Design of a microwave low-temperature plasma reactor used for
simulating plasma interactions with mixed materials targets
G. Lombardi, L. Colina Delacqua, M. Redolfi, D. Vrel, A. Michau, K. Hassouni, C. Arnas, X. Bonnin
Laboratoire des Sciences des Procédés et des Matériaux (LSPM), UPR 3407 CNRS, Université Paris 13, 93430
Villetaneuse, France
Abstract: We have developed a low-temperature plasma reactor to simulate some
of the plasma/surface processes occurring under the divertor dome of tokamaks,
with an emphasis on mixed materials targets and dust production. We wish to
address issues related to the chemistry of erosion products, along with transport,
and redeposition in parasitic plasma environments, as expected in ITER. We
detail the design steps to build the plasma source using a new multi-bipolar ECR
source technology, with plasma ignition through pencil sources arranged in a
circle, providing for an elevated electron temperature and sustained plasma
density. Two erosion targets, located above and below the sources, are exposed
to the plasma. These targets types are considered: 1/ Pie-shaped multiple sectors
of single material (C, W, and/or a Be-like element); 2/ Single sector made up of a
suitable alloy; 3/ A tungsten or carbon blank on which powder samples of mixed
materials have been deposited. These latter samples are to be obtained by
mechano-synthesis (stoichiometric compositions out of chemical equilibrium).
Keywords: plasma-surface interaction, carbon, dust, ECR sources
1. Introduction
An ITER divertor dome simulator, the
CASIMIR II (Chemical Ablation, Sputtering,
Ionization, Multi-wall Interaction and Redeposition) low temperature high density
plasma reactor, has been developed [1]. It
reproduces some of the features of the ITER
scrape-off layer (SOL) plasma and, more
accurately, the secondary parasitic discharges
expected to occur in remote areas of a tokamak
[2] [3], in particular in the space under the
hollow divertor dome leading to the pumping
ducts. The objective is to explore and
understand effects due to preferential sputtering
and material migration of first carbon materials
targets then mixed material (carbon/tungsten)
exposed to an hydrogen plasma, as well as
characterization of the deposits and dust that
such exposure may create.
2. Description of the experimental set-up and
diagnostics
2.1. Reactor simulating dust production
under the divertor dome
With respect to our former device CASIMIR I,
the CASIMIR II reactor presents strongly
improved potentialities in terms of particle
density: we increased the electron density ne
from ne ~ 1.3×1010 cm-3 to ~ 2×1011 cm-3 in
hydrogen plasma. This was possible by making
use of new generation of multi-dipolar ECR
(Electron Cyclotronic Resonance at 875 Gauss)
sources, representing a technological breakdown
in the microwave technology [4]. This new
generation of low temperature high density
plasma sources represents a versatile alternative
to other technologies such as ICP or Helicon
sources [5]. Electron density measured in
CASIMIR II is in correct agreement to those
obtained on parasitic plasma under the divertor
roof baffle of ASDEX-Upgrade (ne ~1010 cm-3)
[2]. One of the aims of the paper is to describe
the potentialities of the apparatus through
preliminary studies on carbon material.
A schematic drawing of the device is given in
figure 1. The reactive chamber (Volume ~ 3 L)
is filled with different gases: H2 (or D2) to
obtain the chemical etching of a target by H (or
D) atoms [6] [7]. The working pressure ranges
from 1 to 10 Pa. The plasma is ignited and
maintained by 16 multi-bipolar ECR sources
[4]. A 2.45 GHz microwave generator is used to
supply 3kW to the ECR sources (about 190W is
coupled per source). These sources are arranged
in a circle providing an elevated electron
temperature and sustained plasma density and
confinement. Two carbon (or tungsten) targets
are exposed to the plasma, and located above
and below the sources. These targets can be
negatively biased (up to - 1000V) to increase the
energy of the ions impacting the surface and
thus favouring the physical erosion.
Figure 1. Schematic of the CASIMIR II reactor
2.2. Diagnostics
UV and visible Optical Emission Spectroscopy
(OES) spectra are performed by using a high
resolution Jobin-Yvon spectrometer (1-m focal
length) to detect excited species present in the
discharge, the detection being insured by a
photomultiplier. The light is collected through a
quartz window placed instead of one ECR
source. Heavy neutral and charged species are
analysed using a plasma monitoring system
Hiden® EQP-500. This device allow mass
detection up to a ratio m/z = 500. Neutral
species, radicals, as well as negative and
positive ions can be monitored. The extraction
of the gas sample is performed in-situ, just in
the middle of the etching chamber. Scanning
electron microscopy micrographs are taken from
the carbon targets to perform a morphological
characterization before and after discharge.
Infrared spectroscopy analyses are performed
with a Shimadzu IR Prestige 21 spectrometer.
The wavenumbers measuring range varies from
3500 to 500 cm-1 with a resolution range of 4
cm-1. A 20 m-multi-pass absorption cell (model
Tornado T20 manufactured by Eurolabo) is
coupled with FTIR spectrometer for gas phase
analysis and Attenuation Total Reflectance
system is used for the measurement of solid
materials.
3. Results
3.1. Morphological characterization of
carbon target surface
Two carbon targets were exposed in CASIMIR
II during 60 min to a low temperature high
density (ne ~ 2×1011 cm-3 measured by
Langmuir probes) D2 plasma ignited and
maintained at 3 kW, and at 1 Pa.
After discharge an ex-situ morphological
characterization
was
performed.
SEM
micrographs were taken from the carbon targets
before and after discharge. Comparing the two
carbon surface states (figure 2), we can observe
changes in surface morphology: the surface had
been eroded and fiber shape deposits have been
formed. The erosion suffered by the carbon
target can be physical, chemical or both. To
determine this, a chemical analysis of the gas
phase was performed.
Figure 2. SEM micrographs showing nanoparticles
aggregates on the target surface
3.2. Chemical analysis of gas phase
Mass spectrometer measurements have been
performed all along the experiment. Only after a
few minutes, neutral carbon containing species
(such as C2, CD, C2D2, C3D3...) have been
detected. To confirm that these species are
products of erosion, the mass spectrum,
obtained during the discharge, was confronted
to a mass spectrum of the gas phase without
plasma (figure 3). The presence of both species
containing only carbon (C, C2) and species
containing carbon and deuterium (CD, C2D2) let
us assume that the carbon target had suffered
both physical and chemical erosion.
plasma off
plasma on
100000
C2D2
C2D4
I (c/s)
10000
CD
C
1000
100
0
5
in the UV domain was performed: between 420
nm and 430 nm the Gerö band spectrum of the
CD molecule was observed. Moreover Doppler
broadening measurements [8] on D atom
Balmer-α lines (Dα at 656.1 nm, Dβ at 486 nm)
were performed to estimate temperature of the
excited deuterium species, together with the
rotational distribution of the Fülcher-α (d3Πu –
a3Σg+) molecular deuterium band (Q-branch, 0-0
transition from 601 to 606.5 nm, 1-1 transition
from 612 to 616.5 nm) [9] and the
corresponding Boltzmann plot to estimate the
rotational molecular temperature. Thus we
found at working conditions (3kW, 1Pa) TD ~
2000 K and Trot ~ 700K.
This chemical analysis highlights the presence
of lights hydrocarbon species in the plasma
phase. At this level three scenarios can be
envisaged, the hydrocarbon species formed: 1)
will not interact with the plasma and be simply
pumped; 2) will suffer in volume chemical,
physical and transport processes that can lead to
the synthesis of soot and their redeposition after;
3) will be redeposited and then grow on surface
of the carbon target.
3.3. Deposit/dust formation
When opening the reactor after discharge, a
large formation of brown deposits and flakes on
the walls of the etching chamber as on the ECR
source magnets was observed (figure 4).
C3D3
C2
10 15 20 25 30 35 40 45 50 55
m/z (amu)
Figure 3: Mass spectra obtained in etching chamber of
the CASIMIR II reactor
To further confirm the presence of carbonaceous
compounds in the gas phase, an OES spectrum
Figure 4: Picture of brown deposit formed in CASIMIR
II.
Infrared analysis was performed to identify the
structure and chemical functions present in the
deposit. FTIR spectrum given in figure 5 allows
highlighting the presence of some characteristic
bands such as those corresponding to C-C bonds
(1050 cm-1) and Csp3-H (2860 cm-1 and 2920
cm-1) characteristics of saturated aliphatic
molecules. This analysis allows to exclude the
presence of the C=C and Csp2-H characteristics
of unsaturated aliphatic molecules or molecules
containing aromatic rings. The CN vibration
bands (1060 cm-1 and 1630 cm-1) correspond to
amine or amide molecules (Nitrogen atom may
come from impurities in the feed gas).
Transmittance (%)
100
CN
98
Csp3-H
CN
96
3500
C-C
3000
2500 2000 1500
-1
Wavenumber (cm )
1000
Figure 5: FTIR spectrum of the deposit
4. Conclusion
In this paper, a low temperature high density
plasma reactor, envisioned as an ITER dome
simulator has been described. Plasma
characteristics measured have showed that
CASIMIR II reactor is able to generate plasma
flow in accordance with the plasma
characteristics simulated for parasitic discharges
under the divertor roof baffle of tokamaks.
Studies, in this new and improved reactor, have
just begun and first experiments were performed
in a D2 plasma in contact of a carbon target. Our
final goal is to study the full lifecycle of
codeposit coming from mixed materials targets
C/W, including
techniques
proposal
of
abatement
References
[1] G. Lombardi, et al., J. Nucl. Mater. 390-391,
(2009) 196.
[2] V. Rohde, M. Mayer, ASDEX Upgrade
Team, J. Nucl. Mater. 313-316 (2003) 337.
[3] K. Matyash, R. Schneider, X. Bonnin, D.
Coster, V. Rohde, H. Kersten, J. Nucl. Mater.
337-339 (2005) 237.
[4] T.V. Tran, Ph.D. Thesis, LPSC, Université
de Grenoble (2006).
[5] Oleg A. Popov, High density plasma
sources, (1995)
[6] P. S. Krstic, C. O. Reinhold, S. J. Stuart,
New J. Phys. 9 (2007) 209.
[7] J. Zhou, I. T. Martin, R. Ayers, E. Adams,
D. Liu, E. R. Fisher, Plasma Sources Sci.
Technol. 15 (2006) 714.
[8] B. Xiao, B. Wan, J. Huang, J. Nucl. Mater.
363-365 (2007) 849.
[9] S. Brezinsek, Ph. Mertens, A. Pospieszczyk,
G. Sergienko, P. T. Greenland, Contrib. Plasma
Phys. 42 (2002) 668.
Acknowledgements
This work is supported by the ANR (French
National
Agency
for
Research)
contracts
JC05_42075 and ANR-09-BLAN-0070-01, the
Fusion Federation contract V3580.001 and
Actions FR-FCM WP10-PWI-02-02 et WP10PWI-06-01