Transcript Influence of material choice on the deposition/erosion mechanisms

Influence of material choice on
the deposition/erosion
mechanisms affecting optical
reflectivity of metallic mirrors
G. De Temmermana, V.S. Voitsenyab, R.A. Pittsc M. Maurera,
L. Marota, and P. Oelhafena
a Institute
of Physics, University of Basel, Switzerland;
b Institute
of Plasma Physics, NSC KIPT, Akademischa St. 1, 61108 Kharkhov, Ukraine;
c Centre
de Recherches en Physique des Plasmas, Association EURATOM, Conférédation Suisse,
EPFL, 1015 Lausanne, Switzerland
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
1
Introduction
• Experiment in Tore Supra with plasma facing mirrors showed strong differences
in the sputtering yields of Mo, SS and Cu under similar exposure conditions
• Surface of the SS mirrors appeared to have been protected from sputtering
• A difference in the stickiness of carbon has been proposed to explain these
findings (V. Voitsenya, ITPA-9)
• Experiment was initiated in the Univ. Basel to clear up the behaviour of SS and
Cu mirrors submitted to D2 glow discharge with controlled partial pressure of
methane.
• In parallel, tests of different candidate materials were made in TCV where
samples are exposed in the divertor region
• Differences in the deposition efficiency of carbon on different substrates were
noticed
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
2
Laboratory experiments (Uni Basel)
• Exposure of metallic mirrors to a low temperature deuterium plasma with controlled
partial pressures of methane in the gas mixture
• 2 substrate materials: copper and stainless steel prepared in IPP Kharkov,
Ukraine
U= -200 V
Water cooled sample
holder
f CH 4 =0 / 1.8 / 3.5 %
Samples characterization:
In situ measurement of the reflectivity using laser reflectometry (532 nm)
Weight measurement: determination of the eroded/deposited depth
SEM: surface morphology
Total and diffuse reflectivity (250-2500 nm); Spectroscopic ellipsometry (350-2300 nm)
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
3
Reflectivity during plasma exposure
Stainless steel
Copper
1.0
1.0
0.8
Pure D2
fCH4=1.8%
0.6
R normalized
R normalized
0.8
fCH4=3.5%
0.4
=532 nm
0.2
0.6
D2 pure
0.4
f
=1.8%
CH4
fCH4=3.5%
0.2
=532 nm
Stainless steel
0.0
0.0
0
2
4
6
19
-2
Fluence (x10
cm )
8
0.0
0.5
1.0
1.5
2.0
19
-2
Fluence (x10
cm )
For f CH 4=3.5%, appearance of
interferences typical from the
growth of a:CH layer
Strong correlation between the carbon
content in the plasma and the
degradation rate of R
No carbon on the other samples
(EDX measurements)
All samples are "carbon free"
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
4
Evolution of the surface morphology
Stainless steel
Copper
f CH 4  0%
f CH 4  0%
Ra  7nm
Ra  4nm
f CH 4  1.8%
f CH 4  1.8%
Ra  6nm
Ra  26nm
f CH 4  3.5%
f CH 4  3.5%
Ra  70nm
No significant effect of physical
sputtering
Deterioration of the reflectivity by an
increase of the roughness
Carbon protection effect
Appearance of the crystallographic grains
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
5
Erosion/deposition measurements
• Eroded/ deposited depth estimated both from mass loss measurements and
profilometry
Surface thickness change (nm)
400
Stainless steel
Copper
200
Deposition
0
Erosion
-200
-400
-600
0
1
2
3
4
fCH4(%)
Different behaviour of both substrates towards erosion/deposition
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
6
Reflectivity after exposure
90
80
70
60
50
40
30
20
10
0
100
Stainless steel
Stainless steel
Specular reflectivity
Virgin mirror
Pure D2
fCH4=1.8%
fCH4=3.5%
500
1000
1500
2000
2500
Wavelength (nm)
90
Copper
Specular reflectivity
80
70
60
50
Virgin mirror
Pure D2
40
30
fCH4=1.8%
20
fCH4=3.5%
10
0
500
1000
1500
2000
2500
Wavelength (nm)
Degradation of the reflectivity due
to absorption of light in the
deposited layer
G. De Temmerman
Specular reflectivity (%)
Specular reflectivity (%)
• Reflectivity measured with a UV-Vis-NIR spectrophotometer equipped with an
integrating sphere
Copper
Degradation of the reflectivity due
to an increase of the surface
roughness
ITPA 10 meeting, Moscow, April 06
7
Reflectivity of linearly polarized light
• Measurement of the reflectivity of linearly polarized light using a spectroscopic
elipsometer at various incidence angles.
• Rs: E field perpendicular to the plane of incidence
• Rp: E field parallel to the place of incidence
• Wavelength range: 350-2300 nm
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
8
Reflection of polarized light (400 nm)
Stainless steel
2.8
Virgin mirror
D2 pure
2.4
fCH4=1.8%
Copper
3.2
2.8
2.4
2.0
400 nm
stainless steel
1.6
Rs/Rp
fCH4=3.5%
Rs/Rp
400 nm
Copper
2.0
1.6
Virgin mirror
D2 pure
1.2
0.8
1.2
fCH4=1.8%
0.4
0.8
fCH4=3.5%
0.0
40
50
60
70
Incidence angle (°)
G. De Temmerman
80
40
50
60
70
80
Incidence angle (°)
ITPA 10 meeting, Moscow, April 06
9
Reflection of polarized light (800 nm)
Stainless steel
Copper
3.6
1.2
3.2
Virgin mirror
D2 pure
2.8
fCH4=1.8%
0.8
2.0
800 nm
stainless steel
1.6
Rs/Rp
fCH4=3.5%
2.4
Rs/Rp
1.0
0.6
Virgin mirror
D2 pure
0.4
fCH4=3.5%
0.2
1.2
0.8
800 nm
Copper
fCH4=1.8%
0.0
40
50
60
70
80
Incidence angle (°)
40
50
60
70
80
Incidence angle (°)
Polarization of the light strongly
affected by the carbon layer.
A drastic increase of the surface
roughness has only a slight effect
on the polarization
Deposition of impurities appears to be a more serious problem for
diagnostics using polarized light
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
10
Exposure of mirrors in TCV (1)
• TCV (Lausanne), 90% carbon coverage of the first wall
Mirrors located in the divertor region and recessed
below the surface of divertor tiles, no direct contact
with the plasma
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
11
Exposure of mirrors in TCV (2)
• No shutter installed, the sample manipulator is electrically insulated from the
torus
• Sample exposures integrated over short experimental campaign periods of few
weeks including He glow discharge conditioning
Magnetic equilibrium of the standard single
null diverted discharge. The red arrow
indicates the mirror location.
• Mirrors exposed to a variety of diverted plasma configurations (many plasma
configurations can be achieved at TCV
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
12
Deposition efficiency
• Test of different materials and different distances
Experiment Material
1
2
3
4
5
Mo
Mo/W
Mo/W
Mo
Si
Mo
Si
Distance below
the tile surface
(mm)
15
10
50
323
19
214
Glow
discharge
(hrs)
33.44
1.47
21.54
50
223
24.5
50
820
90.5
Number
of shots
Deposited
thickness
(nm)
4.7
1
0.85
1.3
15.89
4
24
Thickness determined by ellipsometry/SIMS/ profilometry
Deposited layer consists mainly of carbon and deuterium
Strong differences in the thickness measured on Si and Mo
samples under similar exposure conditions
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
13
Summary/ Conclusions
• Both laboratory experiments and sample exposures in the TCV tokamak have
shown the material dependence of the erosion/deposition patterns affecting the
reflectivity of metallic mirrors (with carbon as impurity)
• Monte Carlo simulation (SDTRIMSP) have confirmed the differences observed for
the various substrates (not shown here)
• These different features are only of importance until a certain deposited thickness is
reached (after this the deposition rate on the various metals is the same)
• Further experiments are needed to test other materials
The material choice not only influences the resistance of mirrors
towards erosion but also their sensitivity to impurity deposition
G. De Temmerman
ITPA 10 meeting, Moscow, April 06
14