o
2 λ1 λ2
Cr
where CCr(r) is the average concentration at the distance r from a Cr
A2
atom, CoCr is the average matrix Cr concentration, 2 is the amplitude
2C oCr
of the concentration fluctuations around the average CoCr, λ1 is the wave-
length of the fluctuation, and λ2 is the decay length of the fluctuation.
The amplitude of the Cr fluctuations, which does not bear a one to one
relation to the discrete core concentration, but rather reflects a quanti-
fied trend in how discrete Cr cluster compositions varied with the
irradiation conditions (Fig. 3a). Notably, the amplitudes of the ion-
irradiated clusters were significantly lower than those measured in
the neutron-irradiated case. This reflects the much lower Cr concentra-
tions in the ion-irradiated clusters compared to those observed in the
neutron-irradiated condition. The amplitudes appeared to saturate
with increasing dose. Further, the decay length appeared to be a good
approximation of cluster radius (Fig. 3b). For example, it was in very
good agreement with previously measured cluster radii for 1.8 dpa
A2
neutron data [6]. For a given dose (2 ± 1 dpa), the amplitude ( 2)
2C oCr
and cluster size (λ2) decreased with increasing dose rate, while the
number density, represented as λ−3 1 increased (Fig. 3b). We note that
the clusters were slightly smaller after ion irradiation and much larger
in the Tissot et al.’s electron irradiation data [23], compared to the
neutron irradiated conditions [6,12].
 E.R. Reese et al. / Scripta Materialia 146 (2018) 213–217
Fig. 2. 2 nm thick atom maps from reconstructed APT volumes obtained after ion and neutron irradiations. Color gradient is associated with local Cr concentration measured in 1 nm
diameter spheres.
Most prior ion irradiations focused on probing only intermediate
depths where dose and dose rate gradients are minimal and that pro-
vide the best compromise between near surface and near end of range
effects. Unlike a previous study [20], no significant end of range effect
was obvious at the peak damage location. The diffusion length scale is
of the order of the inter-cluster distances (~ 5–15 nm), thus a given
depth represents an individual experimental condition. The larger
range of depths that can be probed using the high-energy 70 MeV Fe
ions provides a notable opportunity lab-on-a-chip approach to irradia-
tion experiments. In combination with previously reported neutron
and electron results, the ion irradiation data obtained in this study, pro-
vide an extremely rich basis for identifying key mechanisms controlling
Cr clustering and precipitation over a wide range of irradiation
conditions.
We first consider RED that depends on dose rate and particle type,
and argue that these factors are not sufficient to fully explain the
present results and that a significant contribution from ballistic mixing
is needed.
Cr diffusion is accelerated by RED as mediated by excess concentra-
tions of vacancies under irradiation [27]. The steady-state concentration
of vacancies can be estimated from a rate theory balance of vacancy and
self-interstitial atom (SIA) generation and loss (annihilation) rates both
at sinks and by vacancy-SIA mutual recombination (self-healing). Re-
combination during long-range defect diffusion increases with increas-
ing dose rate and decreasing temperature [27]. In the recombination
dominated limit, RED scales with the square root of dose rate. The alloy
chemistry and particle types also affect RED rates since recombination
is strongly enhanced by solute vacancy traps, as well as by persistent
cascade fragment defect sinks [27–29]. Recombination-mediated dose
rate effects on RED exist in all of the irradiations in this study. Since
ions and neutrons both produce cascades, their RED mechanisms are
qualitatively similar. However, it is noted that the high energy ions
produce lower energy primary knock on atoms, hence smaller cascades;
215
this may affect RED along the ion penetration path as well as the ballistic
mixing. Further, there may be effects of injected interstitials near the end
of range. However, is beyond the scope of the current letter paper to
address these effects in detail.
As seen in [23], RED rates under electron irradiations, that produce
only point defect Frenkel pairs, are far higher than neutrons and ions
that generate cascades. The enhanced RED efficiency is due to the
combination of higher defect production of electrons per dpa, by a factor
of ≈3 [26], and the absence of cascade sinks. Neutron irradiation at
3 × 10− 7 dpa/s also shows increased Cr-cluster growth rates compared
to ion irradiation at higher dose rate (10−5 dpa/s). However, RED itself
does not explain the low Cr content of the clusters, which instead can
be rationalized by considering ballistic mixing. Atomic recoils in
displacement cascades continuously eject Cr cluster solutes back into
the matrix and Fe matrix atoms into the cluster. At low dose rates, the
predominant fate of the dynamic mixing is radiation-enhanced back dif-
fusion of Cr and Fe atoms so as to recover the near equilibrium composi-
tion of the cluster. At high dose rates, there is not enough time for full
diffusional recovery, eventually leading to the establishment of a steady
state between the mixing and unmixing processes at lower than thermo-
dynamic equilibrium cluster Cr contents. In the limit, clusters may
become unstable, or at least undetectable. The present observations
spanning three orders of magnitude in dose rate are clearly consistent
with a ballistic mixing mechanism. The amplitude of the radial distribu-
tion function in Fig. 3a saturates, i.e. reaches steady-state, at doses
N0.6 dpa for the three ranges of dose rates examined (3 × 10 − 7 ,
0.6-3 × 10− 5, and 0.6-2 × 10 − 4 dpa/s). The saturated Cr amplitude
decreases approximately linearly with the log of the dose rate.
Additional measurements would be needed to confirm the high dose
behavior at 10−6 dpa/s.
There are at least three possible reasons for the low measured
Cr concentrations: APT artifacts; non-equilibrium transition phase
thermodynamics-kinetics; and, ballistic mixing. It was previously
216 E.R. Reese et al. / Scripta Materialia 146 (2018) 213–217

wide range of dose rates from 10−8 to 10−3 dpa/s. While the results
cannot be compared on a one-to-one basis at this time, the model pre-
dictions are remarkably quantitatively consistent with the experimental
results reported here.
In summary, this work presents a novel experimental approach
using very high energy self-ion irradiations to probe dose and dose
rate effects. Decomposition of the supersaturated Cr solid solution
under ion irradiation results in composition fluctuations (amplitude
and length scales) that vary strongly with dose rate. We propose that
the difference with respect to thermal aging or neutron and electron ir-
radiation is a result of cascade ballistic mixing, or enhanced mixing, that
leads to a steady-state non-equilibrium clusters with lower Cr contents.
The authors acknowledge technical support and funding from: the
U.S. Department of Energy (DOE) Integrated Research Project award
DE-NE0000639; the DOE Office of Nuclear Energy Idaho Operations
Office Contract DE-AC07-051D14517 for the ATR National Scientific
User Facility UCSB ATR-1 irradiation experiment and post irradiation
sample and FIB access at the Center For Advanced Energy Studies, the
University of Michigan Center for Materials Characterization, the
Michigan Ion Beam Laboratory, the DOE Office of Fusion Energy Sciences
(DE-FG03-94ER54275), the Center for Accelerator Mass Spectrometry
at Lawrence Livermore National Laboratory (performed under the
auspices of the U.S. Department of Energy by Lawrence Livermore
National Laboratory under Contract DE-AC52-07NA27344).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.

org/10.1016/j.scriptamat.2017.11.040.

References
Fig. 3. A) dose dependence of the fluctuation amplitude parameter (A) for the neutron and [1] R.L. Klueh, A.T. Nelson, J. Nucl. Mater. 371 (2007) 37–52.
ion irradiation conditions differentiated by dose rate. B) fluctuation amplitude parameter [2] P.J. Grobner, Transactions 4 (1973) 251–260.
(A), cluster size (λ2, nm), cluster number density (λ−3 23 3
1 , 10 /m ), for neutron and ion [3] J.J. Kai, R.L. Klueh, J. Nucl. Mater. 230 (1996) 116–123.
irradiation conditions at doses from 1 to 3 dpa. Neutron irradiation conditions are [4] M.H. Mathon, Y. de Carlan, G. Geoffroy, X. Averty, A. Alamo, C.H. de Novion, J. Nucl.
represented by diamonds, the 5 MeV ion irradiation condition by a triangle, and the Mater. 312 (2003) 236–248.
70 MeV ion irradiation conditions by disks, except for the 2.4 dpa condition represented [5] V. Kuksenko, C. Pareige, C. Genevois, F. Cuvilly, M. Roussel, P. Pareige, J. Nucl. Mater.
by squares. 415 (2011) 61–66.
[6] M. Bachhav, G.R. Odette, E.A. Marquis, Scr. Mater. 74 (2014) 48–51.
[7] P.D. Edmondson, S.A. Briggs, Y. Yamamoto, R.H. Howard, K. Sridharan, K.A. Terrani,
K.G. Field, Scr. Mater. 116 (2016) 112–116.
argued that common APT artifacts associated with element and feature- [8] R.D. Griffin, D.S. Gelles, R.A. Dodd, G.L. Kulcinski, J. Nucl. Mater. 179 (1991) 714–717.
[9] D.S. Gelles, J. Nucl. Mater. 148 (1987) 136–144.
specific evaporation fields and trajectory aberrations limit accurate
[10] P. Dubuisson, D. Gilbon, J.L. Séran, J. Nucl. Mater. 205 (1993) 178–189.
compositional measurements for α′ precipitates at small sizes [16]. [11] S.I. Porollo, A.M. Dvoriashin, A.N. Vorobyev, Y.V. Konobeev, J. Nucl. Mater. 256
This effect was evaluated in detail for neutron irradiation data [12]. (1998) 247–253.
[12] E. Anderson, M. Bachhav, P. Wells, T. Yamamoto, G.R. Odette, E.A. Marquis, J. Nucl.
APT artifacts were judged to be the most likely reason for the wide
Mater. (2017) (Submitted to).
range of α′ compositions reported in the literature for neutron irradia- [13] S.S. Brenner, M.K. Miller, W.A. Soffa, Scr. Metall. 16 (1982) 831–836.
tions and thermal annealing (~50 to 95 at.%) conditions. Thermal an- [14] S. Novy, P. Pareige, C. Pareige, J. Nucl. Mater. 384 (2009) 96–102.
nealing and electron irradiation experiments reported an increase in [15] S.V. Rogozhkin, O.A. Korchuganova, A.A. Aleev, Inorg. Mater. Appl. Res. 7 (2016)
210–213.
the α′ Cr content with time [14,15,22] and dose [23], respectively. [16] W.-Y. Chen, Y. Miao, Y. Wu, C.A. Tomchik, K. Mo, J. Gan, M.A. Okuniewski, S.A. Maloy,
While non-equilibrium thermodynamics might contribute to an initial J.F. Stubbins, J. Nucl. Mater. 462 (2015) 242–249.
evolution of cluster Cr compositions [38], we note that the cluster [17] F. Bergner, C. Pareige, V. Kuksenko, L. Malerba, P. Pareige, A. Ulbricht, A. Wagner, J.
Nucl. Mater. 442 (2013) 463–469.
sizes considered in these previous studies also increased from b1 nm [18] C. Pareige, V. Kuksenko, P. Pareige, J. Nucl. Mater. 456 (2015) 471–476.
up to 3 nm with increasing time or dose; and the cluster Cr content [19] E.A Marquis, B. Wirth, G. Was, Characterization and Modeling of Grain Boundary
tracked the size. Here, however, the significantly higher amplitude Chemistry Evolution in Ferritic Steels under Irradiation. United States, 2016 https://
doi.org/10.2172/1248953 http://www.osti.gov/scitech/servlets/purl/1248953.
after neutron irradiation at 1.8 dpa compared to the ion-irradiated [20] O. Tissot, C. Pareige, E. Meslin, B. Décamps, J. Henry, Mater. Res. Lett. (2016) 1–7.
condition at 6.0 dpa support a clear dose rate influence on the effects [21] K. Fujii, K. Fukuya, J. Nucl. Mater. 440 (2013) 612–616.
of ballistic mixing. Both conditions exhibited similar cluster sizes [22] O.A. Korchuganova, M. Thuvander, A.A. Aleev, S.V. Rogozhkin, T. Boll, T.V. Kulevoy, J.
Nucl. Mater. 477 (2016) 172–177.
(~ 1.2–1.5 nm) that would presumably be affected by roughly equiva-
[23] O. Tissot, C. Pareige, E. Meslin, B. Decamps, J. Henry, Scr. Mater. 122 (2016) 31–35.
lent APT artifacts and thermodynamic effects. [24] H. Takahashi, S. Ohnuki, T. Takeyama, J. Nucl. Mater. 104 (1981) 1415–1419.
We realize that the necessarily very brief discussion above is qualita- [25] Z.J. Jiao, V. Shankar, G.S. Was, J. Nucl. Mater. 419 (2011) 52–62.
[26] G.S. Was, Fundamentals of Radiation Materials Science: Metals and Alloys, Springer,
tive, and not fully satisfying. These issues are being addressed in a
New York, 2007.
detailed phase field based model developed by Ke and co-workers, [27] G.R. Odette, T. Yamamoto, D. Klingensmith, Philos. Mag. 85 (2005) 779–797.
and that will soon be submitted for publication [39]. The model [28] G.R. Odette, R.K. Nanstad, JOM 61 (2009) 17–23.
addressed dose rate effects in a Fe\\15Cr alloy that are associated [29] G.R. Odette, E.V. Mader, G.E. Lucas, W.J. Phythian, C.A. English, 16th International
Symposium on the Effects of Irradiation on Materials, ASTM STP, 1175, 1995 373.
with both ballistic mixing and RED and was calibrated with the neutron [30] H. Ke, P. Wells, P.D. Edmondson, N. Almirall, L. Barnard, G.R. Odette, D. Morgan, Acta
data in Ref [6] and applied to both ion and electron irradiation over a Mater. 138 (2017) 10–26.
 E.R. Reese et al. / Scripta Materialia 146 (2018) 213–217 217

[31] G.R. Odette, T. Yamamoto, E.D. Eason, R.K. Nanstad, Proceedings of the International
Symposium on Research for Aging Management of Light Water Reactors, Institute of
Nuclear Safety Systems, 2008 279.
[32] D.S. Gelles, J. Nucl. Mater. 225 (1995) 163–174.
[33] J.F. Ziegler, J. Biersack, U. Littmark, The Stopping and Range of Ions in Matter,
Pergamon Press, 1985.
[34] E. Getto, Z. Jiao, A.M. Monterrosa, K. Sun, G.S. Was, J. Nucl. Mater. 465 (2015)
116–126.
[35] D. Haley, in: A. Ceguerra (Ed.), 3Depict - Visualization & Analysis for Atom Probe,
2010.
[36] J. Zhou, J. Odqvist, M. Thuvander, P. Hedström, Microsc. Microanal. 19 (2013)
665–675.
[37] L. Couturier, F. De Geuser, A. Deschamps, Mater. Charact. 121 (2016) 61–67.
[38] V. Svetukhin, P. L'vov, E. Gaganidze, M. Tikhonchev, C. Dethloff, J. Nucl. Mater. 415
(2011) 205–209.
[39] J.H. Ke, D. Morgan, University of Wisonsin, Madison, Private Communication, 2017.