Stochastic simulation of destruction processes in self-irradiated materials
Self-irradiation damages resulting from fission processes are common phenomena observed in nuclear fuel containing (NFC) materials. Numerous α-decays lead to local structure transformations in NFC materials. The damages appearing due to the impacts of heavy nuclear recoils in the subsurface layer...
Збережено в:
| Дата: | 2017 |
|---|---|
| Автори: | , , , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут фізики конденсованих систем НАН України
2017
|
| Назва видання: | Condensed Matter Physics |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/157112 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Stochastic simulation of destruction processes in self-irradiated materials / T. Patsahan, A. Taleb, J. Stafiej, M. Holovko, J.P. Badiali // Condensed Matter Physics. — 2017. — Т. 20, № 3. — С. 33003: 1–11. — Бібліогр.: 46 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-157112 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1571122019-06-20T01:28:49Z Stochastic simulation of destruction processes in self-irradiated materials Patsahan, T. Taleb, A. Stafiej, J. Holovko, M. Badiali, J.P. Self-irradiation damages resulting from fission processes are common phenomena observed in nuclear fuel containing (NFC) materials. Numerous α-decays lead to local structure transformations in NFC materials. The damages appearing due to the impacts of heavy nuclear recoils in the subsurface layer can cause detachments of material particles. Such a behaviour is similar to sputtering processes observed during a bombardment of the material surface by a flux of energetic particles. However, in the NFC material, the impacts are initiated from the bulk. In this work we propose a two-dimensional mesoscopic model to perform a stochastic simulation of the destruction processes occurring in a subsurface region of NFC material. We describe the erosion of the material surface, the evolution of its roughness and predict the detachment of the material particles. Size distributions of the emitted particles are obtained in this study. The simulation results of the model are in a qualitative agreement with the size histogram of particles produced from the material containing lava-like fuel formed during the Chernobyl nuclear power plant disaster. Самоопромiнення, що виникає внаслiдок процесiв подiлу радiоактивних елементiв, є звичним явищем, яке спостерiгається в радiоактивних паливовмiсних матерiалах (ПВМ). Численнi α-розпади призводять до перетворення локальної структури ПВМ. Ушкодження, якi виникають за рахунок ударiв вiддачi осколкiв подiлу важких ядер в приповерхневому шарi, можуть спричинювати вiдокремлення частинок матерiалу. Така поведiнка є подiбною до процесу розпилення, яке спостерiгається пiд час бомбардування поверхнi матерiалу потоком енергетичних частинок. Проте, в ПВМ удари iнiцiюються в об’ємi матерiалу. В цiй роботi ми пропонуємо двовимiрну мезоскопiчну модель, з метою здiйснення стохастичного комп’ютерного моделювання процесiв руйнування, що виникають в приповерхневiй областi ПВМ. Ми описуємо ерозiю поверхнi матерiалу, змiну її шорсткостi з часом i передбачаємо вiдокремлення частинок матерiалу. У цьому дослiдженнi отримано розподiли розмiрiв вiдокремлених частинок. Результати комп’ютерного моделювання якiсно узгоджуються iз гiстограмами розмiрiв розпилених частинок спостережених в лавоподiбних паливовмiсних масах, якi сформувалися пiд час Чорнобильської катастрофи. 2017 Article Stochastic simulation of destruction processes in self-irradiated materials / T. Patsahan, A. Taleb, J. Stafiej, M. Holovko, J.P. Badiali // Condensed Matter Physics. — 2017. — Т. 20, № 3. — С. 33003: 1–11. — Бібліогр.: 46 назв. — англ. 1607-324X PACS: 02.50.-r, 23.70.+j, 05.50.+q, 46.50.+a, 46.65.+g DOI:10.5488/CMP.20.33003 arXiv:1712.01094 http://dspace.nbuv.gov.ua/handle/123456789/157112 en Condensed Matter Physics Інститут фізики конденсованих систем НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
Self-irradiation damages resulting from fission processes are common phenomena observed in nuclear fuel
containing (NFC) materials. Numerous α-decays lead to local structure transformations in NFC materials. The
damages appearing due to the impacts of heavy nuclear recoils in the subsurface layer can cause detachments
of material particles. Such a behaviour is similar to sputtering processes observed during a bombardment of the
material surface by a flux of energetic particles. However, in the NFC material, the impacts are initiated from the
bulk. In this work we propose a two-dimensional mesoscopic model to perform a stochastic simulation of the
destruction processes occurring in a subsurface region of NFC material. We describe the erosion of the material
surface, the evolution of its roughness and predict the detachment of the material particles. Size distributions
of the emitted particles are obtained in this study. The simulation results of the model are in a qualitative
agreement with the size histogram of particles produced from the material containing lava-like fuel formed
during the Chernobyl nuclear power plant disaster. |
| format |
Article |
| author |
Patsahan, T. Taleb, A. Stafiej, J. Holovko, M. Badiali, J.P. |
| spellingShingle |
Patsahan, T. Taleb, A. Stafiej, J. Holovko, M. Badiali, J.P. Stochastic simulation of destruction processes in self-irradiated materials Condensed Matter Physics |
| author_facet |
Patsahan, T. Taleb, A. Stafiej, J. Holovko, M. Badiali, J.P. |
| author_sort |
Patsahan, T. |
| title |
Stochastic simulation of destruction processes in self-irradiated materials |
| title_short |
Stochastic simulation of destruction processes in self-irradiated materials |
| title_full |
Stochastic simulation of destruction processes in self-irradiated materials |
| title_fullStr |
Stochastic simulation of destruction processes in self-irradiated materials |
| title_full_unstemmed |
Stochastic simulation of destruction processes in self-irradiated materials |
| title_sort |
stochastic simulation of destruction processes in self-irradiated materials |
| publisher |
Інститут фізики конденсованих систем НАН України |
| publishDate |
2017 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/157112 |
| citation_txt |
Stochastic simulation of destruction processes in self-irradiated materials / T. Patsahan, A. Taleb, J. Stafiej, M. Holovko, J.P. Badiali // Condensed Matter Physics. — 2017. — Т. 20, № 3. — С. 33003: 1–11. — Бібліогр.: 46 назв. — англ. |
| series |
Condensed Matter Physics |
| work_keys_str_mv |
AT patsahant stochasticsimulationofdestructionprocessesinselfirradiatedmaterials AT taleba stochasticsimulationofdestructionprocessesinselfirradiatedmaterials AT stafiejj stochasticsimulationofdestructionprocessesinselfirradiatedmaterials AT holovkom stochasticsimulationofdestructionprocessesinselfirradiatedmaterials AT badialijp stochasticsimulationofdestructionprocessesinselfirradiatedmaterials |
| first_indexed |
2025-07-14T09:26:06Z |
| last_indexed |
2025-07-14T09:26:06Z |
| _version_ |
1837613886692392960 |
| fulltext |
Condensed Matter Physics, 2017, Vol. 20, No 3, 33003: 1–11
DOI: 10.5488/CMP.20.33003
http://www.icmp.lviv.ua/journal
Stochastic simulation of destruction processes in
self-irradiated materials∗
T. Patsahan1, A. Taleb2,3, J. Stafiej4, M. Holovko1, J.P. Badiali1,3
1 Institute for Condensed Matter Physics of the National Academy of Sciences of Ukraine,
1 Svientsitskii St., 79011 Lviv, Ukraine
2 PSL Research University, Chimie ParisTech— CNRS, Institut de Recherche de Chimie Paris, Paris 75005, France
3 Université Pierre et Marie Curie, Paris, 75231, France
4 Cardinal Stefan Wyszyński University, Department of Mathematics and Natural Sciences, Warsaw, Poland
Received June 19, 2017, in final form August 6, 2017
Self-irradiation damages resulting from fission processes are common phenomena observed in nuclear fuel
containing (NFC) materials. Numerous α-decays lead to local structure transformations in NFC materials. The
damages appearing due to the impacts of heavy nuclear recoils in the subsurface layer can cause detachments
of material particles. Such a behaviour is similar to sputtering processes observed during a bombardment of the
material surface by a flux of energetic particles. However, in the NFC material, the impacts are initiated from the
bulk. In this work we propose a two-dimensional mesoscopic model to perform a stochastic simulation of the
destruction processes occurring in a subsurface region of NFC material. We describe the erosion of the material
surface, the evolution of its roughness and predict the detachment of the material particles. Size distributions
of the emitted particles are obtained in this study. The simulation results of the model are in a qualitative
agreement with the size histogram of particles produced from the material containing lava-like fuel formed
during the Chernobyl nuclear power plant disaster.
Key words: nuclear fuel containing material, self-irradiation damage, destruction, roughness, stochastic
computer simulation, sputtering
PACS: 02.50.-r, 23.70.+j, 05.50.+q, 46.50.+a, 46.65.+g
1. Introduction
A study of nuclear fuel containing (NFC) materials and problems related to their storage are of
a special interest due to the great importance with respect to the safety of environment. The nuclear
waste materials encapsulating actinides (e.g., U, Pu) undergo self-irradiation during long periods of time.
Therefore, their chemical and physical properties change crucially. Permanent α-decay events occurring
in a material lead to transformations of material structure around the points where these events appear,
producing the so-called thermal spikes [1–4]. Such local structural changes damage a material and can
lead to its destruction if the relaxation time related to healing processes is too slow. In homogeneous NFC
materials where nuclear fuel is distributed randomly, the damages also appear randomly in the material
bulk producing local defects [3]. On the other hand, if these defects occur deep in the bulk they can
be healed with time [3, 4], while the defects produced near the material surface can erode this surface
which leads to a detachment of the particles formed by material pieces. As a consequence, a highly
radioactive surface may spontaneously disseminate and disperse a solid state phase into the environment.
In connection with these phenomena, long term predictions are required. However, predictions cannot
result from simplified extrapolations of the experimental data measured in limited time scales, which are
∗We dedicate this article to the memory of late Dr. Jean-Pierre Badiali who inspired our research and made a significant
contribution to this work.
This work is licensed under a Creative Commons Attribution 4.0 International License . Further distribution
of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI.
33003-1
https://doi.org/10.5488/CMP.20.33003
http://www.icmp.lviv.ua/journal
http://creativecommons.org/licenses/by/4.0/
T. Patsahan et al.
much shorter than the storage period of the nuclear waste. Therefore, theoretical description appears to
be an unavoidable complement to experimental data, and computer simulations become helpful in this
case.
There is a large literature describing the damage processes inside nuclear waste materials [5–25]. Self-
destruction mechanisms of pure nuclear fuel are also studied. Usually, they try to predict the conditions
under which a material bulk becomes unstable (fragile). In most theoretical studies, an atomistic level
of modelling is used. For instance, in [8] a method of molecular dynamics is used to simulate the
damage processes in zirconolite structures containing radioactive Pu. Unfortunately, such studies cannot
be performed on large time scales as well as one cannot make any predictions for the system where many
nuclear decays occur, since the microscopic level description imposes a restriction on a space scale as
well. Therefore, it is necessary to develop a model allowing us to consider the system at a mesoscopic
level.
In the present study, we propose a simple mesoscopic model of surface destruction processes in NFC
materials, which is implemented on two-dimensional lattice with the use of a stochastic algorithm similar
to the cellular automata concept [26, 27]. It is assumed that the detachment of nanoparticles at the NFC
material surface results from the formation of damaged regions and microfracturing in the vicinity of the
locationswhere nuclear decays take place. It is also assumed that the heavy nuclear recoils due toα-decays
are responsible for the formation of the damaged regions inside the NFC material. The technique used
in the study allows us to run simulations at a mesoscopic level and to describe the surface evolution in
large time scales. Thus, the characteristics such as chemical roughness, the mean-square front width and
the yield of a material can be easily obtained from simulations. Our main goal is to get histograms of the
particles formed during the surface destruction of NFC material. We compare the obtained histograms
with the experimental results obtained for the NFC material formed in the well-known heavy nuclear
accident at the Chernobyl nuclear power plant [28].
This paper is organized as follows. In section 2, we discuss the physical aspects and review some
literature related to the damage and destruction processes in NFCmaterials. A stochastic model developed
in our study is described in section 3. In section 4, we present a discussion of the obtained results. Finally,
we conclude the paper in the last section.
2. Physical aspects
To date, many studies can be found in literature which are devoted to the study of the damage and
destruction processes of the solid material surface due to external factors such as a high energy particle
bombardment, while there is a lack of theoretical studies of the same processes for the NFC materials
where the internal factors cause a surface destruction. However, one can notice similarities between the
processes occurring near the surface in the self-irradiated material and the sputtering processes due to a
surface bombardment with a flux of energetic particles. In the case of collisions between the beam and
the target, one can observe a formation of cascades of atom displacements, and as a consequence, the
creation of new structures and defects inside the material, which lead to a surface reconstruction and
particles emission. For energetic incident beams, the sputtered particles can be large clusters of atoms.
These clusters are formed in the near-surface region due to a multiple breaking of bonds [29, 30]. The
magnitude of these effects specifically depends on the nature of the particles forming the beam and on
their energy as well as on the nature of a target material. When a target and a beam contain the same
chemical species, this process is called a self-sputtering. The self-sputtering is also widely studied for
materials bombarded with projectiles of different energies [31–33] and sizes [34, 35].
In the case of sputtering phenomena as well as in the case of self-irradiated materials, we deal
with energy dissipation, which leads to the formation of atomic displacement cascades [4, 30]. The
main difference is that in the first case, the cascades are initiated at the surface, and the atoms or
atomic clusters escape from the surface immediately. In the second case, the cascades start in the bulk.
Due to α-decays occurring in NFC material and a heavy nuclear recoil appears. As a consequence,
an energy in the range of 40−100 keV disperses causing a local structure transformation. As a result
of elastic collisions between atoms and heavy nuclei, the cascades of atomic displacements around α-
decay locations must occur[2, 4, 7, 36, 37]. The formation of atomic displacement cascades produced
33003-2
Destruction processes in self-irradiated materials
by an elastic collision with some energetic particles was studied by molecular dynamics [7, 8, 38] and
also observed from experiment [39]. The sizes of atomic collision cascades depending on the value of
the particle energy estimated in these investigations are in the range of 25–30 nm. The same result was
obtained by Baryakhtar et al. [28] for lava-like fuel containingmaterials (LFCM) formed in the Chernobyl
heavy nuclear accident [40, 41]. A microscopic structure of a damaged region is disordered. At the same
time, it is known that such disordered regions cause local tensions, which are responsible for the defects
and microfractures formation in NFC materials [3–5, 18] and they can crucially reflect on the material
durability [4, 36]. However, the fractures produced far from the material surface may not play an essential
role in the material destruction due to two factors: i) because they are confined in the stable material
parts; ii) they can be healed in a quite short time due to relaxation processes [42]. On the other hand,
when fractures appear close to the surface, their boundaries can reach a surface and detach parts of the
material. A size of a detached particle depends on geometrical sizes of the fractures which are defined
by mechanical properties of the material considered as well as by the collision energy.
3. Modelling
3.1. General concept
We propose a simple model sufficient to describe the cluster formation in terms of two parameters
model. First we assume that we have a lattice with the sites from which nuclear events can be initiated.
When one of the events occurs at some given site, the damage randomly expands producing defects in the
vicinity of this site. For instance, a formation of a cascade of atom displacements (damaged region) takes
place. In this case, one of the model parameters should relate to an area of a contact surface between the
damaged region and the rest of the material forming a fracture. Another parameter corresponds to the
frequency of nuclear events per volume of the NFCmaterial, which directly depends on the concentration
of nuclear fuel. A fracture can lead to the detachment of a piece of the material (particle detachment)
when both two conditions are valid: i) boundaries of the fracture totally belong to the surface; ii) the
fracture is open toward the surface, e.g., it should have a hemisphere-like or a cap-like shape with a basis
directed toward the surface. The sequential particle detachments lead to the formation of a specific front
surface which is characterized by its roughness.
To facilitate our simulations, we consider a two-dimensional model of the system presented as its
projection. Therefore, the NFC material is described by its profile in XY coordinates, where the surface
destruction process proceeds along Y -axis. In this case, the area of a fracture is reduced to the length of
its projection, and for a particle detachment it should be of an arc-like shape. Moreover, the front surface
formed during the destruction process in two dimensions reduces to a line, and hereafter we refer to it as
a front line (FL).
3.2. Stochastic model
A two-dimensional model is presented as a square lattice with 4 or von Neumann connectivity [43].
The simulation box is of the size of Nx × Ny and the periodical boundary conditions are applied along
X-axis. The sites of the lattice are of two types corresponding to two kinds of species. The first kind of
sites represents the regions of pristine solid material and the second type represents the regions containing
nuclear fuel inclusions, which correspond to impact centers (ICs) where the damage processes can start
from. We reduce the number of characteristic lengths of the system to only one — the lattice constant,
a, and it is subsequently our unit of length. The IC sites are located randomly on the lattice. We denote
their concentration or number density as fd. Each of the IC sites is chosen randomly to build paths of a
fracture initiated from this IC (OA and OB in figure 1). Within the model proposed, the fracture is built
by two different paths, which evolve in opposite directions. Such a fracture formation can potentially lead
to a detachment of a piece of material surface. In this case, two possibilities to start the fracture creation
are considered: OA is directed to the left from the chosen IC site while OB goes to the right, or OA goes
up while OB goes down. The paths of fracture propagation is modelled as a random walk. Due to the
assumption of a limited length of the fracture that can be formed in the material, the number of steps of
33003-3
T. Patsahan et al.
A B
OA’
A"
B’
B"
IC
Figure 1. (Color online) Example of fracture prop-
agation. Filled blue cells denote random walk paths
to build a fracture. The cental red cell (site “O”)
corresponds to the impact center (IC).
Figure 2. (Color online) Cluster detachment for
Nsrw = 6. Filled grey cells belong to the particle to
be detached. The dotted lines denote random walk
paths initiated from the IC site (big point) and the
thick red line denotes the corresponding fracture.
a random walk cannot exceed the specified parameter Nsrw, which is equal to a half the maximum length
of fractures. Another restriction introduced in the model requires that the distances between the ends of
the fracture and its initial point should increase during its propagation as shown in figure 1: OA′′ > OA′
and OB′′ > OB′. This condition prevents the self-crossing of the fracture. In this way, the ends of a
fracture propagate away one from another forming an arc-like fracture. It is quite reasonable to give some
preferable direction for the fracture propagation dependent on a deformation geometry around the IC
considered (especially when the maximum length of a fracture is rather small). If both ends of the fracture
reach the surface, then the constructed fracture leads to a detachment of the material cluster (figure 2).
Otherwise, the event is discarded, i.e., when the fracture paths stop somewhere in the bulk. In this case,
we forget about it supposing that the relaxation processes are fast enough for the fracture to be healed [42].
To obtain the results with a sufficient accuracy for the surface roughness for each pair of the parameters
Nsrw and fd, we perform an average over 30 independent simulation runs. Each simulation run takes
200000 steps. One simulation step corresponds to one particle detachment. The size of a lattice taken of
Nx = 1000 and Ny = 1000 is found to be sufficient to minimize the boundary condition effects and to
consider rather large FL widths. The events for the size histogram are collected in a stationary regime
of the material destruction after 50000 simulation steps. To identify the clusters of sites formed during
particle detachments the Hoshen-Kopelman algorithm is used [44].
4. Results and discussion
To describe the properties of the front line (FL) we introduce standard functions, which characterize
the growth processes [45]. First we consider the roughness of the FL as a function of the number of
simulation time steps Nsteps for different values of the parameters fd and Nsrw. The roughness is given by
the factor defined as a ratio of the number of sites Ns in the FL to the initial number Nx of sites in the
smooth surface Nx = 1000. After Nstep, the roughness factor is given by
r(Nsteps) =
Ns(Nsteps)
Nx
. (4.1)
We may notice that according to our algorithm, the Ns sites are only formed by the host material
(free of any nuclear fuel inclusions). Thus, there is no IC site on the surface, by construction. For the
initial FL there is no roughness, i.e., r(0) = 1. After some simulation steps, we observe an increase of
the roughness factor, due to the particle detachment process (figure 3). After 50000 simulation steps we
get a stationary regime characterized by a constant value r for all the parameters considered in our study.
In figure 4 we plot r as a function of fd and for two values of Nsrw. We also observe that r increases
when the density of ICs decreases and about fd = 0.80 the roughness becomes almost independent of
Nsrw. We expect that this weakening related to the average distance between ICs becomes less than the
33003-4
Destruction processes in self-irradiated materials
0 50000 100000 150000 200000
1
2
3
4
5
6
Nsrw = 3
r
Steps
0.19
0.20
0.30
0.50
0.80
fd =
0 50000 100000 150000 200000
1.2
1.6
2.0
2.4
2.8
3.2
Nsrw = 5
r
Steps
0.10
0.20
0.30
0.50
0.80
fd =
Figure 3. Variation of the ratio r versus a number of simulation steps obtained for Nsrw = 3 and Nsrw = 5,
and for the different density fd.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
1
2
3
4
5
6
N
srw
= 3
N
srw
= 5
r
fd
Figure 4. The dependence of roughness r on the fd for Nsrw = 3 (square symbols) and Nsrw = 5 (circle
symbols).
mean length of a fracture. This can be also explained by the fact that the mean size of detached particles
increases monotonously when we decrease fd (table 1). It is also worth noting that there exists a critical
value fc such that for fd < fc the destruction process stops. This is related to the distances between ICs,
which are too large in order to provide continuous detachments of particles, and at some step one quickly
faces a situation when Nsrw is not sufficient to build a path from any IC in the system to the surface. For
Nsrw = 3 we obtain fc about 0.19, and for Nsrw = 5, the critical density fc is found at lower values of
fd which are near 0.08. One can observe in figure 4 that in the vicinity of fc, the roughness r increases
drastically, and since the critical density fc is higher for smaller Nsrw, the roughness increases faster with
a decrease of fd. Therefore, we always obtain larger values of the roughness for smaller Nsrw. This looks
surprising because one would expect a larger roughness in the case corresponding to a larger particles
detachment. This is certainly true if a single particle detachment at a flat surface is considered. However,
we have a series of detachments leading to a formation of a complex rough surface. It is really difficult
to predict for which exactly Nsrw, the roughness is higher without doing simulation. Moreover, it should
be noted that Nsrw is just the maximum length of a fracture, and it means that actual lengths of fractures
causing particle detachments can be smaller than the given Nsrw. Therefore, one can expect a distribution
of lengths of such fractures. At the same time, the roughness results from the averaging along FL and
33003-5
T. Patsahan et al.
Table 1. The mean sizes of particles for the given parameters.
fd Nsrw = 3 Nsrw = 5
0.10 – 14.78
0.20 10.06 13.58
0.30 9.72 12.49
0.50 8.60 11.70
0.80 8.05 11.32
the real contribution of fractures of different sizes to the total surface formation is unclear. On the other
hand, the behaviour of the roughness at different Nsrw can be explained by that for a larger Nsrw where the
destruction process proceeds more intensively than for smaller Nsrw, and this leads to a better smoothing
of FL.
A constant value for the roughness factor r means that the length of the FL is constant but nevertheless
different shapes can be observed for the FL. The deformations at a constant length can be visualized
with a flexible cord rearranged without stretching or squeezing. Therefore, the roughness factor r does
not reflect all the geometrical features of the FL. Another parameter associated with the roughness is the
width of the FL defined as the root of mean-square deviation σ of the FL from its mean position [45]:
σ(Nsteps) =
{
1
Nx
Nx∑
i=1
[hi − h(Nsteps)]
2
}1/2
, (4.2)
where hi is the height of the pile of material sites in the i-th column from the bottom of the simulation
box up to the first environment site in this column. The mean position of the FL is defined as usual [45]:
h(Nsteps) =
1
Nx
Nx∑
i=1
hi . (4.3)
The dependence of σ(Nsteps) on the number of simulation steps is shown in figure 5. As expected, the
average width of the front profile increases when the density of ICs in the system decreases. Qualitatively,
the evolution of σ is similar to that of r , but the convergence to the stationary regime is noticeably slower.
The time evolution of the mean position of the FL calculated from the expression (4.3) gives an
estimation of degradation rate for the material. In our algorithm, one detachment of the particle occurs
0 50000 100000 150000 200000
0
20
40
60
80
100
Steps
Nsrw = 3
0.19
0.20
0.30
0.50
0.80
f
d
=
0 50000 100000 150000 200000
0
10
20
30
40
50
Steps
Nsrw = 5
0.10
0.20
0.30
0.50
0.80
f
d
=
Figure 5. The evolution of the front width as given by the root of mean-square deviation from the mean
position. The parameters are the same as in figure 3.
33003-6
Destruction processes in self-irradiated materials
per one simulation step. The simulation step can be related to real time if we know the so-called dust
productivity parameter which is known in some experiments. Unfortunately, we have no such parameter
in the case of the LFCM. Therefore, we limit the analysis of the FL to its dependence on the number of
simulation steps. We observe the existence of a stationary regime for the displacement of the FL. After a
large number of simulation steps, the mean position of FL is given by pNstep and the slopes p have been
estimated (table 2). It is observed that the lower is the value of fd the larger is the slope. Due to a single
particle detachment per a simulation step, the higher slopes indicate that the larger sizes of particles are
detached. As shown above, the lower is fd the higher is the roughness and the larger are the sizes of the
particles emitted. Therefore, the results presented in table 2 are consistent with that in table 1. Indeed,
they are in a very good agreement for the values of fd higher than 0.30 and Nsrw = 3 as well as for
Table 2. The slopes of the dependencies of the mean position of FL for the same parameters as in table 1.
fd
Nsrw = 3 Nsrw = 5
×10−3
×10−3
0.10 – 19.12
0.20 13.24 13.74
0.30 9.92 12.51
0.50 8.60 11.70
0.80 8.05 11.32
0 5 10 15 20 25 30 35 40
0.00
0.05
0.10
0.15
0.20
0.25
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 3
f
d
= 0.19
0 5 10 15 20 25 30 35 40
0.00
0.05
0.10
0.15
0.20
0.25
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 3
f
d
= 0.20
0 5 10 15 20 25 30 35 40
0.00
0.05
0.10
0.15
0.20
0.25
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 3
f
d
= 0.30
0 5 10 15 20 25 30 35 40
0.00
0.05
0.10
0.15
0.20
0.25
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 3
f
d
= 0.50
Figure 6. The distributions of the emitted particles for Nsrw = 3 and different densities fd.
33003-7
T. Patsahan et al.
fd higher than 0.20 and Nsrw = 5 (marked bold). Near the critical values of ICs density, fc, there is a
difference observed between these two cases.
The events for the size histogram are collected in a stationary regime of the material destruction
after 50000 simulation steps. From the simulations we can calculate the total number of the emitted
particles and estimate the relative frequency of a given particle size. By such procedure, the histogram is
normalized and from this histogram we may determine the mean size of the emitted particles (table 1).
In figures 6 and 7 we present the calculated histograms. One can see that the dispersion of the particle
sizes is not essentially affected by the IC density while the mean size of the detached particles depends
on fd as it is shown in table 1. At low fd, there are two sharp peaks on the histograms (figure 6).
When fd increases, the second peak is suppressed and the histogram becomes smoother. At a sufficiently
high density ( fd = 0.50), the second peak disappears completely. Simultaneously, the first peak shifts
to smaller sizes. The dispersion of the size distribution is strongly affected by the Nsrw parameter. In
figure 7, the histograms for the cases of Nsrw = 2 and Nsrw = 5 at fd = 0.50 are presented. For Nsrw = 2,
the dispersion of particle sizes is much smaller than for Nsrw = 5. Hence, for longer fractures we may
obtain larger particles.
The projected histogram calculated for fd = 0.19−0.3 and Nsrw = 3 (figure 6) is in a remarkable
qualitative agreement with that shown in figure 8 for the data of Baryakhtar et al. [28]. Herein above we
have given the relation between the effective radius, the mean number of sites and the lattice constant a. If
0 5 10 15 20 25 30 35 40
0.0
0.1
0.2
0.3
0.4
0.5
0.6
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 2
f
d
= 0.50
0 5 10 15 20 25 30 35 40
0.00
0.02
0.04
0.06
0.08
0.10
0.12
R
el
at
iv
e
fr
eq
ue
nc
y
Particle size
N
srw
= 5
f
d
= 0.50
Figure 7. The distributions of the emitted particles for Nsrw = 2 and Nsrw = 5 at the density fd=0.50.
Figure 8. The distribution of the emitted particles obtained from the experiment for thematerial containing
a lava-like nuclear fuel by Baryakhtar et al. [28].
33003-8
Destruction processes in self-irradiated materials
we fit the effective radius to the value Rmax associated with the maximum of the experimental histogram
and if for the number of sites we take its value nmax obtained from the calculated histogram, we can get
an estimation of the lattice constant as
a ≈ Rmax
√
π/nmax = 20.1 nm , (4.4)
where Rmax = D/2, while D = 60 nm is the diameter of the particles at the first maximum in figure 8,
and nmax = 7 is the corresponding size of the particles in terms of a number of sites obtained from the
simulation. The calculated value of a has the same order of magnitude as the damaged regions 25−30 nm
observed in [28] and related to a size of the elementary cluster that can be detached. The value of the
parameter Nsrw giving the maximum length of the fracture can be derived as l = La = (2Nsrw + 1)a ∼
141 nm for Nsrw = 3. The value of l is far lower than the critical one when the fracture starts to propagate
continuously [46]. Due to this, the immediate destruction of the material only appears in its subsurface
region or when fractures form an infinite network inside the material bulk. In the given study, we consider
the first case. As it is shown above, the surface roughness keeps constant during this process. For a more
thorough verification of the model, a comparison of the experimentally determined and simulated surface
roughness can be useful. Such results would allow us to adjust the model parameters so far undetermined
in the absence of the experimental data.
5. Conclusions
We propose a stochastic model to simulate the destruction processes in nuclear fuel containing (NFC)
materials. In this study, we focus on the surface erosion processes leading to the emission of material
particles. It is assumed that the erosion of the NFC material surface results from numerous α-decay
events in the subsurface region, which are accompanied with heavy nuclear recoils producing damages in
the material. The approach used in the study allows us to run simulations at the mesoscopic level and to
describe the surface evolution in large time scales. A quantitative description of this model is presented
as its projection on a two-dimensional lattice.
The model is defined by two parameters, one of which is related to the number density of the nuclear
fuel in the material, and another characterizes a maximum size of the defects (fractures) produced by
heavy nuclear recoils. Different pairs of these parameters have been checked and the corresponding
characteristics, such as surface roughness, mean-square width of the front and the yield of material are
calculated. These results were obtained in a stationary regime of the NFC material destruction process.
However, it should be noted that the stationary regime can be achieved only when the number density fd
and the maximum length of fractures 2Nsrw are large enough. For instance, for each value of the parameter
Nsrw, one can find a critical concentration fc below which the destruction process cannot proceed. On
the contrary, for fd higher than fc, we find at longer times a stationary regime. It is observed that for high
values of fd, the roughness of NFC material surface is rather small. However, a decrease of fd leads to
an increase of the roughness as well as of the mean-square width of the front and the mean particle size.
Moreover, it is noticed that the maximum values of these characteristics are reached if fd tends to fc.
The size histograms of the particles produced during the erosion of the material surface are built
during simulations. These distributions are characterized by different dispersions and mean values of the
particle sizes depending on fd and Nsrw chosen. It is shown that our theoretical predictions are in a good
agreement with the experimental histograms of particle sizes obtained due to destruction processes of
the material containing lava-like nuclear fuel formed during the Chernobyl catastrophe [28]. We have
also checked that the parameters of the model in this case are of a correct order of magnitude.
Acknowledgement
The simulations have been done on the computing cluster of the Institute for Condensed Matter
Physics (Lviv, Ukraine).
33003-9
T. Patsahan et al.
References
1. Matzke Hj., Radiat. Eff. Defects Solids, 1982, 64, 3, doi:10.1080/00337578208222984.
2. Weber W.J., J. Mater. Res., 1990, 5, 2687, doi:10.1557/JMR.1990.2687.
3. Weber W.J., Ewing R.C., Catlow C.R.A., Diaz de la Rubia T., Hobbs L.W., Kinoshita C., Matzke Hj.,
Motta A.T., Nastasi M., Salje E.K.H., Vance E.R., Zinkle S.J., J. Mater. Res., 1998, 13, No. 6, 1434–1484,
doi:10.1557/JMR.1998.0205.
4. Ewing R.C., Weber W.J., Clinard F.W. (Jr.), Prog. Nucl. Energy, 1995, 29, 63,
doi:10.1016/0149-1970(94)00016-Y.
5. Evron R., Kimmel G., Eyal Y., J. Nucl. Mater., 1994, 217, 54–66, doi:10.1016/0022-3115(94)90304-2.
6. Fayek M., Anovitz L.M., Cole D.R., Bostick D.A., Geochim. Cosmochim. Acta, 2011, 75, 3677–3686,
doi:10.1016/j.gca.2011.03.040.
7. Trachenko K., Dove M.T., Salje E.K.H., J. Phys.: Condens. Matter, 2001, 13, 1947,
doi:10.1088/0953-8984/13/9/317.
8. Trachenko K., Dove M.T., Artacho E., Todorov I.T., Smith W., Phys. Rev. B, 2006, 73, 174207,
doi:10.1103/PhysRevB.73.174207.
9. Trachenko K., Zarkadoula E., Todorov I.T., Dove M.T., Dunstan D.J., Nordlund K., Nucl. Instrum. Methods
Phys. Res., Sect. B, 2012, 277, 6–13, doi:10.1016/j.nimb.2011.12.058.
10. Bishop C.L., Grimes R.W., Parfitt D.C., Nucl. Instrum. Methods Phys. Res., Sect. B, 2010, 268, 2915–2917,
doi:10.1016/j.nimb.2010.05.007.
11. Van Brutzel L., Rarivomanantsoa M., Ghaleb D., J. Nucl. Mater., 2006, 354, 28–35,
doi:10.1016/j.jnucmat.2006.01.020.
12. Van Brutzel L., Vincent-Aublant E., Delaye J.-M., Nucl. Instrum. Methods Phys. Res., Sect. B, 2009, 267,
3013–3016, doi:10.1016/j.nimb.2009.06.078.
13. Martin G., Maillard S., Van Brutzel L., Garcia P., Dorado B., Valot C., J. Nucl. Mater., 2009, 385, 351–357,
doi:10.1016/j.jnucmat.2008.12.010.
14. Martin G., Garcia P., Van Brutzel L., Dorado B., Maillard S., Nucl. Instrum. Methods Phys. Res., Sect. B, 2011,
269, 1727–1730, doi:10.1016/j.nimb.2010.12.075.
15. Tian X.F., Gao T., Long C., Li J.K., Jiang G., Xiao H., Nucl. Instrum. Methods Phys. Res., Sect. B, 2011, 269,
1771–1776, doi:10.1016/j.nimb.2011.04.115.
16. Delaye J.M., Peuget S., Bureau G., Calas G., J. Non-Cryst. Solids, 2011, 357, 2763–2768,
doi:10.1016/j.jnoncrysol.2011.02.026.
17. Dewan L., Hobbs L.W., Delaye J.M., J. Non-Cryst. Solids, 2012, 358, 3427–3432,
doi:10.1016/j.jnoncrysol.2012.04.033.
18. Kieu L.H., Delaye J.M., Stolz C., J. Non-Cryst. Solids, 2012, 358, 3268–3279,
doi:10.1016/j.jnoncrysol.2012.07.037.
19. Devanathan R., Corrales L.R., Weber W.J., Chartier A., Meis C., Nucl. Instrum. Methods Phys. Res., Sect. B,
2005, 228, 299–303, doi:10.1016/j.nimb.2004.10.060.
20. Heinisch H.L., Weber W.J., Nucl. Instrum. Methods Phys. Res., Sect. B, 2005, 228, 293–298,
doi:10.1016/j.nimb.2004.10.059.
21. Chung B.W., Thompson S.R., Lema K.E., Hiromoto D.S., Ebbinghaus B.B., J. Nucl. Mater., 2009, 385, 91–94,
doi:10.1016/j.jnucmat.2008.09.031.
22. Baclet N., Oudot B., Grynszpan R., Jolly L., Ravat B., Faure P., Berlu L., Jomard G., J. Alloys Compd., 2007,
444–445, 305–309, doi:10.1016/j.jallcom.2006.10.156.
23. Dremov V.V., Sapozhnikov F.A., Samarin S.I., Modestov D.G., Chizhkova N.E., J. Alloys Compd., 2007,
444–445, 197–201, doi:10.1016/j.jallcom.2006.10.003.
24. Robinson M., Kenny S.D., Smith R., Storr M.T., McGee E., Nucl. Instrum. Methods Phys. Res., Sect. B, 2009,
267, 2967–2970, doi:10.1016/j.nimb.2009.06.113.
25. Geisler T., Trachenko K., Ríos S., Dove M.T., Salje E.K.H., J. Phys.: Condens. Matter, 2003, 15, L597–L605,
doi:10.1088/0953-8984/15/37/L07.
26. Taleb A., Stafiej J., Chaussé A., Messina R., Badiali J.P., J. Electroanal. Chem., 2001, 500, 554,
doi:10.1016/S0022-0728(00)00433-2.
27. Patsahan T., Taleb A., Stafiej J., Badiali J.P., Condens. Matter Phys., 2007, 10, 579–585,
doi:10.5488/CMP.10.4.579.
28. Baryakhtar V., Gonchar V., Zhidkov A., Zhydkov V., Condens. Matter Phys., 2002, 5, 449,
doi:10.5488/CMP.5.3.449.
29. Wucher A., Garrison B.J., J. Chem. Phys., 1996, 105, 5999, doi:10.1063/1.472451.
30. Betz G., Husinsky W., Philos. Trans. R. Soc. London, Ser. A, 2004, 362, 177, doi:10.1098/rsta.2003.1302.
33003-10
https://doi.org/10.1080/00337578208222984
https://doi.org/10.1557/JMR.1990.2687
https://doi.org/10.1557/JMR.1998.0205
https://doi.org/10.1016/0149-1970(94)00016-Y
https://doi.org/10.1016/0022-3115(94)90304-2
https://doi.org/10.1016/j.gca.2011.03.040
https://doi.org/10.1088/0953-8984/13/9/317
https://doi.org/10.1103/PhysRevB.73.174207
https://doi.org/10.1016/j.nimb.2011.12.058
https://doi.org/10.1016/j.nimb.2010.05.007
https://doi.org/10.1016/j.jnucmat.2006.01.020
https://doi.org/10.1016/j.nimb.2009.06.078
https://doi.org/10.1016/j.jnucmat.2008.12.010
https://doi.org/10.1016/j.nimb.2010.12.075
https://doi.org/10.1016/j.nimb.2011.04.115
https://doi.org/10.1016/j.jnoncrysol.2011.02.026
https://doi.org/10.1016/j.jnoncrysol.2012.04.033
https://doi.org/10.1016/j.jnoncrysol.2012.07.037
https://doi.org/10.1016/j.nimb.2004.10.060
https://doi.org/10.1016/j.nimb.2004.10.059
https://doi.org/10.1016/j.jnucmat.2008.09.031
https://doi.org/10.1016/j.jallcom.2006.10.156
https://doi.org/10.1016/j.jallcom.2006.10.003
https://doi.org/10.1016/j.nimb.2009.06.113
https://doi.org/10.1088/0953-8984/15/37/L07
https://doi.org/10.1016/S0022-0728(00)00433-2
https://doi.org/10.5488/CMP.10.4.579
https://doi.org/10.5488/CMP.5.3.449
https://doi.org/10.1063/1.472451
https://doi.org/10.1098/rsta.2003.1302
Destruction processes in self-irradiated materials
31. Robinson M.T., J. Appl. Phys., 1983, 54, 2650, doi:10.1063/1.332267.
32. Voskoboinikov R.E., Osetsky Yu.N., Bacon D.J., Nucl. Instrum. Methods Phys. Res., Sect. B, 2006, 242, 68–70,
doi:10.1016/j.nimb.2005.08.166.
33. Salonen E., Nordlund K., Keinonen J., Wu C.H., J. Nucl. Mater., 2003, 313–316, 404,
doi:10.1016/S0022-3115(02)01397-1.
34. Li A., Watts R.O., J. Chem. Phys., 1995, 103, 7187, doi:10.1063/1.470347.
35. LindenblattM., Heinrich R.,Wucher A., Garrison B.J., J. Chem. Phys., 2001, 115, 8643, doi:10.1063/1.1404982.
36. Day D.H., Hughes A.E., Leake J.W., Marples J.A.C., Marsh G.P., Rae J., Wade B.O., Rep. Prog. Phys., 1985,
48, 101, doi:10.1088/0034-4885/48/1/003.
37. Palenik C.S., Nasdala L., Ewing R.C., Am. Mineral., 2003, 88, 770–781, doi:10.2138/am-2003-5-606.
38. Kubota A., Caturla M.J., Payne S.A., Diaz de la Rubia T., Latkowski J.F., J. Nucl. Mater., 2002, 307–311, 891,
doi:10.1016/S0022-3115(02)01008-5.
39. Kubota A., Caturla M.J., Stolken J., Sadigh B., Reyes S., Diaz de la Rubia T., Latkowski J.F., Nucl. Instrum.
Methods Phys. Res., Sect. B, 2003, 202, 88, doi:10.1016/S0168-583X(02)01837-2.
40. Pazukhin E.M., Radiochemistry, 1994, 36, 97.
41. Zhidkov A.V., Problems of Chornobyl, 2001, 7, 23.
42. Wronkiewicz D.J., Preprint ANL/CMT/CP–80028, Argonne National Laboratory, 1993.
43. Chopard B., Droz M., Cellular Automata Modeling of Physical Systems, Cambridge University Press, Cam-
bridge, 1998.
44. Hoshen J., Kopelman R., Phys. Rev. B, 1976, 14, 3438, doi:10.1103/PhysRevB.14.3438.
45. Barabasi A.L., Stanley H.E., Fractal Concepts in Surface Growth, Cambridge University Press, Cambridge,
1995.
46. Stashchuk M.H., Stashchuk A.M., Mater. Sci., 1998, 34, 383, doi:10.1007/BF02355628.
Стохастичне комп’ютерне моделювання процесiв
руйнування в самоопромiнених матерiалах
Т. Пацаган1, А. Талеб2,3, Я. Стафiєй4,М. Головко1, Ж.П. Бадiалi1,3
1 Iнститут фiзики конденсованих систем Нацiональної академiї наук України,
вул. Свєнцiцького, 1, 79011 Львiв, Україна
2 Дослiдницький унiверситет науки та лiтератури Парижу, Chimie ParisTech— CNRS, Iнститут хiмiчних
дослiджень Парижу, Париж, Францiя
3 Унiверситет П’єра iМарiї Кюрi, Париж, 75231, Францiя
4 Унiверситет кардинала Стефана Вишинського, факультет математики i природничих наук,
Варшава, Польща
Самоопромiнення, що виникає внаслiдок процесiв подiлу радiоактивних елементiв, є звичним явищем,
яке спостерiгається в радiоактивних паливовмiсних матерiалах (ПВМ). Численнi α-розпади призводять до
перетворення локальної структури ПВМ. Ушкодження, якi виникають за рахунок ударiв вiддачi осколкiв
подiлу важких ядер в приповерхневому шарi, можуть спричинювати вiдокремлення частинок матерiалу.
Така поведiнка є подiбною до процесу розпилення, яке спостерiгається пiд час бомбардування поверхнi
матерiалу потоком енергетичних частинок. Проте, в ПВМ удари iнiцiюються в об’ємi матерiалу. В цiй ро-
ботi ми пропонуємо двовимiрну мезоскопiчну модель, з метою здiйснення стохастичного комп’ютерного
моделювання процесiв руйнування, що виникають в приповерхневiй областi ПВМ. Ми описуємо еро-
зiю поверхнi матерiалу, змiну її шорсткостi з часом i передбачаємо вiдокремлення частинок матерiалу.
У цьому дослiдженнi отримано розподiли розмiрiв вiдокремлених частинок. Результати комп’ютерного
моделювання якiсно узгоджуються iз гiстограмами розмiрiв розпилених частинок спостережених в лаво-
подiбних паливовмiсних масах, якi сформувалися пiд час Чорнобильської катастрофи.
Ключовi слова: радiоактивнi паливовмiснi матерiали, самоопромiнююче ушкодження, руйнування,
шорсткiсть, стохастичне комп’ютерне моделювання, процес розпилення
33003-11
https://doi.org/10.1063/1.332267
https://doi.org/10.1016/j.nimb.2005.08.166
https://doi.org/10.1016/S0022-3115(02)01397-1
https://doi.org/10.1063/1.470347
https://doi.org/10.1063/1.1404982
https://doi.org/10.1088/0034-4885/48/1/003
https://doi.org/10.2138/am-2003-5-606
https://doi.org/10.1016/S0022-3115(02)01008-5
https://doi.org/10.1016/S0168-583X(02)01837-2
https://doi.org/10.1103/PhysRevB.14.3438
https://doi.org/10.1007/BF02355628
Introduction
Physical aspects
Modelling
General concept
Stochastic model
Results and discussion
Conclusions
|