Pilot experiments for the International Thermonuclear Experimental
Reactor active beam spectroscopy diagnostic
M. von Hellermann,
a)
M. de Bock, R. Jaspers, and K. Jakubowska
FOM-Institute for Plasma physics “Rijnhuizen,” Association EURATOM, Trilateral Euregio Cluster, 3430 BE
Nieuwegein, NL
R. Barnsley, C. Giroud, N. C. Hawkes, and K. D. Zastrow
UKAEA Culham Laboratory Euratom Association, Abingdon, United Kingdom
P. Lotte and R. Giannella
CEA Cadarache, Association Euratom, France
A. Malaquias
IAEA, Vienna, Austria
E. Rachlew
Euratom Association-VR, KTH Stockholm, Sweden
S. Tugarinov and A. Krasilnikov
TRINITI Troitsk, Russia
A. Litnovsky, V. Philipps, and P. Wienhold
Institute for Plasma Physics, FZ-Juelich, Euratom Association, 52425 Juelich, Germany
P. Oelhafen and G. De Temmerman
Institut für Physik, University of Basel, 4056 Basel, Switzerland
L. Shmaenok
Phystex, Vaals, NL and loffe Physical Technical Institute, St. Peterburg, Russia
(Presented on 19 April 2004; published 1 October 2004)
Supporting pilot experiments and activities which are currently considered or already performed for
the development of the International Thermonuclear Experiment Reactor active beam spectroscopy
diagnostic are addressed in this article. Four key issues are presented including optimization of
spectral instrumentation, feasibility of a motional Stark effect (MSE) evaluation based on line ratios,
“first-mirror” test-bed experiments at the tokamak TEXTOR, and finally the role of integrated data
analysis for the conceptual layout of the change exchange recombination spectroscopy and MSE
diagnostic. © 2004 American Institute of Physics. [DOI: 10.1063/1.1787950]
I. INTRODUCTION
A comprehensive package of active beam based spec-
troscopy tools for the International Thermonuclear Experi-
mental Reactor (ITER) has been developed and a final sum-
mary was recently completed.
1
The package encompasses
charge exchange recombination spectroscopy (CXRS)
2,3
for
the measurement of the main impurity ion densities (includ-
ing helium ash), ion temperature and torodial as well as po-
loidal plasma rotation. Quantitative use of beam emission
spectroscopy (BES)
4
based on comprehensive atomic model-
ing is proposed as an indispensable cross-calibration tool for
absolute local impurity density measurements
5
and monitor-
ing of the neutral beam power deposition profile. Finally, a
full exploitation of the motional Stark effect (MSE) pattern is
proposed to deduce local pitch angles, total magnetic fields
6,7
and radial electric fields. CXRS on slowing-down alphas as
demonstrated. on the Tokamak Fusion Test Reactor
8
is also
considered for the ITER DNB and HNB (100 keV/amu,
2.2 MW diagnostic neutral beam and 500 keV/amu, 17 MW
heating neutral beam respectively) and was discussed
recently.
2
More quantitative studies and modeling will be
required in this field. A similar diagnostic challenge refers to
the measurement of slowing-down beam ions as produced by
the HNB. Medium energy velocity distribution functions rep-
resenting slowing-down helium beam ions were observed at
the Joint European Tours (JET).
9
A number of issues have
been recognized over the last years which need to be ad-
dressed in advance of the next stage, the diagnostic “procure-
ment phase.” Four of those are addressed in this article.
(1) The optimization of spectral instruments and detectors
for the active beam spectroscopy diagnostic package
needs to be tested on existing fusion devices in order to
establish suitable combinations for high-resolution and
broadband applications. Moreover, multi-tasking recom-
mends practical maintainable instrumental solutions.
(2) A converging interest in motional Stark diagnostics,
beam emission spectroscopy, and charge exchange spec-
a)
Author to whom correspondence should be addressed; electronic mail:
mgvh@rijnh.nl
REVIEW OF SCIENTIFIC INSTRUMENTS VOLUME 75, NUMBER 10 OCTOBER 2004
0034-6748/2004/75(10)/3458/4/$22.00 3458 © 2004 American Institute of Physics
Downloaded 02 Jan 2007 to 134.94.122.39. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp
troscopy has highlighted the needs for characterization
of the polarization transfer properties of optical
periscopes.
(3) A central issue is the survival and protection of the first
mirrors used in observation periscopes. The effects of
carbon and beryllium coatings on reflectivity and polar-
ization characteristics of metallic mirrors have been
modeled and presented at several meetings (e.g., Ref. 7).
First results describing erosion effects of monocrystal-
line molybdenum mirrors were reported recently.
10
(4) Finally, data evaluation procedures are closely linked to
feasibility studies and will ultimately decide on achiev-
able performances.
II. CXRS AND BES INSTRUMENTATION
Very early on it has been recognized that the substantial
attenuation of the diagnostic beam requires high-optical
throughput spectrometers in order to compensate for a low
CX signal in the presence of huge background of continuum
radiation. For the case of a modulated DNB the effective
signal to noise is then determined by the ratio of CX signal
and fluctuation of the background. A crucial criterion for the
acceptance of CXRS as a viable tool for ITER is the capa-
bility of measuring local helium ash densities in the plasma
core. At TRINITI a high-optical-throughput, high-resolution
spectrometer was developed especially for CXRS
application.
11
A unique combination of a large f-number
共F/3兲,f =480 mm, and high linear dispersion value 共D
=0.25–0.3 nm/mm兲 make it a particularly suited instrument
for the envisaged ITER CXRS system. The system’s etendue
共A·⍀ =4⫻ 10
−1
mm
2
sr兲 is preserved by the periscope fiber
bundle assembly, and a DNB slab of about 100⫻ 100 mm is
imaged for each radial channel onto the entrance slit 共1
⫻ 5mm兲 of individual spectrometers. The role of the CXRS/
BES diagnostic as a multi-tasking requires the simultaneous
measurement of the main ion species: helium ash at 468 nm,
intrinsic impurities C and Ne at 529 nm, and bulk D and T
ions at 650 nm (cf. Fig. 1). The proposed scheme is to make
use of color filters and different orders of the TRINITI ech-
elle grating instruments for different representative wave-
lengths. Pilot experiments are required addressing achievable
sensitivities, stray-light suppression, and color control. A
critical issue will also be the spectral purity of the underlying
continuum radiation and its quantum noise photon statistics
in order to verify predicted spectral signal-to-noise reactor
(SNR) values (cf Ref. 1). The detection of nonthermal broad-
band features depends critically on clear distinction of CX
induced features and passive background.
III. MSE MEASUREMENTS BY RATIOMETRY
The combination of CXRS and beam emission
spectroscopy
5
is seen today as the sole path to obtain abso-
lute ion densities (helium ash) on ITER. BES refers here
explicitly to a quantitative exploitation of measured and
modeled line intensities and wavelength separations and not
to the measurement of plasma fluctuations. A natural spin-off
of BES on the DNB is to exploit the D-alpha spectrum and
its amplitudes for magnetic diagnosis. This will be comple-
mentary to MSE polarimetry on the HNB.
6,12,13
In fact, the
complementary exploitation of the entire MSE polarization
pattern will ultimately optimize signal accuracy and spatial
resolution. Whereas the viewing geometry of the HNB MSE
periscopes is optimized for a maximum sensitivity to a rota-
tion of the polarization pattern, the DNB periscopes will only
be sensitive to changes of the angle between Lorentz-field
vector and viewing line. For the ITER magnetic field B
t
=5.2 T, and a DNB energy of 100 keV/amu, the MSE com-
ponents (6
and 3
components, respectively) will be well
separated 共⬎5Å兲 and the SNR compared to the background
fluctuation level (assuming photon statistics of continuum
radiation) should be in excess of 50 (see also Refs. 1 and 2).
The pitch angle B
p
/B
t
derived from the dipole intensity
ratio of
and
group in MSE spectrum, which, in the case
of statistical population, is determined by the angle
be-
tween direction of l.o.s. and Lorentz-vector E
L
=
v
⫻ B
I
I
=
sin
2
1 + cos
2
.
Maximum sensitivity for observation angles
close to 45°.
For the ITER CXRS top-port periscope the angle
on axis
共B
p
=0兲 is 32.6°. Figure 2 shows q values derived from ratio
I
/I
for the case of a simplified circular magnetic flux sur-
face and the ITER CXRS top-port periscope. In contrast to
polarimetry, no active or passive polarizing element is in-
volved in ratiometry. However, a key issue will be isotropic
reflection factors ensuring the preservation of I
/I
for the
entire observation system including multiple mirrors in the
periscope, the fiber link, and finally in the spectrometer it-
self. A MSE ratiometry pilot experiment is presently imple-
mented on TEXTOR in ITER-like geometry.
IV. FIRST-MIRROR EXPERIMENTS ON TEXTOR
Metallic mirrors which are foreseen for ITER optical
periscopes were recently investigated on TEXTOR. The aim
of the experiments was to investigate optical properties of
plasma-facing mirrors in dependence on erosion, deposition,
and particle implantation.
FIG. 1. Schematic layout of CXRS/
BES/MSE spectroscopic instrumenta-
tion. Multi-tasking of periscopes for
simultaneous evaluation of CXRS
(impurity and bulk ions), BES on
injected neutrals and finally for MSE
ratiometry.
Rev. Sci. Instrum., Vol. 75, No. 10, October 2004 Plasma diagnostics 3459
Downloaded 02 Jan 2007 to 134.94.122.39. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp
Large polycrystalline metal mirrors (molybdenum and
tungsten) on inclined target holders were exposed in the
scrape-off layer (SOL) plasma of TEXTOR in both erosion
and deposition-dominated zones. The accumulated plasma
exposure time of mirrors was about 900 s. Particle energies
and particle affluence were of the same order of magnitude
as those expected for ITER. Figure 3 shows a schematic
layout of the mirror test and one sample exposed in the
deposition-dominated zone. The reflectivity was measured
before and after exposure in the SOL. The wavelength re-
gime of measurements ranged from 250 to 2500 nm. First
experimental evaluations
14
can be summarized as:
(1) after plasma exposure the reflectivity was decreased by
up to 35% in the deposition areas. The reflectivity was
increased by 12% in the plasma near zones (erosion
zones);
(2) fringes due to constructive and destructive interference
are observed if /4 corresponds to the optical thickness
of a deposited film;
(3) the degradation of reflectivity of the exposed mirror
samples is likely due to film growth observed on the
mirror surface; and
(4) the increase of reflectivity of mirrors noticed on some
areas is possibly due to annealing effects during the tem-
perature excursions and/or cleaning by plasma ions.
V. ACTIVE BEAM SPECTROSCOPY DATA ANALYSIS:
CHEAP FOR ITER
A collateral activity to ongoing conceptual design and
feasibility studies is dedicated to the development of suitable
data analysis packages including spectral analysis and phys-
ics evaluation. At JET the self-consistent CX evaluation
package CHEAP (charge exchange analysis package) is
used routinely and is potentially applicable in inter-shot
performance.
The CHEAP concept is based on a comprehensive
atomic modeling
15
of all emission processes (beam emission,
active and passive charge exchange recombination) making
use of a full set of data for the plasma environment. An
essential first step is the introduction of a common mapping
grid for all physical parameters. This strategy has enabled us
to treat the beam-plasma interaction (i.e., beam stopping and
beam emission processes) in a consistent fashion. Multiple
step processes and contributions from excited states are taken
into account. For ITER the concept of integrated data CX
analysis implies a comprehensive diagnostic coverage of the
main plasma ions (intrinsic and seeded impurities and bulk
ions). From this point of view, the issue of CX measurement
of helium ash densities on ITER is inseparable from a simul-
taneous measurement of all ions. Moreover, the issue of a
density measurement is linked directly to a measurement of
ion temperature and plasma rotation.
Highlights of future activities in this field are listed
below:
(1) Super Fit: The present strategy of spectral analysis is
usually based on a step-by-step process addressing each
spectrum individually. A new concept is to involve a full
set of spectra representing, for example, a complete ion
temperature profile and solve all spectra in one go.
(2) Integrated Data Analysis: From a similar point of view, a
spectral analysis is ultimately to be seen as part of a
global data consistency strategy. Ion temperature and
density lead to ion pressure and combined with electron
pressure to a match of kinetic and diamagnetic energy
which can be used as a constraint on spectral analysis.
(3) The stochastic occurrence of plasma edge line emission,
e.g., edge localized modes ELMs is a potential hazard to
a perfect background suppression scheme using beam
modulation. In this case, complex spectra will have to be
evaluated requiring advanced atomic modeling and suit-
able extraction tools.
1
M. von Hellermann, “Active Charge Exchange Spectroscopy (CXRS) and
Beam Emission Spectroscopy 共BES+MSE兲 with Diagnostic Neutral
Beam (DNB)” (under EFDA contract; not publicly available, contact:
mgvh@rijnh.nl)
2
M. von Hellermann, C. Giroud, N. C. Hawkes, R. Jaspers, A. Krasilnikov,
P. Lotte, G. McKee, A. Malaquias, M. O’. Mullane, E. Rachlew, S.
FIG. 2. Safety factor q as derived from MSE intensity ratio for different
minor radii and ITER top-portconfiguration. Note, a simplified case off cir-
cular flux geometry has been assumed for illustration purposes.
FIG. 3. (Color) Large polycrystalline metal mirrors (molybdenum and tung-
sten) on inclined target holders were exposed in the SOL plasma of TEX-
TOR in both erosion and deposition-dominated zones.
3460 Rev. Sci. Instrum., Vol. 75, No. 10, October 2004 von Hellermann
et al.
Downloaded 02 Jan 2007 to 134.94.122.39. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp
Tugarinov, and K.-D. Zastrow, 30th EPS Conf. Contr. Fus. Pl. Phys., St.
Petersburg, 2003, ECA Vol. 27A, O-4.2D.
3
A. Malaquias, A. Costley, A. Gorshkov, N. Hawkes, M. v. Hellermann,
M. Kuldkepp, P. Lotte, E. Rachlew, S. Tugarinov, G. Vayakis, and C.
Walker, 15th Topical Conference on High Temperature Plasma Diagnos-
tics, San Diego, 19 April 2004.
4
W. Mandl, R. Wolf, M. von Hellermann, and H. P. Summers, Plasma Phys.
Controlled Fusion 35, 1373 (1993).
5
M. von Hellermann, R. Jaspers, H. P. Summers, and K-D. Zastrow, Ad-
vanced Diagnostics for Magnetic and Inertial Fusion, edited by P. E.
Stott, A. Wootton, G. Gorini, E. Sindoni, and D. Batani (Plenum, New
York, 2001),p.205.
6
P. Lotte, M. Von Hellermann, A. Malaquias, R. Giannella, N. Hawkes, E.
Rachlew, R. Jaspers, P. Nielsen, S. Tugarinov, and C. Walker, in Ref. 2,
P1–183.
7
A. Malaquias, M. von Hellermann, P. Lotte, S. Tugarinov, and V. Voit-
senya, in Ref.2 .
8
G. McKee, R. J. Fonck, B. Stratton, R. Budny, Z. Chang, and A. Ramsey,
Nucl. Fusion 37, 501 (1997).
9
M. von Hellermann, W. G. F. Core, J. Frieling, L. D. Horton, R. W. T.
König, W. Mandl, and H. P. Summers, Plasma Phys. Controlled Fusion
35, 799 (1993).
10
K. Vukolov and V. Voitsenya, Sixth ITPA Topical Group Meeting on ITER
Diagnostics, Part I, Naka, Japan, February 2004.
11
S. Tugarinov, A. Krasilnikov, V. Dokouka, R. Khayrutidinov, I. Beigman,
I. Tolstikhina, L. Vainshtein, M. von Hellermann, and A. Malquias, Rev.
Sci. Instrum. 74, 2075 (2003).
12
N. C. Hawkes, A. Malaquias, P. Lotte, M. von Hellermann, M. Brix, R.
Giannella, M. Kuldkepp, E. Rachlew, C. Negus, and E. Surrey, Fifth ITPA
Topical Group Meeting on ITER Diagnostics, St. Petersburg, Russia, 16
July 2003.
13
F. Levinton, R. Fonck, G. Gammel, R. Kaita, H. Kugel, E. Powell, and D.
Roberts, Phys. Rev. Lett. 63, 2060 (1989).
14
P. Wienhold, A. Litnovsky, V. Philipps, P. Oelhafen, G. De Temmerman,
W. Schneider, and B. Emmoth. (unpublished).
15
http://adas.phys.strath.ac.uk..
Rev. Sci. Instrum., Vol. 75, No. 10, October 2004 Plasma diagnostics 3461
Downloaded 02 Jan 2007 to 134.94.122.39. Redistribution subject to AIP license or copyright, see http://rsi.aip.org/rsi/copyright.jsp