Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events
This paper is focused on experimental analysis of melt layer erosion and droplet splashing of tungsten and other metals under heat loads typical for ITER FEAT off-normal events, such as disruptions and VDE’s. Plasma pressure gradient action on melt layer results in erosion crater formation with moun...
Збережено в:
| Дата: | 2002 |
|---|---|
| Автори: | , , , , , , , , , , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
2002
|
| Назва видання: | Вопросы атомной науки и техники |
| Теми: | |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/77871 |
| 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: | Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events / I.E. Garkusha, A.N. Bandura, O.V. Byrka, V.V. Chebotarev, V.V. Garkusha, N.V. Kulik, V.A. Makhlaj, D.G. Solyakov, V.V. Stal’tsov, V.I. Tereshin, I. Landman, H. Wuerz // Вопросы атомной науки и техники. — 2002. — № 5. — С. 30-32. — Бібліогр.: 8 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-77871 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-778712015-03-09T03:01:54Z Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events Garkusha, I.E. Bandura, A.N. Byrka, O.V. Chebotarev, V.V. Garkusha, V.V. Kulik, N.V. Makhlaj, V.A. Solyakov, D.G. Stal’tsov, V.V. Tereshin, V.I. Landman, I. Wuerz, H. ITER and fusion reactor aspects This paper is focused on experimental analysis of melt layer erosion and droplet splashing of tungsten and other metals under heat loads typical for ITER FEAT off-normal events, such as disruptions and VDE’s. Plasma pressure gradient action on melt layer results in erosion crater formation with mountains of displaced material at the crater edge. It is shown that macroscopic motion of melt layer and surface cracking are the main factors responsible for tungsten damage. Weight loss measurements of all exposed materials demonstrate inessential contribution of evaporation process to metals erosion. 2002 Article Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events / I.E. Garkusha, A.N. Bandura, O.V. Byrka, V.V. Chebotarev, V.V. Garkusha, N.V. Kulik, V.A. Makhlaj, D.G. Solyakov, V.V. Stal’tsov, V.I. Tereshin, I. Landman, H. Wuerz // Вопросы атомной науки и техники. — 2002. — № 5. — С. 30-32. — Бібліогр.: 8 назв. — англ. 1562-6016 PACS: 52.40.Hf; 52.55.Rk http://dspace.nbuv.gov.ua/handle/123456789/77871 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 |
ITER and fusion reactor aspects ITER and fusion reactor aspects |
| spellingShingle |
ITER and fusion reactor aspects ITER and fusion reactor aspects Garkusha, I.E. Bandura, A.N. Byrka, O.V. Chebotarev, V.V. Garkusha, V.V. Kulik, N.V. Makhlaj, V.A. Solyakov, D.G. Stal’tsov, V.V. Tereshin, V.I. Landman, I. Wuerz, H. Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events Вопросы атомной науки и техники |
| description |
This paper is focused on experimental analysis of melt layer erosion and droplet splashing of tungsten and other metals under heat loads typical for ITER FEAT off-normal events, such as disruptions and VDE’s. Plasma pressure gradient action on melt layer results in erosion crater formation with mountains of displaced material at the crater edge. It is shown that macroscopic motion of melt layer and surface cracking are the main factors responsible for tungsten damage. Weight loss measurements of all exposed materials demonstrate inessential contribution of evaporation process to metals erosion. |
| format |
Article |
| author |
Garkusha, I.E. Bandura, A.N. Byrka, O.V. Chebotarev, V.V. Garkusha, V.V. Kulik, N.V. Makhlaj, V.A. Solyakov, D.G. Stal’tsov, V.V. Tereshin, V.I. Landman, I. Wuerz, H. |
| author_facet |
Garkusha, I.E. Bandura, A.N. Byrka, O.V. Chebotarev, V.V. Garkusha, V.V. Kulik, N.V. Makhlaj, V.A. Solyakov, D.G. Stal’tsov, V.V. Tereshin, V.I. Landman, I. Wuerz, H. |
| author_sort |
Garkusha, I.E. |
| title |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events |
| title_short |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events |
| title_full |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events |
| title_fullStr |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events |
| title_full_unstemmed |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events |
| title_sort |
melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating iter off-normal events |
| publisher |
Національний науковий центр «Харківський фізико-технічний інститут» НАН України |
| publishDate |
2002 |
| topic_facet |
ITER and fusion reactor aspects |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/77871 |
| citation_txt |
Melt layer macroscopic erosion of tungsten and other metals under plasma heat loads simulating ITER off-normal events / I.E. Garkusha, A.N. Bandura, O.V. Byrka, V.V. Chebotarev, V.V. Garkusha, N.V. Kulik, V.A. Makhlaj, D.G. Solyakov, V.V. Stal’tsov, V.I. Tereshin, I. Landman, H. Wuerz // Вопросы атомной науки и техники. — 2002. — № 5. — С. 30-32. — Бібліогр.: 8 назв. — англ. |
| series |
Вопросы атомной науки и техники |
| work_keys_str_mv |
AT garkushaie meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT banduraan meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT byrkaov meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT chebotarevvv meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT garkushavv meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT kuliknv meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT makhlajva meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT solyakovdg meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT staltsovvv meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT tereshinvi meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT landmani meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents AT wuerzh meltlayermacroscopicerosionoftungstenandothermetalsunderplasmaheatloadssimulatingiteroffnormalevents |
| first_indexed |
2025-07-06T02:07:19Z |
| last_indexed |
2025-07-06T02:07:19Z |
| _version_ |
1836861505357742080 |
| fulltext |
MELT LAYER MACROSCOPIC EROSION OF TUNGSTEN
AND OTHER METALS UNDER PLASMA HEAT LOADS
SIMULATING ITER OFF-NORMAL EVENTS
I.E. Garkusha, A.N. Bandura, O.V. Byrka, V.V. Chebotarev, V.V. Garkusha, N.V. Kulik,
V.A. Makhlaj, D.G. Solyakov, V.V. Stal’tsov, V.I. Tereshin, I. Landman*, H.Wuerz*
Institute of Plasma Physics of the NSC KIPT, Kharkov, 61108, Ukraine
* Forschungszentrum Karlsruhe, IHM, 76021 Karlsruhe, Germany
This paper is focused on experimental analysis of melt layer erosion and droplet splashing of tungsten and other metals
under heat loads typical for ITER FEAT off-normal events, such as disruptions and VDE’s. Plasma pressure gradient action
on melt layer results in erosion crater formation with mountains of displaced material at the crater edge. It is shown that
macroscopic motion of melt layer and surface cracking are the main factors responsible for tungsten damage. Weight loss
measurements of all exposed materials demonstrate inessential contribution of evaporation process to metals erosion.
PACS: 52.40.Hf; 52.55.Rk
1. INTRODUCTION
Materials irradiation with plasma streams generated by
powerful plasma accelerators [1,2], which can simulate at
least in magnitude of heat load, the conditions expected for
ITER off-normal events, is used at present for numerical
models validation and for experimental simulation of metal
targets erosion under high heat loads.
During such off-normal events as disruptions and vertical
displacement events (VDEs) energy flux at the armour
material reaches values sufficient for melting of metal
surfaces. The melt layer is subjected to external forces such
as surface tension, gradients of both plasma pressure and
recoil pressure of evaporating material, Lorentz force and
others. Melt motion driven by external forces may produce
significant macroscopic erosion of materials [3,4]. It is
expected that under the VDE rather large area of surface will
be heated with a practically constant heat load. In contrast to
VDE, during disruptions heating occurs with a characteristic
heat load profile having the peak value at the separatrix
strike point (SSP). As a consequence, the external pressure
and surface tension depend on the position along the melt
surface. Results of numerical simulations of disruption have
shown that the pressure profiles of the plasma shield after 4
ms are between 4 and 7 bar and the pressure profiles have a
half width of 4 cm only [5].
This paper presents the experimental analysis of
contribution of different erosion mechanisms to the material
damage under high heat loads simulating these features of
VDE and disruption.
2. EXPERIMENTAL SETUP
Different metal targets were exposed to perpendicular
and inclined plasma impact in QSPA Kh-50, which is
described elsewhere [1]. The parameters of the free hydrogen
plasma stream at the target position were as follows: average
density of 4x1016 cm-3, plasma stream energy density up to
25-30 MJ/m2, ion energy below 0.6 keV, discharge duration
t= 0.3 ms and power pulse duration (half height width) –
0.10-0.14 ms. Plasma stream maximal pressure in near the
axis region achieved (1.6-1.8)x106 Pa. The diameter of the
QSPA plasma stream was 10-12 cm. The total energy of the
plasma stream exceeded 160 kJ. A guiding magnetic field of
0.54 T was applied in experiments (average β ~ 0.3-0.4).
A profilometer with an accuracy of 0.4 µm was used for
analysis of the surface of the melt layer. As the sensitive
element, a small ball was applied instead of a diamond pin
for surface profile measurements to avoid surface roughness
contribution to the profilograms. The unexposed part of the
target was used as a reference for profilometry. Surface
analysis was carried out with an optical microscope. X-ray
diffraction analysis and weight loss measurements were
performed also.
Radial distributions of plasma stream pressure measured
by piezodetector at the target position are presented in Fig. 1
As follows from this figure, in spite of high values of
pressure achieved in QSPA plasma stream, rather small
pressure gradient is in the central part of the plasma stream
corresponding to the targets position usually used.
That is why, for simulation of disruptions the targets
were exposed through molybdenum diaphragms of different
diameters to impose a pressure gradient along the target that
mimics the pressure gradient found at the strikepoint
locations in a tokamak disruption. Such experimental scheme
is described in details in [6].
0 2 4 6 8 10 12 14
0
4
8
12
16
20
τ = 40 µ s
τ = 80 µ s
P,
b
ar
R, cm
Fig.1. Radial distributions of plasma stream pressure at
the target position
3. MELT MOTION
The profile of tungsten melt surface is presented in Fig.
2,a for perpendicular exposure with 20 pulses through the 2
cm hole. As it follows from profilometry, mountains of melt
materials, indicating the melt motion, arise at the melt edge.
The height of the mountains achieves 65 µm. The high value
of the surface roughness Rz ~ 30 µm masks an erosion crater
between the mountains and the ball sensor scans only the
roughness peaks. Only tendency of crater formation is
registered. It should be mentioned that even for a target
thickness of 6 mm the influence of bending is seen on the
profilograms. Sagging of the center of the target is about 10
µm, although initially it has a good flatness. Profile of melt
spot of inclined tungsten target (α=200) irradiated with 20
pulses through the 2 cm hole is presented in Fig. 2,b. The
specific heat flux for exposure of an inclined target became
essentially less (as compare with the perpendicular impact)
due to the increase of the plane projection of the plasma
30 Problems of Atomic Science and Technology. 2002. № 5. Series: Plasma Physics (8). P. 30-32
stream to the target surface Formation of the mountain peak
of 28 µm in height under the melt motion is observed at the
downstream part of the melt spot only. Therefore the melt
motion is dominated by plasma pressure. Contribution of
surface tension gradient is insignificant and not seen against
a background of triggered surface roughness.
Dynamics of the erosion crater and mountains
formation in dependence on irradiation dose is shown in
Fig.3 for irradiation of Ti sample through the 1 cm hole. The
erosion crater with uniform depth is clearly registered even
for exposure with one pulse because of the lower level of
surface roughness and more pronounced melt motion for Ti.
Fig.2. Erosion profiles of tungsten targets exposed with
20 pulses through the 2 cm hole; a- perpendicular
impact, b- inclined plasma impact with α=200
a
b
Fig. 3. Growth of erosion profile for Ti target under
perpendicular plasma impact; a- 1 pulse, b- 5 and 20 pulses
Due to the thermo-mechanical properties of Ti (density,
heat conductivity and melt temperature) higher melt
velocities and melt layer depth are realized. Depth of
erosion crater (i.e. erosion by melt motion) achieved 10-12 µ
m/pulse. The height of ridge arisen as result of first pulse
action is up to 70 µm. There is an approximate balance
between material loss in the crater area and mountain
material. It was obtained that erosion crater depth and the
distance between mountain peaks increased with the number
of pulses and achieved 100 µm and 15 mm respectively after
15 pulses. The height of the mountains achieved 130 µm,
their width was also increased up to 4-6 mm. With a further
increase of the irradiation dose growth of the erosion crater
and mountains became essentially slower, because of the
resolidified mountain, restricts movement of the melt
initiated by consequent pulses. Erosion evaluated from
weight loss measurements for titanium targets is about 0.08
µm/pulse i.e. is negligible in comparison with erosion by
melt movement
Increase of hole diameter leads to changes in pressure
distribution along the target surface. Region with practically
constant pressure appears in the central part (Fig.4). In this
case registered erosion crater is nonuniform with a maximal
depth at the periphery (region of maximal pressure gradient).
Fig.4. Melt layer profiles for titanium target irradiated with
10 pulses (a) and with 20 pulses (b), with overlay of plasma
pressure profile (c). Vertical scale for pressure is 3.5
Bar/div. Hole diameter is 3 cm.
4. TUNGSTEN CRACKING
X-ray diffraction analysis of irradiated tungsten has
shown a decrease of the tungsten crystal lattice period. It has
appeared equal 3.1622 Å, while for initial nonirradiated
surface the lattice spacing is 3.1653 Å. This is the result of
compressive stresses, which arise in the resolidified layer
and are accompanied by plastic deformation of material.
Appearance of macrostresses leads to crack formation on the
tungsten surface. Both fine intergranular and large size
cracks are seen on target surfaces (Fig. 5). Thus alongside
with melt motion, the surface cracking is very important
process from the point of view tungsten damage.
Nevertheless, measurements of weight loss shows negligible
role of evaporation process in tungsten erosion. In terms of
erosion crater weight losses are about 0.04 µm/pulse. To
explain this result it should be noted that droplet splashing
(including one caused by melt motion and surface cracking)
is practically not contribute to weight loss because of
overwhelming majority of ejected droplets remain on the
sample surface. Lava flow observed at the edge of melt spot.
Large cracks are not registered in this region. Droplets
splashed to nonirradiated surface are observed also. Some
tracks of droplets ejected from the melt are seen at the
distance up to 1 cm from the melt spot. The size of droplets
is varied in the range of 1µm-100µm. The quantity of
droplets is in inverse proportion on their size. Analysis of
droplets tracks have shown that droplets registered far from
the melt edges were ejected with extremely high velocities
(at least tens meters per second)
One of possible ways to improve the durability of
tungsten against a cracking is use of tungsten coatings on
31
5 mm 50 µ m 5 mm 50 µ m 5 mm 50 µ m 5 mm 50 µ m 50 µ m
50 µm 5 mm
20 µm 5 mm
a
b
Fig.5. Images tungsten melt surface and splashed droplets near and far from the melt edge
40µm80µm50µm
other materials instead of massive tungsten target. Such
coatings are widely used in present-day tokamaks. As the
first step in investigations of W-coatings erosion, the
experiments on perpendicular and inclined irradiation of Cu
targets with W-coatings have been started. Tungsten coatings
were deposited with the planar ECR plasma source [7]. The
thickness of W-coatings was 2 µm. The main results of
coatings exposure can be summarized as follows:
Coating demonstrates rather high adhesion with
substrate. High heat load do not lead to shelling and
exfoliation of coating. In regimes of irradiation with energy
loads below the coating melting point, excellent durability of
the coating is observed. Even in conditions of melting under
the high heat load no shelling and exfoliation were
registered. As result of the surface melting under the plasma
exposure some mixing of coating with material substrate is
observed. Due to relatively small thickness of the coating
molten copper appears on the surface at the places of grain
boundaries. Erosion by melt motion for W-coated samples is
similar to tungsten targets. However surface cracking is
essentially less pronounced. Last result can be considered as
important advantage of coatings in comparison with
monolithic target.
5. SIMULATION OF VDE
Simulation of VDE is carried out with targets irradiation
without diaphragms. In this case the heat load and plasma
pressure are only slightly changed along the target surface
and melt zone up to 8-10 cm in diameter can be achievable.
Experiments have shown that the surfaces of the resolidified
melt layers have a considerable roughness with microcraters and
a ridge like relief on the surface. Melt layer erosion by melt
motion was clearly identified only for exposure of composite
targets [8]. Because of small value of pressure gradient in
plasma shield the melt motion is masked by boiling, bubble
expansion and bubble collapse and by formation of a Kelvin-
Helmholtz instability. It is not clearly seen against a
“background” of triggered high surface roughness, which
became the main erosion factor.
6. CONCLUSIONS
The disruption simulation experiments have shown that
metals erosion is dominated by melt motion. For
perpendicular plasma impact the melt layer motion driven by
plasma pressure gradient results in erosion crater formation
with rather large mountains of the resolidified material at the
crater edges. Analysis of plasma pressure distributions along
the surface exposed through the diaphragms with different
holes allows us to conclude that the most pronounced melt
motion (and maximal erosion crater) is registered in the
regions of the maximum gradient of plasma pressure.
Effect of deceleration for the erosion crater depth and
mountain growth with increasing the number of exposures
was observed. It is concluded that resolidified mountain,
formed by previous pulses, restricts movement of the melt
initiated by the consequent pulses.
Melt motion of metals is accompanied by droplet
splashing. The droplets size is varied from 1 to 100 microns.
The droplet velocity was evaluated on the basis of the droplet
size, the distance of their displacement and the duration of
the incident plasma stream exposure. The estimated velocity
depends on their size and is typically more than 5.102 cm/s.
Analysis of tungsten droplets tracks have shown that droplets
registered far from the melt edges were ejected with
extremely high velocities.
Exposure of inclined targets results in formation of
mountain only at the downstream part of the target surface.
This is a clear indication of plasma pressure influence. Under
the conditions realized in the QSPA the surface tension
gradient is not a determining factor for mountains formation
and melt motion is dominated by plasma pressure.
Weight loss measurements of all exposed materials
demonstrate inessential contribution of evaporation process
to metals erosion.
Tungsten targets show highest erosion resistance in
comparison with other metals. Nevertheless melt layer
motion and surface cracking are the main factors responsible
for tungsten damage. For ITER disruptions with much longer
duration of plasma exposure, the melt motion can be very
serious problem, especially in the case of additional action of
Lorentz force due to the currents flowing in the melt
Erosion by melt motion for W-coated samples is similar
to tungsten targets. However surface cracking is
essentially less pronounced.
Triggered surface roughness became the dominating
erosion factor only in the case of small values of driving
forces for the melt motion initiation.
ACKNOWLEDGEMENTS
This work has been performed within the WTZ project
UKR-02-009.
REFERENCES
[1] V.V. Chebotarev et al., J. Nucl. Mat. 233-237 (1996)
736.
[2] N.I. Arkhipov et al., Fus. Eng. and Design. 49-50 (2000)
151.
[3] H.Wuerz et al., J. Nucl. Mater. 290-293 (2001) 1138.
[4] A.M. Hassanein, Fusion Technology. 15 (1989) 513.
[5] H. Wuerz et al., J. Nucl. Mater. 307-311P1 (2003).
[6] V.I. Tereshin et al. Proc. of PSI-15. Submitted to J. Nucl.
Mater.
[7] V.D. Fedorchenko et al. Proc. of the Intern. Conf. and
School on Plasma Physics and Controlled Fusion.
Alushta, Ukraine, September,16-21, 2002. p.185.
32
[8] A.N. Bandura et al., J. Nucl. Mater. 307-311P1 (2003)
106.
Irradiation of composite copper SS- copper target was
performed to make visible the material displacement by melt
motion and to estimate the value of such displacement.
Stainless steel rod of 5 mm in diameter was molded into the
hole in copper target. Displacement of SS material due to the
melt motion is 1.8-2 mm for perpendicular exposure with 20
pulses. Melt displacement in downstream direction for
inclined with 300 target is about 1.5-1.8 mm after 40 pulses.
33
4. Tungsten Cracking
Coating demonstrates rather high adhesion with substrate. High heat load do not lead to shelling and exfoliation of coating. In regimes of irradiation with energy loads below the coating melting point, excellent durability of the coating is observed. Even in conditions of melting under the high heat load no shelling and exfoliation were registered. As result of the surface melting under the plasma exposure some mixing of coating with material substrate is observed. Due to relatively small thickness of the coating molten copper appears on the surface at the places of grain boundaries. Erosion by melt motion for W-coated samples is similar to tungsten targets. However surface cracking is essentially less pronounced. Last result can be considered as important advantage of coatings in comparison with monolithic target.
|