Effect of increasing surface roughness on sputtering and reflection
In this work, the SDTrimSP-2D code was used for numerical simulation of the interaction of ions with a 2D periodical structure as idealized test system to investigate the influence of surface roughness on sputtering. Sputtering yield and reflection coefficient have been studied as a function of the...
Gespeichert in:
| Datum: | 2012 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | English |
| Veröffentlicht: |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
2012
|
| Schriftenreihe: | Вопросы атомной науки и техники |
| Schlagworte: | |
| Online Zugang: | http://dspace.nbuv.gov.ua/handle/123456789/109143 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Zitieren: | Effect of increasing surface roughness on sputtering and reflection / I. Bizyukov, A. Mutzke, R. Schneider // Вопросы атомной науки и техники. — 2012. — № 6. — С. 111-113. — Бібліогр.: 6 назв. — англ. |
Institution
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-109143 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1091432016-11-21T03:02:54Z Effect of increasing surface roughness on sputtering and reflection Bizyukov, I. Mutzke, A. Schneider, R. Динамика плазмы и взаимодействие плазмы со стенкой In this work, the SDTrimSP-2D code was used for numerical simulation of the interaction of ions with a 2D periodical structure as idealized test system to investigate the influence of surface roughness on sputtering. Sputtering yield and reflection coefficient have been studied as a function of the size of the pitch grating structure. Simulations show that the most important changes in ion-surface interactions occur when the structure size gets approximately equal to the size of the collisional cascade.. Код SDTrimSP-2D использовался для моделирования взаимодействия ионов с двухмерной поверхностью, которая взята в качестве идеализированной тестовой системы для исследования влияния шероховатости. Коэффициенты распыления и отражения изучались как функции характерного размера структуры дифракционной решетки. Моделирование показало, что наиболее важные изменения во взаимодействии ионов с поверхностью происходят тогда, когда размер структуры приблизительно равен размеру столкновительного каскада. Код SDTrimSP-2D використовувався для моделювання взаємодії іонів з двомірною поверхнею, яка обрана у якості ідеалізованої тестової системи для дослідження впливу шорсткості. Коефіцієнти розпилення і відбиття вивчалися як функціі характерного розміру структури дифракційної решітки. Моделювання показало, що найбільш важливі зміни у взаємодії іонів з поверхнею відбуваються тоді, коли розмір структури приблизно дорівнює розміру каскаду зіткнень. 2012 Article Effect of increasing surface roughness on sputtering and reflection / I. Bizyukov, A. Mutzke, R. Schneider // Вопросы атомной науки и техники. — 2012. — № 6. — С. 111-113. — Бібліогр.: 6 назв. — англ. 1562-6016 PACS: 79.20 Rf http://dspace.nbuv.gov.ua/handle/123456789/109143 en Вопросы атомной науки и техники Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
Динамика плазмы и взаимодействие плазмы со стенкой Динамика плазмы и взаимодействие плазмы со стенкой |
| spellingShingle |
Динамика плазмы и взаимодействие плазмы со стенкой Динамика плазмы и взаимодействие плазмы со стенкой Bizyukov, I. Mutzke, A. Schneider, R. Effect of increasing surface roughness on sputtering and reflection Вопросы атомной науки и техники |
| description |
In this work, the SDTrimSP-2D code was used for numerical simulation of the interaction of ions with a 2D periodical structure as idealized test system to investigate the influence of surface roughness on sputtering. Sputtering yield and reflection coefficient have been studied as a function of the size of the pitch grating structure. Simulations show that the most important changes in ion-surface interactions occur when the structure size gets approximately equal to the size of the collisional cascade.. |
| format |
Article |
| author |
Bizyukov, I. Mutzke, A. Schneider, R. |
| author_facet |
Bizyukov, I. Mutzke, A. Schneider, R. |
| author_sort |
Bizyukov, I. |
| title |
Effect of increasing surface roughness on sputtering and reflection |
| title_short |
Effect of increasing surface roughness on sputtering and reflection |
| title_full |
Effect of increasing surface roughness on sputtering and reflection |
| title_fullStr |
Effect of increasing surface roughness on sputtering and reflection |
| title_full_unstemmed |
Effect of increasing surface roughness on sputtering and reflection |
| title_sort |
effect of increasing surface roughness on sputtering and reflection |
| publisher |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| publishDate |
2012 |
| topic_facet |
Динамика плазмы и взаимодействие плазмы со стенкой |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/109143 |
| citation_txt |
Effect of increasing surface roughness on sputtering and reflection / I. Bizyukov, A. Mutzke, R. Schneider // Вопросы атомной науки и техники. — 2012. — № 6. — С. 111-113. — Бібліогр.: 6 назв. — англ. |
| series |
Вопросы атомной науки и техники |
| work_keys_str_mv |
AT bizyukovi effectofincreasingsurfaceroughnessonsputteringandreflection AT mutzkea effectofincreasingsurfaceroughnessonsputteringandreflection AT schneiderr effectofincreasingsurfaceroughnessonsputteringandreflection |
| first_indexed |
2025-07-07T22:37:33Z |
| last_indexed |
2025-07-07T22:37:33Z |
| _version_ |
1837029501432758272 |
| fulltext |
ISSN 1562-6016. ВАНТ. 2012. №6(82) 111
EFFECT OF INCREASING SURFACE ROUGHNESS ON SPUTTERING
AND REFLECTION
I. Bizyukov1, A. Mutzke2, R. Schneider3
1V.N. Karazin Kharkov National University, Kharkov, Ukraine;
2Max-Planck-Institut für Plasmaphysik, Wendelsteinstr. 1, 17491 Greifswald, Germany;
3Ernst-Moritz-Arndt University, Felix-Hausdorff-Str. 6, 17489 Greifswald, Germany
E-mail: ivan.bizyukov@mail.ru
In this work, the SDTrimSP-2D code was used for numerical simulation of the interaction of ions with a 2D periodi-
cal structure as idealized test system to investigate the influence of surface roughness on sputtering. Sputtering yield
and reflection coefficient have been studied as a function of the size of the pitch grating structure. Simulations show
that the most important changes in ion-surface interactions occur when the structure size gets approximately equal to the
size of the collisional cascade..
PACS: 79.20 Rf
INTRODUCTION
The sputtering of surface atoms by ion bombardment
is a well-known process [1], which is utilized by many
plasma technological applications. Most experiments are
performed with a polished and smooth surface; the simu-
lations assumed also that the surface is perfectly plane.
Up to now, the influence of surface roughness on sputter-
ing is not understood well, because there was no suitable
model, which was able to provide a comprehensive de-
scription. Only few attempts had been made in the past.
Ruzic has modified the TRIM.SP code to study the sput-
tering of the surface with fractal geometry [2]. Kuestner,
Eckstein and co-authors had considered the rough surface
as an aggregate of simple surfaces at inclined angles [3].
Later, the SDTrimSP-2D code [4] has been developed to
simulate interaction of ions with the 2-D non-planar sur-
faces. It is a powerful tool for the study of surface mor-
phology effects. The validation of the code has been per-
formed exposing a Si pitch grating with typical dimen-
sions of 200…250 nm to 6 keV Ar ion beam [5, 6].
In this work 6 keV Ar ion projectiles are bombarding
a Si pitch grating with a periodic 2D structure of varying
size representing surface roughness. This rather idealized
system has been used for numerical investigation of the
sputtering yield and reflection coefficient as a function of
the size of the surface roughness. The characteristic size
of the surface morphology is varied in the range of
1…100 nm. Previous studies [5, 6] have validated the
code for this particular target-projectile combination and
it was confirmed that such structure exhibits all the effects
expected for rough surface: local increase of sputtering
due to inclined incidence of ions, contribution from sput-
tering by reflected projectiles, strong influence of the re-
deposition, etc.
1. METHODS
The simulations have been performed by the
SDTrimSP-2D code [4]. The surface is shaped in two
dimensions (vertical and lateral) and extended in the third
direction. The cross-section of the surface is shown in
Fig. 1,a; one can see that the typical dimension h charac-
terizes width and height of the structure.
Fig. 1. Cross-section of the Si surface with 2-D structure:
a – typical sizes of the structure in units of h;
b – structure scaled according to different h values
The simulations have been run in static mode, i.e. pro-
jectiles and collisional cascades do not change the struc-
ture and elemental composition of the irradiated surface.
Each simulation was run with a particular value of the
typical dimension h, varied between 1 and 100 nm. While
the typical size h is changed, the shape of the structure is
preserved; Fig. 1,b shows the size of cross-sections of the
structure for different h values.
As results, one can obtain the dependence of sputter-
ing yield and reflection coefficient on the typical rough-
ness size h. Moreover, code diagnostics delivers the par-
tial sputter yields from different parts of the surface: left,
right, top and bottom parts of the structure. Therefore, one
can analyze the contributions of different surface parts to
the total sputtering yield and reflection coefficient.
2. RESULTS AND DISCUSSION
Fig. 2 shows typical trajectories of projectiles and re-
coils, forming a collisional cascade for an impact at nor-
mal incidence. The absolute maximum of depth profiles is
30 nm and the absolute maximum of lateral spreads, R, is
16 nm for all calculated trajectories. The average of the
maxima of depth profiles for all cascades is 16.5 nm and
the average of all maxima of lateral spreads, R, is
a
b
ISSN 1562-6016. ВАНТ. 2012. №6(82) 112
10.2 nm. These two values were calculated from the point
of the impact of projectile to the depth position where one
has the maximum number of intermediate points of trajec-
tories. While the typical size h of the structure is increas-
ing, the mean size R of the collisional cascade remains the
same (see Fig. 2). The influence of one cascade is more
local if the size h increased. Fig. 3 shows the dependence
of sputtering yield and reflection coefficient on typical
structure size h.
Fig. 2. Trajectories of projectile and recoils during the
development of the collisional cascade:
a – h=1 nm; b – h=5 nm; c – h=10 nm; d – h=100 nm
When R>>h, the collisional cascade develops under
the bottom surface of the structure, as one can see in
Fig. 2,a. One should expect that this surface behaves as a
planar one with regard to sputtering and reflection. The
number of atoms reaching the surface is only slightly in-
creased in comparison to the case of a planar surface. The
effect of inclined surfaces is small. Furthermore, the mean
free path of the projectile is 0.27 nm. The influence of
target geometry in this case is negligible. Therefore, geo-
metrical effects should have rather small influence on
sputtering and reflection. This is confirmed by the simula-
tion, which indicates that the sputtering yield of a planar
surface differs only marginally from the rough one
(Yplanar=1.4 and Yrough=1.5). Simulations show that the
sputtering yield and the reflection coefficient grow only
slightly, when h<1 nm (see Fig. 3).
At R≈h, the situation becomes different. Here, the size
of the collisional cascade and the structure are similar.
The cascade spreads over approximately one structure
period, as one can see in Fig. 2,b and c. The number of
recoils, which reach the surface at the inclined side of the
structure, increases. They may leave the surface as sput-
tered atoms and, if re-deposition is avoided, these atoms
contribute to the overall sputtering yield. This is compa-
rable with the effect of bombardment at an inclined inci-
dent angle in the planar case.
The strongest growth of the sputtering yield and re-
flection coefficient occurs, until the typical structure size
does no longer exceed the size of the collisional cascade
(mostly for 1 nm<h<10 nm). However, different parts of
the surface behave different in terms of sputtering and
reflection. Fig. 3 shows the partial sputtering yields and
reflection coefficients for top, bottom and inclined sur-
faces. Sputtering is possible for two cases: either a projec-
tile reaches the particular surface or an event occurs on
this particular surface due to particles originating from
projectiles impacting at different locations. The simula-
tion shows that the yields for these two cases are not
equal and one can extract additional information on the
development of collisional cascades on rough surfaces.
Fig. 3. Sputtering yield as a function of typical size h of
the structure. The partial values of the parameters have
been obtained according to the location, from which the
atom leaves the structure; the lines marked as “incident”
in the legend show the sputtering yield according to the
location, where the projectile enters the surface (a).
Reflection coefficient as a function of typical size h of the
structure; the partial values of the parameters have been
obtained according to the location, from which the atom
leaves the structure (b)
Projectiles bombarding the top of the structure pro-
duce the highest partial sputtering yields (see Fig. 3 in the
range of structure size h of 1 nm<h<10 nm). In contrast,
there are much less atoms sputtered from this location.
Collisional cascades develop due to bombardment of the
top surface, which produces recoils. These leave the sur-
face by reaching side and bottom structures. The sputter-
ing yield produced by direct bombardment of the sides is
lower than that produced indirectly from atoms leaving
through the side structure. Extra sputtering events are
produced by collisional cascades initiated on other sur-
faces (obviously, the top one). Similar behavior is seen on
the bottom of the structure. Summarizing, one can con-
clude that the collisional cascades are initiated mostly on
the top of the structure and their development can produce
sputtered atoms on the sides. The same effect is less pro-
nounced if cascades from the sides are considered. Fi-
nally, on the bottom surface the collisional cascades
mainly go deep into the structure material and additional
sputtering is produced by recoils originating from colli-
sional cascades initiated on other surfaces.At large struc-
ture sizes, when R<<h, the collisional cascade is much
, nm
, n
m
, nm
h, nm
a
b
a b
c d
ISSN 1562-6016. ВАНТ. 2012. №6(82) 113
smaller than the structure; a typical example is shown in
Fig. 2,d. Now, the interaction of ions with the structure
can be well described in a 1-D approximation, i.e. one can
calculate sputtering yield and reflection coefficient as-
suming that projectiles interact with aggregates of the
inclined surfaces, as it has been performed by Kuestner et
al. [3]. However, the redeposition has not been taken into
account in this 1-D approximation. The analysis of the
data presented in Fig. 3 shows that the contribution from
the top surface to the sputtering is strongly reduced.
The reflection coefficient has a similar dependency as
the sputtering yield (see Fig. 3); the inclined surfaces pro-
duce more reflected projectiles due to the effective inci-
dent angle and to multiple reflections. One example of
possible trajectories is shown in Fig. 2,d. The reflection
coefficient from the top surface remains constant, while
the structure size is growing. This is explained by the fact
that most projectiles are obviously scattered from the top
surface through single reflection events. In contrast, pro-
jectiles incident on side surfaces are reflected towards the
other surfaces of the structure. Therefore, there are projec-
tiles, which experience multiple scattering before finally
leaving the surface. While the structure size is increasing,
one can observe that the last reflection of the scattered
projectiles occurs mostly at the side surface. Some of the
scattered projectiles leave the structure reflected from the
bottom of the structure. This explains why the reflection
coefficient from the bottom location is growing with
structure size h.
CONCLUSIONS
In this work, the SDTrimSP-2D code was used to in-
vestigate the interaction of ions with a rough surface. We
used an idealized representation, namely a 2D periodical
structure as test system to clarify the basic physics trends.
Such a 2D periodic Si pitch grating has been exposed to
an ion flux of 6 keV Ar and the sputtering yield and re-
flection coefficient have been studied as a function of the
size of the pitch grating structure. It has been shown that
the bombardment of the surface at normal angle of inci-
dence (relatively to the macroscopic plane) results in in-
creases of both the sputtering yield and reflection coeffi-
cient with increasing structure size. The largest increase
of the sputtering yield is observed in the range between 1
and 10 nm, which corresponds to the typical size of colli-
sional cascades, initiated by the 6 keV Ar projectiles.
Therefore, if the roughness size is larger than the typical
size of the collisional cascade, one gets the highest possi-
ble sputtering yield. Another benefit of large surface
roughness is the reproducibility of sputtering yields or
deposition rates, if the deposition utilizes sputtering (like
in magnetron sputter deposition).
REFERENCES
1. R. Behrisch, W. Eckstein (ed). Sputtering by Particle
Bombardment. Experiments and Computer Calculations
from Threshold to MeV Energies. Berlin, Springer, 2007.
2. D. Ruzic. The effects of surface roughness character-
ized by fractal geometry on sputtering // Nucl. Instrum.
Meth. 1990, v. B47, p. 118-125.
3. M. Kuestner, W. Eckstein, E. Hechtl, and J. Roth. An-
gular dependence of the sputtering yield of rough beryl-
lium surfaces // J. Nucl. Mater. 1999, v. 265, p. 22-27.
4. A. Mutzke, R. Schneider. SDTrimSP-2D: Simulation of
Particles Bombarding on a Two Dimensional Target Ver-
sion 1.0. IPP Report 12/4. Garching, Max-Planck-Institute
5. I. Bizyukov, A. Mutzke, R. Schneider, and J. Davis.
Evolution of the 2D surface structure of a silicon pitch
grating under argon ion bombardment: Experiment and
modeling. // Nucl. Instr. and Meth. 2010, v. B268,
p. 2631-2638.
6. I. Bizyukov, A. Mutzke, M. Mayer, H. Langhuth,
K. Krieger, R. Schneider. Macroscopic parameters of the
interaction of an Ar+ ion beam with a Si pitch grating //
Nucl. Instr. and Meth. 2012, v. B278, p. 4-7.
Article received 20.09.12
ВЛИЯНИЕ ШЕРОХОВАТОСТИ ПОВЕРХНОСТИ НА РАСПЫЛЕНИЕ И ОТРАЖЕНИЕ
И. Бизюков, А. Муцке, Р. Шнайдер
Код SDTrimSP-2D использовался для моделирования взаимодействия ионов с двухмерной поверхностью,
которая взята в качестве идеализированной тестовой системы для исследования влияния шероховатости.
Коэффициенты распыления и отражения изучались как функции характерного размера структуры дифракцион-
ной решетки. Моделирование показало, что наиболее важные изменения во взаимодействии ионов с поверхно-
стью происходят тогда, когда размер структуры приблизительно равен размеру столкновительного каскада.
ВПЛИВ ШОРСТКОСТІ ПОВЕРХНІ НА РОЗПИЛЕННЯ ТА ВІДБИТТЯ
І. Бізюков, А. Муцке, Р. Шнайдер
Код SDTrimSP-2D використовувався для моделювання взаємодії іонів з двомірною поверхнею, яка обрана у
якості ідеалізованої тестової системи для дослідження впливу шорсткості. Коефіцієнти розпилення і відбиття
вивчалися як функціі характерного розміру структури дифракційної решітки. Моделювання показало, що
найбільш важливі зміни у взаємодії іонів з поверхнею відбуваються тоді, коли розмір структури приблизно
дорівнює розміру каскаду зіткнень.
|