Evaluation of damage in neutron irradiated boron

carbide
Dominique Gosset, Patrick Herter, Vianney Motte

To cite this version:

Dominique Gosset, Patrick Herter, Vianney Motte. Evaluation of damage in neutron irradiated boron
carbide. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with
Materials and Atoms, 2018, 434, pp.66-72. �10.1016/j.nimb.2018.08.021�. �cea-02430090�

HAL Id: cea-02430090

https://cea.hal.science/cea-02430090
Submitted on 7 Jan 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

Contents lists available at ScienceDirect

Nuclear Inst. and Methods in Physics Research B

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

Evaluation of damage in neutron irradiated boron carbide T

a,⁎ b a
Dominique Gosset , Patrick Herter , Vianney Motte
a
CEA, Université Paris-Saclay, DEN-SRMA, F-91191 Gif sur Yvette, France
b
CEA, Université Paris-Saclay, DEN-DISN-VALO, F-91191 Gif sur Yvette Cedex, France

A R T I C LE I N FO A B S T R A C T

Keywords: When irradiated in a reactor, neutron absorber materials undergo two different damages induced, first, by the
Neutron absorber elastic scattering of neutrons; second by the neutron absorption reactions. In boron carbide irradiated in a fast
Damage rate neutron flux, neutron scattering and the slowing-down of He and Li arising from the (n,α) absorption reaction
Boron carbide both lead to the displacement of B and C atoms which energy ranges up to a few MeV. Simulating the neutron
Ion implantation simulation
damage by ion irradiation requires the calculation of the damage produced in both cases. In this paper we
propose an estimation of the actual defect production rate resulting from the fast neutron irradiation. For this,
we first estimate the energy spectra of both the primary knocked-on atoms and atoms created by the absorption
reactions, here in a Phenix-like neutron spectrum. We then deduce from SRIM calculations the actual energy
distribution of all the atoms displaced along the displacements cascades induced by those primary projectiles. At
last, we obtain an estimation of the number of displaced atoms per produced helium, about 305, most of them
resulting from neutron scattering. This is far from a Kinchin-Pease or NRT estimation, of the order of several
thousands, this arising from the fact that most of the energy is dissipated through electronic interactions. Such
results are then used in order to perform ion irradiations aiming at a realistic simulation of synergetic effects of
helium implantation and damage production.

1. Introduction one. In fast neutron reactor, helium can be produced in large quantities,
about 1022/cm3 (i.e. concentration about 0.1), leading to swelling and
Ion implantations are widely used to simulate the damage and the cracking of the material. 4He and 7Li are emitted with energies in the
composition change arising in materials irradiated in nuclear plants MeV range (mass defect of the main absorption reaction). Owing to the
[1,2]. However, some parameters can hardly be accounted for; for ex- low atomic mass of the constituents of the material, they can then
ample, a reactor irradiation most often last up to a few years to be produce structural defects. On the other hand, in a fast neutron flux, the
compared to a few hours for an ion implantation, this making kinetics primary knocked-on atoms (pka) produced by neutron scattering have
effects difficult to handle. Another parameter is the way the energy is an energy spectrum up to a few MeV, high enough to displace the atoms
dissipated into the material. The slowing-down of the particles happens of the material. In both cases, neutron absorption or scattering, the
according to two different processes [3]; either an interaction with the nuclei emerging from the reaction are light ones (B, C, Li, He). They will
electrons (electronic slowing-down, associated with the electronic slow-down in a matrix also constituted of light nuclei. In that case, it is
stopping power, Se) or with the nuclei (nuclear stopping power, Sn). known that most of the pka energy is dissipated by electronic interac-
The former is of low consequences in metals (energy dissipation in the tion. However, some atomic displacements arise, mostly at the end of
electron sea) but can induce important effects in insulators or semi- the ion paths. The questions which arise are then the ratio of produced
conductors [4]. Moreover, in some conditions, nuclear reactions helium to the total number of displaced atoms, the way energy is dis-
happen (fuel, neutron absorber, tritium blankets in ITER, spallation sipated, and the consequence of the damage on the kinetics of helium,
sources…). In those cases, first the composition of the material is e.g. nucleation and growth of clusters.
modified, but second the reaction products can be energetic enough to Simulating the behavior of helium in boron carbide with ion im-
produce defects. plantations then leads to wonder if the associated damage in reactor can
We are here interested in the behavior of helium in boron carbide. be reproduced as well as the gas concentration. For that, it is necessary
Boron carbide is widely used as a neutron absorber, either in thermal or to perform ion irradiations which parameters such as the damage to
fast neutron flux [5]. The main absorption reaction is the 10B(n,α)7Li helium concentration ratio or the Se/Sn ratio are precisely known and

⁎
Corresponding author.
E-mail addresses: [email protected] (D. Gosset), [email protected] (P. Herter), [email protected] (V. Motte).

https://doi.org/10.1016/j.nimb.2018.08.021
Received 20 November 2017; Received in revised form 7 August 2018; Accepted 17 August 2018
Available online 25 August 2018
0168-583X/ © 2018 Elsevier B.V. All rights reserved.
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

Fig. 1. Neutron spectrum at the center of the late French Phenix LMFBR [10]. Fig. 2. Neutron scattering cross sections on the isotopes to be considered, 10
B,
11
B, 12C, 7Li, 4He, from [11].
representative. Such estimations can be done e.g. with tools such as
SRIM [6]. But the comparison with the reactor conditions requires
making this estimation with the actual reactor parameters. The avail-
able data are the material characteristics, the neutron flux in the reactor
and the interaction cross sections. Here, we detail the route we have
followed from those nuclear data to the atoms displacement rate (ob-
taining the helium production rate is straightforward, by convoluting
the neutron spectrum and the absorption cross section). The first step is
calculating the so-called primary damage spectrum [7–9], i.e. the en-
ergy distribution of the pka and of the created atoms by the absorption
reactions. We then use the SRIM software to estimate the number of
displaced atoms as a function of the pka’s energy and the way energy is
dissipated. At last, those results allow us to estimate the helium to
atomic displacements ratio and the electronic stopping power range.

2. Data

We consider a neutron spectrum equivalent to the one in the center

of the late French Phenix LMFBR (sodium cooled fast neutron reactor)
[10]. The spectrum is divided in 36 groups (Fig. 1). The total flux is
about 3.1015 n.cm–2.s–1. 10
Fig. 3. Neutron absorption cross section for the B(n,α)7Li and 10
B(n,2α)3H
The neutron interaction cross sections (neutron scattering and ab- reactions, from [11].
sorption) are taken from the ENDF data base [11]: Figs. 2 and 3. For the
absorption reactions, we consider only the 10B(n,α)7Li one. We then
neglect the 10B(n,2α)3H reaction. As a matter, the cross section for this The initial 10B enrichment of boron is taken as 48%, which reduces to
reaction is about 10−6 lower than the (n,α) one, excepted in the fast 39% for a 1022.cm−3 burnup.
neutron range (where it is rather poorly known [12]), where the neu- Estimating the atom displacement rates requires the knowledge of
tron flux significantly decreases, leading to a production rate about one the displacement energy threshold (TDE) for all the atoms of the ma-
thousand lower than the (n,α) one. Second, it has a lower mass defect, terial. Here we consider only mean values. For B and C, we have con-
about 0.23 MeV to be compared to about 2.8 MeV for the (n,α) reaction, sidered the values proposed in SRIM, which are coherent with the es-
leading to a much lower energy of the pka’s. Excepted in the fast timations performed by Zuppiroli et al. [13]. For He, we have used a
neutron range, the (n,α) absorption cross section varies as σ = 611/√E most recent estimation of the thermal diffusion energy [14]. For Li, we
(σ in barn, E in eV). Convoluting the neutron spectrum and the (n,α) have used an arbitrary, intermediate value: the only available value is
absorption cross section leads to the helium production rate: for lithium aluminate, about 22 eV, i.e. in a quite different electronic
environment with strong ionic bondings we have supposed higher than
τHe = 4, 01 × 10−9 at−1. s−1 in B4C [15]. Now, on a formal point of view, this value is of low im-
portance: Li emitted with energy close to the TDE is unable to displace
where ‘at’ stands for an equivalent boron carbide atom (initially full
B or C and is of low efficiency on He. The values are then the followings:
density B0.8C0.2, 48% 10B enriched, 1022/cm3 burnup: see here under).

• B: 25 eV
In order to have a good estimation of the damage on all the atoms,
including the neutron capture products, we consider a highly irradiated
boron carbide sample. We have chosen a 1022.cm−3 capture density • C: 28 eV
• He: 2 eV
(most often called burnup). The boron concentration in non-irradiated
B4C boron carbide is about 1.1x1023.cm−3. As a result, the mean atomic • Li: 15 eV
composition of the irradiated material is taken as B0.72C0.2He0.08Li0.08.

67
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

3. Primary damage spectra

The primary damage spectra are the sum of two components:

• The pka arising from the ballistic scattering of the neutron,

• The atoms emerging from the neutron absorption reactions.
The first term requires considering the kinematics of the neutron –
atom scattering reaction. The second one is on a first step given by the
energy taken by He and Li but has to take into account the energy of the
impinging neutron. In both cases, we consider for simplicity the neu-
tron – nuclei reactions cross sections are isotropic.

3.1. Neutron – atom scattering reaction

From classical results of kinematics, a pka of mass M elastically hit

by a neutron of mass m and kinetic energy E will get a kinetic energy E '
which probability density function f(E’,E) is uniform from 0 to a max-
imum energy given by:
Fig. 4. Primary spectrum for He, showing the contribution of the neutron
′ = γE
Emax scattering (elastic) and absorption (for a He content = 1022/cm3).

4Mm
with γ = (M + m)2 2
⎛ ' ⎞
Integrating the contribution of all the neutrons through the whole E0 ⎜ E + 1⎟
'
⎝ E0 ⎠
spectrum ϕ with σelast the elastic cross section leads to the number of pka dN ' ϕ (E ) σabs (E ) dE
per second and pka energy unit given by: dE '
= ∫ (at−1. s−1. eV −1)
2 4E0' E
⎛ ' ⎞ E0
E0 ⎜ E − 1⎟
dN ′ 1 dE −1 −1 '
⎝ E0 (3)
= ∫E ⩾ E′/ γ ϕ (E ) σelast (E ) (at . s . eV−1) ⎠
dE′ γ E (1)
Similar formulas are derived for Li.
Calculation are performed by a numerical method with a log scale
for the energies (lethargy units).
3.2. Energy spectra of produced He and Li
4. Results
The He pka kinetic energy E′ resulting from fission induced by a
neutron absorption of 10B does not only depend on the fission energy 4.1. Primary spectra
but also on the impacting neutron kinetic energy E . In the center of
mass referential, He is emitted with about 1.8 MeV and Li with 1.0 MeV. The primary spectra for the different isotopes are reported on Figs. 4
Those energies have to be evaluated to take into account the energy of and 5. They span from the eV range (depending on the threshold dis-
the impinging neutron and its space distribution. We assume that the placement energies) to the MeV range (depending on the atom mass
loss of mass due to the fission is negligible. Conservation laws of energy over neutron mass ratio). In the case of helium (lithium is similar),
and momentum lead to formulas depending both on the masses of the Fig. 4 clearly shows the two components: displacements by the neutrons
parent isotope 10B (M ) and the fission products 4He (mHe ) and 7Li (mLi ). (“elastic” component) and energy distribution of He produced by the
Let the two characteristic energies be: absorption reaction (“fission” component). The latter is proportional to
m
E0' = Efission m +Lim (He energy resulting from a neutron of null en- He concentration and is here calculated for a He concentration of 1022/
He Li
ergy, equal to the value in the center of mass referential) cm3. Taking into account the energy distribution of the absorbed neu-
mLi M + m
E0 = Efission m (neutron energy leading to a He staying at rest tron, He atoms (resp. Li) are emitted with an energy distribution
He m
after its creation) spanning over several energy groups of the neutron spectrum. On Fig. 5,
The kinetic energy E ' of the created He may be expressed in the the curves are limited on the low energy side by the threshold dis-
following form: placement energies. The steep decrease of the curves on the high energy
side justifies not having taken into account the 10B(n,2α)T reaction.
E E Integrating the primary spectra curves shows that most of the pkas have
E ' = E0' ⎡1 + +2 cos (2πX ) ⎤
⎢ E E ⎥ energy above 10 keV (Fig. 6).
⎣ 0 0 ⎦
This calculation can then be used to estimate the pka formation
where is a uniform random number between −1 and +1 (due to the rates (Table 1).
isotropic recoil direction in the mass center coordinates), which leads to
a probability density function of the pka energy E ' given by the formula: 4.2. Displacements induced by pka’s and created atoms

1 After calculating the primary spectra, it is necessary to estimate the

f (E ', E ) =
4E0'. E
number of displaced atoms of each species as a function of the pka’s
E0 (2)
energy. For that, we used the SRIM program in the “full cascade” mode
2 2 [6]. This allows an estimation of the number of the displaced atoms by a
where E ' ranges between E0' ⎛ E −1⎞ and E0' ⎛ E + 1⎞ given incident ion, including the sub-cascades, and of the way energy is
⎝ E0 ⎠ ⎝ E0 ⎠
For a whole neutron spectrum ϕ and an absorption cross section σabs dissipated. Calculations are performed with 5000 ions in each case. The
this leads to a number of He created per second and He energy unit material is simulated with the composition we detailed above, then
given by with an equivalent boron atom with a 10.6 atom mass. For the sake of

68
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

Fig. 5. Primary spectra for all the isotopes in irradiated B4C, for an initial 10B enrichment = 48% and a 1022/cm3 burnup. a: per isotope; 12
C and 11
B show very
similar profiles. b: reported to an equivalent B4C atom with 1022 (n,α)/cm3.

those atoms correspond to ones emitted in a backward direction. Their

energy can however be high enough to produce defects in a bulk ma-
terial. To take into account this contribution, we have extrapolated the
‘range’ profiles as calculated by SRIM towards the negative paths, as-
suming Gaussian-like profiles (Fig. 9). The displacements rates have
then been increased in the surface of the Gaussian over surface of the
range curve ratio.
The results we obtained for each impinging atom are reported on
Fig. 10. Here again a high mobility is observed on He, due to its low
atom mass. Polynomial fittings of the results have then been performed.
Those polynomials are then convoluted with the primary spectra, this
leading to the total number of displaced atoms: Table 2.
As mentioned in Section 2, the helium production rate is about
4.01.10−9 s−1 per equivalent B4C atom. From those results, we deduce
the number of displaced B and C atoms per produced helium (i.e.
(973 + 252)/4.01) is about 305, and 280 when neglecting the damage
induced by the slowing down of He and Li produced in the (n,α) re-
actions. This second contribution is then less than 10% (it is worth
noting that this contribution would become the main one in a thermal
neutron flux). This is far from an estimation in the Kinchin-Pease or
Fig. 6. Integral weighed primary spectra for B4C (1022/cm3 burnup) in a
Phenix-like neutron spectrum. Most of the pka’s have energy above 10 keV. The
NRT (Norgett-Robinson-Torrens) approximations [16,17] which lead to
steps at about 1 MeV on the He and Li curves correspond to the (n,α) con- a displaced atoms over produced helium ratio of several thousands (at
tribution. first order, the pka yield is evaluated by the pka energy divided by
twice the threshold displacement energy). Again, this is due to the fact
that most of the energy is dissipated through electronic processes.
Table 1
Regarding the electronic energy losses, their intensity has to be
pka formation rate for all the isotopes in irradiated boron carbide for an
examined. As a matter, depending on this intensity, this contribution
equivalent B4C atom (10B enrichment = 39%, He = Li = 1022.cm−3).
can either produce damage (e.g., tracks in insulators), heal the material
Pka.s−1 (x109) 12
C 11
B 10
B 7
Li 4
He [18] or have no noteworthy effect. Here, the electronic stopping power
is at maximum 160 eV.Å−1 for C, 125 eV.Å−1 for B, 75 eV.Å−1 for Li
2.34 4.90 2.83 0.79 0.82
and 45 eV.Å−1 for He for the highest primary atom energy. This can
certainly not lead to damage (the usual threshold value for producing
simplicity, those calculations have been performed on a few pka en- damage is about or higher than 1 keV.Å−1) but could possibly allow a
ergies evenly distributed along a pka energy logarithmic range from low healing of some defects. This could be studied by tuning the Se over
100 eV to 4 MeV. Sn ratio at constant Sn value: this could be achieved by performing dual
On Fig. 7 we reported the results of the SRIM calculation in the case beam experiments (cf. infra).
of a 12C ion with a 1 MeV initial energy. As emphasized above, it ap-
pears most of the energy is dissipated via electronic interactions: Fig. 8. 4.3. Application
Atoms displacements occur mainly at the end of the C path or for a C
primary atom energy lower than 10 keV. Due to the mass ratio and We have then evaluated the damage produced by different ions of
threshold displacement energies, the relative displacement yield of different energy (see [19] for the choice of some ions) to estimate the
helium appears much higher than the B or C ones. possibility of reproducing the damage evaluated in the previous section
In the case of low-energy pka’s, SRIM leads to a non-negligible (Table 3.).
number of back-scattered atoms. In the case of a reactor irradiation, From the previous estimations, it appears that reproducing the

69
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

12
Fig. 7. SRIM calculation for a C atom with initial energy 1 MeV in B4C with 1022/cm3 He and Li. a: number of displaced atoms along the ion range. b: electronic
losses.

Fig. 9. Range of 100 eV 4He in B4C (with 1022/cm3 burnup). The back-scattered
Fig. 8. Partition of the energy losses for C atoms in ‘B4C’ as a function of C atoms yield is estimated by extrapolating a Gaussian profile fitted on the helium
initial energy. Electr.: electronic losses. Ball.: ballistic losses. distribution into the material.

damage arising in a fast neutron reactor with ion irradiation in boron atoms in the reactor, i.e. B and C with energy up to a few MeV.
carbide requires a high ballistic component together with low elec- Preliminary tests have been performed [20,21] aiming at reprodu-
tronic losses. This can be achieved by using slow heavy ions. But in this cing the helium to damage ratio together with a significant helium
case, the ion range is very short and the damage zone is strongly su- concentration. We have irradiated a high density boron carbide sample
perimposed to the implantation one, which is not desirable. This is of with a dual beam on the Jannus-Saclay facility [22]:
course no longer acceptable if helium has to be implanted in the da-
mage zone. An acceptable compromise is to use heavy ions with in- • 10 16
/cm2, 500 keV He (peak at 1.2 µm depth, FWHM ∼ 0.1 µm)
termediate energy (a few MeV): in that case, the ballistic damage can be • 10 15
/cm2, 10 MeV Au (peak at 2.2 µm, FWHM ∼ 0.25 µm).
made realistic and the electronic slowing down is a bit too high but still
far from the range where specific damage such as ion tracks appear The implantations have been performed at 500 °C, close to the
[19]. In the case of dual-beam experiments, it is worth noting that we temperature range boron carbide undergoes in a sodium fast neutron
are interesting in creating a realistic damage in the zone where helium reactor. The flux ratio has been chosen to be close the He to dpa ratio
is implanted. We then only consider the atom displacements in the we estimated here above, taking into account the accelerators re-
damage region of the slowing-down of the heavy ions, not in the quirements. The maximum He concentration is about 7.1020/cm3. The
stopping region. As a consequence, the structure of the damage, i.e. the total damage (Au + He) in the He implanted zone is about 1.2 dpa, the
space distribution of the displaced atoms does not include all the cas- damage to He ratio is then about 200. After annealing at 1100 °C for
cades. This is an important point to consider since it is known that the one hour to allow the formation of helium clusters, transmission elec-
largest cascades are produced at the end of the heavy ion course: in the tron microscopy has been performed on samples prepared with FIB. The
damage region, most of displacements arise from small cascades which single beam implanted sample show a low density of parallel platelets
structure is rather close to those created by the stopping of the matrix in the implanted zone. The dual-beam implanted sample show a high

70
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

Fig. 10. Number of displaced atoms in B4C (with burnup 1022.cm−3) per impinging atom (C, B, Li, He) as a function of the impinging atom energy as evaluated by
SRIM. Dots: calculated values. Lines: polynomial fitting.

Table 2 Table 3
Atom displacements rates (s−1 × 109) for an equivalent ‘B4C’ atom irradiated in Main characteristics of some ion irradiations of B4C boron carbide. Collision
Phenix reactor, for a 1022.cm−3 burnup. With/without (n,α): taking into ac- (density of displaced atoms), Se (electronic stopping power) and Sn (ballistic
count or neglecting the contribution coming from the slowing-down of He and stopping power) are given at mid-range.
Li produced in the (n,α) reactions.
ion Energy Range Straggling Collision Se Sn
B C He Li (MeV) (µm) (µm) (at./Å) (eV/Å) (eV/Å)
(1022.cm−3) (1022.cm−3)
Au 1 0.188 0.023 2.5 440 350
with (n,α) 973 252 365 150 4 0.650 0.065 1.5 550 250
without (n,α) 890 233 333 136 10 1.67 0.16 0.8 550 160
Bi 0.8 0.155 0.017 2.8 400 300
Ar 0.8 0.509 0.070 0.22 140 27
C 0.6 0.735 0.063 0.025 80 2
density of small bubbles (Fig. 11). Note they also both show bubbles in
Fe 10 3.14 0.18 0.08 280 10
the grain boundaries. This clearly evidences the role of damage in the S 100 21.5 0.15 0.0015 500 0.15
formation of the helium bubbles, this explaining the high density of
bubbles observed after irradiation in reactors [23]. As mentioned
above, helium mobility is much higher than the B and C ones, which helium production in boron carbide irradiated in a fast neutron spec-
stimulates the nucleation and growth of the clusters. trum. A value about 305 atom displacements per produced helium is
obtained. Most of the energy of the displaced particles is dissipated via
electronic processes, with stopping power Se lower than or around
5. Conclusion 150 eV/Å. Such results are to be used in order to simulate the behavior
of helium produced in boron carbide. Dual-beam ion implantations can
We have estimated the ratio of the number of displaced atoms to

71
D. Gosset et al. Nuclear Inst. and Methods in Physics Research B 434 (2018) 66–72

References

[1] G.S. Was, Z. Jiao, E. Getto, K. Sun, A.M. Monterrosa, S.A. Maloy, O. Anderoglu,
B.H. Sencer, M. Hackett, Emulation of reactor irradiation damage using ion beams,
Scr. Mater. 88 (2014) 33–36.
[2] S.J. Zinkle, S. Kojima, Helium-assisted cavity formation in ion-irradiated ceramics,
J. Nucl. Mater. 179–181 (1991) 395–398.
[3] W. Wesch, E. Wendler (Eds.), Ion Beam Modification of Solids, Springer, 2016.
[4] K. Yasuda, C. Kinoshita, S. Matsumura, Effects of simultaneous displacive and io-
nizing radiation in ionic and covalent crystals, Defect Diffus. Forum 206–207
(2002) 53–74.
[5] D. Gosset, Absorber materials for Gen-IV reactors, in: P. Yvon (Ed.), Structural
Materials for Generation IV Nuclear Reactors, Elsevier, 2017Ch. 15.
[6] J.F Ziegler, J.P. Biersack, M.D. Ziegler, SRIM, The Stopping and Range of Ions in
Matter, http://www.srim.org/.
[7] R. E. Stoller, K. Nordlund, S.P. Simakov, Summary Report of the Technical Meeting
on Primary Radiation Damage: from nuclear reaction to point defects, IAEA reports,
INDC(NDS)-0624, 2012.
[8] L. Lunéville, D. Simeone, C. Jouanne, Calculation of radiation damage induced by
neutrons in compound materials, J. Nucl. Mater. 353 (2006) 89–100.
[9] R.E. Stoller, M.B. Toloczko, G.S. Was, A.G. Certain, S. Dwaraknath, F.A. Garner, On
the use of SRIM for computing radiation damage exposure, Nucl. Instrum. Methods
B 310 (2013) 75–80.
[10] P. Filliatre, C. Jammes, B. Geslot, L. Buiron, In vessel neutron instrumentation for
sodium-cooled fast reactors: type, lifetime and location, Ann. Nucl. Energy. 37
(2010) 1435–1442.
[11] Evaluated Nuclear Data File (ENDF) Database, Version of 2017-06-01, https://
www-nds.iaea.org/exfor/endf.htm.
[12] T.A. Ivanova, et al., Cross-section of the Production of Tritium in Interactions of
Neutrons with 10B Nuclei, Phys. Part. Nucl. Lett. 10–4 (2013) 353–356.
Fig. 11. Helium bubbles in irradiated boron carbide, from [20,21]. a: single He [13] L. Zuppiroli, R. Kormann, D. Lesueur, Etude fondamentale du carbure de bore in-
dustriel, CEA report CEA-R-5237, 1983.
beam. b: dual He-Au beam. Bubbles also form in grain boundaries (orange lines:
[14] V. Motte, D. Gosset, T. Sauvage, H. Lecoq, N. Moncoffre, Evaluation of the diffusion
delimitation of the He implanted zone). (For interpretation of the references to coefficient of helium in implanted boron carbide by a NRA method, to be published.
colour in this figure legend, the reader is referred to the web version of this [15] H. Tsuchihir, T. Oda, S. Tanaka, Molecular-dynamics simulation of threshold dis-
article.) placement energies in lithium aluminate, Nucl. Instrum. Methods -B 269–14 (2011)
1707–1711.
[16] H. Kinchin, R.S. Pease, The displacement of atoms in solids by radiation, Rep. Progr.
then be performed with He and a second ion, the role of which is Phys. 18 (1955) 1–51.
creating a significant and representative damage. The calculation we [17] M.J. Norgett, M.T. Robinson, I.M. Torrens, A proposed method of calculating dis-
placement dose rates, Nucl. Eng. Des. 33 (1975) 50–54.
performed here allowed choosing the right parameters of the irradia- [18] L. Thomé, et al., Radiation effects in nuclear materials: role of nuclear and elec-
tion, second ion nature, energies, relative fluence, allowing a realistic tronic energy losses and their synergy, Nucl. Instrum. Methods B 307 (2013) 43–48.
simulation of the damage to helium ratio. Preliminary TEM observa- [19] G. Victor, Y. Pipon, N. Bérer, N. Toulhoat, N. Moncoffre, N. Djourelov, S. Miro,
J. Baillet, N. Pradeilles, O. Rapaud, A. Maître, D. Gosset, Structural modifications
tions of a dual-beam irradiated sample show helium clusters distribu- induced by ion irradiation and temperature in boron carbide B4C, Nucl. Instrum.
tion is similar to those observed after fast neutron irradiation, quite Methods-B 365 (2015) 30–34.
different from those resulting from a single He implantation. [20] V. Motte, Helium behaviour in implanted boron carbide, thesis University Lyon-1
(France), 08-11-2017.
[21] V. Motte, D. Gosset, N. Moncoffre, G. Gutierrez, Helium cluster nucleation and
growth in implanted B4C boron carbide, to be published.
Acknowledgements [22] S. Pellegrino, P. Trocellier, S. Miro, Y. Serruys, É. Bordas, H. Martin, N. Chaâbane,
S. Vaubaillon, J.P. Gallien, L. Beck, The JANNUS Saclay facility: a new platform for
This study was supported by the CEA-CNRS NEEDS-MATERIAUX materials irradiation, implantation and ion beam analysis, Nucl. Instrum. Methods-
B 273 (2012) 213–217.
collaborative research group. We are very indebted to the JANNUS- [23] T. Stoto, N. Housseau, L. Zuppiroli, B. Kryger, Swelling and microcracking of boron
Saclay team for performing the dual-beam implantations. carbide subjected to fast neutron irradiations, J. Appl. Phys. 68–7 (1990) 3198.

72