HAL Id: jpa-00227574
https://hal.archives-ouvertes.fr/jpa-00227574
Submitted on 1 Jan 1988
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
HIGH RESOLUTION X-RAY SPECTROSCOPY AT JET
R. Giannella
To cite this version:
R. Giannella. HIGH RESOLUTION X-RAY SPECTROSCOPY AT JET. Journal de Physique Col-
loques, 1988, 49 (C1), pp.C1-283-C1-291. �10.1051/jphyscol:1988160�. �jpa-00227574�
JOURNAL DE PHYSIQUE
C O l l O q u e C 1 , S u p p l 6 m e n t a u n 0 3 , Tome 4 9 , Mars 1 9 8 8
HIGH RESOLUTION X-RAY SPECTROSCOPY AT JET
R. GIANNELLA
JET J o i n t U n d e r t a k i n g , GB-Abingdon OX14 3DB, Oxon, G r e a t - B r i t a i n
Abstract
A high resolution X-ray crystal spectrometer w a s installed a t J E T i n 1986. T h e m a i n t a s k of t h i s i n s t r u m e n t , designed a n d built by t h e E N E A Frascati Laboratory in collaboration with J E T , is t h e measurement of t h e ion temperature a n d rotation velocity from Doppler b r o a d e n i n g s a n d shifts. S t r i n g e n t operation r e q u i r e m e n t s considerably conditioned i t s versatility a s a survey i n s t r u m e n t or a tool for fundamental research. The uni ue conditions of t h e J E T plasmas, however, allowed to observe in high resolution a n d identiYy lines from t h e highest ionization s t a g e s (B-like to H-like) of nickel i n t h e vicinity of t h e resonance transitions of t h e one- a n d two-electron systems of this element. These spectra a n d t h e i r relevance to t h e study of J E T physics a r e briefly presented. T h e comparison with atomic physics theoretical computations of He-like a n d H-like ions spectra observed a t J E T i s also shortly presented. T h e most intense a m o n g these lines could regularly be monitored with adequate time resolution a s to allow t h e production of a n extensive d a t a base. T h e experimental spectra a r e simulated by i n t e g r a t i n g t h e theoretical ones along t h e spectrograph line of s i g h t across t h e plasma. Ion t e m p e r a t u r e a s well a s rotation velocity measurements in ohmic plasmas a n d additionally h e a t e d o n e s a r e discussed a n d t h e i r r e l a t i o n s h i p w i t h o t h e r p h y s i c a l q u a n t i t i e s a n d operational parameters a r e shortly reviewed. Details of t h e spectrometer design, i t s lay-out on t h e J E T s i t e a n d i t s operational capabilities a r e also given.
1. INTRODUCTION
Diagnostics of t h e ion distribution functions i n the present generation t o k a m a k devices requires t h e use of several approaches based on a large variety of'experimental techniques a n d relying on a n u m b e r of different physical assumptions. In fact t h e s i m u l t a n e o u s use of different methodologies allows to cross check t h e various results a n d to extend t h e accessible plasma p a r a m e t e r ranges. Also t h e results of o t h e r diagnostics m e a s u r i n g different plasma p a r a m e t e r s a n d a n extensive use of numerical simulations a r e often needed in t h e process of d a t a validation.
Among these techniques, highly valuable a r e t h e spectroscopic o n e s a n a l y s i n g t h e Doppler broadenings a n d s h i f t s of emission l i n e s [I-21. T h e b e n t crystal s p e c t r o m e t e r s o p e r a t i n g i n t h e soft X-rays display very good resolution a n d luminosity performances i n t h i s spectral domain typical of t h e line emission from t h e central hottest regions of these multi-keV plasmas. Furthermore the peculiar simple inverse dependence of t h e t h r o u g h p u t of t h e s e i n s t r u m e n t s on t h e i r focal length allows to devise r a t h e r unconventional large scale lay-outs such a s to meet t h e needs of complicated a n d often hostile operation conditions [3].
'I'he m a i n objective of t h e high resolution ( N A A = 20000) X-ray crystal spectrometer installed a t J E T i n 1986, a 25 metre curvature radius J o h a n n type i n s t r u m e n t designed a n d b u i l t by t h e E N E A Frascati fusion research centre, i s not atomic physics investigations or s t u d i e s of pollution of t h a t plasma by impurities, but diagnostics of ion populations reliably performed on a routine basis a n d for different experimental conditions. T h i s is required both i n t h e present operation phase a n d in t h e planned deuterium-tritium (D-T) one.
In t h i s note a s h o r t description of t h e experimental conditions a s well a s details of t h e i n s t r u m e n t a n d i t s lay-out a r e given i n Secticn 2. In Section 3 some of t h e observed spectra, a n d t h e related simulation studies, a r e shortly visited. Finally i n Section 4 some of t h e r e s u l t s from t h e velocity measurements a n d t h e related phenomenology a r e uutlined.
(l'permanent address : Associazione EIIRATOM-ENE4 sulla Pusione. CRE Frascati. CP 65. 1-00044 Frascsti.
Roma. ltalie
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1988160
C 1-284 JOURNAL DE PHYSIQUE
2. EXPERIMENTAL CONDITIONS
J E T (Joint European Torus) is the tokamak producing the largest reported laboratory plasmas; its parameters are believed to be the closest realized so far to those of t h e future fusion reactor [4]. The main objective of this device is to further improve its performance to reach a regime where i t should be possible, in a deuterium-tritium plasma, to s t u d y t h e confinement of the thermonuclear alpha-particles produced and their contribution to plasma heating.
J E T a n n u l a r discharges are characterized by major radii ran ing between 2.5 and 3.4 m
Q
and circularorD shaped crosssectionsofarea between
-
2 and-
8 m .Electron peakdensitiesin the range 1.5 to 6.1019 m-3, electron and ion peak temperatures in the range 1.8 to 7 keV and 1.5 to 15 keV respectively, plasma currents between 1 and 5 MA a n d toroidal magnetic fields in t h e range 1.7 to 3.5 T are the main arameters of most of the plasmas studied, under different operating conditions, in the first Pour years of operation.Besides the ohmic heating of the plasma due to the electric current, additional heatin i s supplied by several megawatts of electromagnetic waves i n the ion cyclotron range of t e - quency [51. Alternatively or in addition to the radio-frequency waves, beams of energetic ( - 8 0 keV) neutral hydrogen or deuterium atoms are injected into the plasma to increase i t s energy content 161. Reference will be made in the following to the influence of these heating systems
upon the rotation of the plasma about the torus axis of symmetry.
Several measures are taken a t J E T to minimize the lasma pollution with the strongly radiating metal ions. These measures include use of carion limiters on the plasma outer equatorial plane, coverage with protection carbon tiles of large areas on the Inconel vacuum vessel internal wall to avoid contact with the plasma, carbonization treatments [6] of all the surfaces inside t h e discharge chamber. As a consequence the main impurities in the present hydrogen or deuterium plasmas produced are the light ions of carbon and oxygen (normally about 170 of the electrons) [81 while the concentration of nickel (the main constituent of the Inconel alloy) is usually in the range between 10-5 and 5.10-4.
J E T i s a n intense source of neutrons from fusion reactions. Yields up to 3.1015 neutrons per s [ 4 ] , corresponding to peak fluxes of the order of 108 neutrons cm-2 s-1 in the tokamak hall, have been reported. These fluxes, already adverse to the functioning of photon detectors and requiring heavy shieldings of many equipments located in that hall, are expected to increase by some four orders of magnitude in the planned D-T operation phase, resulting i n t h e absolute impossibility to operate most of the diagnostic apparatuses in t h a t environment.
Therefore the analysing and detection sections of any spectrograph suitable for operation in t h a t phase have to be positioned in a n adequately shielded location outside the tokamak hall.
These requirements determined the design of the curved crystal X-ray spectrometer installed a t JET. Its purpose is the ion temperature and plasma rotati011 diagnostics through the analysis of line profiles and shifts in narrow spectral intervals about the resonance transitions of He-like and H-like ions. The lay out of this instrument i s schematically shown in fig. 1. The line of sight, lying in the tokamak equatorial plane, i s not directed along a major radius of the toroidal discharge chamber, but its distance from the torus axis is b = 182 cm resulting in a relative Doppler shift (AhlA)i) proportional to the angular velocity wq, of the emitting ions about the mentioned axis according to the relation
where c is the speed of light.
The bent crystal is mounted on a turntable inside a vacuum box ositioned 'ust outside a penetration in the three metre thick concrete wall t h a t encloses the toRamak hsli; i t views the plasma through a beryllium window 0.15 mm thick located a t one of the main tokamak ports.
The ray path is within a n evacuated 30 cm diameter tube connecting the window to t h e vacuum box. After difiaction from the crystal the X-rays reach the detector via another tube pivoted to a n axis nex~. to the crystal box to match the diffracted X-rays direction according to the Bragg angle fixed by the crystal orientation.
Main geometrical constraints imposed by this lay-out are the following:
source to crystal distance D = 20 m;
Bragg angle range H = 48.75" to 52.75';
.
window extension in the diffraction plane s = 20 cm;J E T
Crystal d i f t r o c r l n ~ ore0 120 x 50- TORUS HALL
DIAGNOSTIC HALL
Fig. 1 Schematic lay-out of the X-ray crystal spectrometer installed at JET. The line of sight through the plasma is also shown.
they imply the following instrument parameters:
crystal radius of curvature R = 25.25 m;
fractional linear dispersion (l/A) (dNdx) = 4-10-4 cm- 1 ; spectral range observable a t one time (ANA)p = 0.8%.
The detector, 9 cm high and 25 cm wide, is a multi-wire proportional chamber where each anode wire is connected to a separate amplification and counting chain, t h u s providing a high count rate (up to 250 kHz per wire) position sensitive photon detection. The anode spacing of 0.127 cm is the dominant resolution limiting factor of the whole instrument. The diffracting a r e a of the crystals is 5 cm high and 12 cm wide. A quartz crystal cut parallel to the planes 2243 h a s been used during the first year of operation of the spectrometer. This crystal h a s been recently substituted with a 440 germanium one in order to increase, d u e to t h e higher reflectivity of germanium, the overall luminosity of the instrument from
-
5.10- 13 m2 s t r to-
2.5.10-12 m2 str. Accordingly the time resolution for the ion temperature measurement from the Voigt fit of He-like Ni resonance line, rangin so far between about 20 ms and a few seconds, i s expected to improve by a factor of five, i f t h e concentration of t h a t impurity inside the plasma will not vary.The large focal length of this instrument easily allows a n accurate direct measurement of the Bragg angle within a few hundredths of a degree provided t h a t care is taken to have precise parallelism between the crystal surface and the atomic lanes. This resulted in a n estimated relative error on the absolute measured wavelengths ofabout 0.06% when also the o t h e r e r r o r contributions (such a s those due to the uncertainty on t h e crystal lattice parameters and the mechanical and encoding inaccuracies) are taken into account.
A more extensive description of this instrument is to be published in the near future.
Cl-286 JOURNAL DE PHYSIQUE
3. SPECTRA
Figure 2 shows a n observed spectrum including t h e resonance line of t h e He-like nickel.
I t w a s recorded by s e t t i n g t h e i n s t r u m e n t a t five different positions i n a sequence of repetitive dischar es. T h e central plasma parameters for these discharges were n, = 2.1019 m-3 a n d T,
= 3.1
$
0.3 keV. T h e wavelength scale, deduced by t h e i n s t r u m e n t setting, i s affected a s mentioned by a n offset error of a b o u t 1 mA, but the error upon t h e wavelength separation between p e a k s i s more t h a n ten times smaller. T h e d a t a i n t h e different segments h a v e been normalized with respect to each other by multiplication by factors r a n g i n g between 1.03 a n d 0.79. All t h e m a i n spectral features have been identified a s lines emitted from nickel ions withFig. 2 Experimental resonance spectrum of He-like nickel (boxes) and simulated spectrum (continuous line). The integration time was 9 s. The inset shows the Voigt profile fitting the w line experimental data.
two, three, four a n d five electrons. The Ni ion temperature a s determined from t h e Voigt fit of t h e resonance line i s 2.4 keV; t h e fit is shown in the figure inset.
I n fig. 3 t h e N i 13-like resonance spectrum i s shown. I t was acquired in two consecutive discharges with T, = 4.7 keV a n d n, = 1.5.1019 m-3. T h e two Ly-a lines clearly emerge from t h e c o n t i n u u m background: t h e i r intensity ratio is r = 0.6 a n d t h e ion temperature from t h e Voigt fit of Ly-a1 line is 4.0 keV.
I n figs 2 a n d 3 simulated spectra a r e overlapped to t h e observed ones. In t h e simulation t h e brigthness Bi, a s viewed by t h e spectrograph, for the i-th line m a k i n g u p t h e spectrum i s computed a s
w h e r e t h e integral is performed along t h e line of sight a n d c i s t h e emissivity depending upon t h e local plasma parameters. In doing this, use i s made of t h e r e s u l t s supplied by o t h e r diagnostics independently m e a s u r i n g t h e electron temperature [9] a n d density [ l o ] a t t h e different plasma positions. These quantities allow to calculate t h e relative abundancies of t h e different ionization s t a g e s a t t h e v a r i o u s positions according to t h e coronal ionization- recombination mechanisms a n d suitable modelling of t h e transport of these ions across t h e
Rg. 3 Experimental resonance spectrum of H-like nickel (boxes) and simulated spectrum (continuous line). The integration time was 10 s.
discharge. T h e width A appearing in t h e equation for t h e spectral brightness a n d t h e c e n t r a l a p p a r e n t wavelength A j i n t h e spectral shape function WA a r e also d e p e n d e n t upon t h e position through t h e local ion temperature a n d bulk velocity respectively. T h e bulk velocity of t h e ions i s assumed directed toroidally and can be s e t to v a r y with t h e torus major radius. For purely ohmically heated discharges like those referred to by figs 2 and 3 t h e rotation velocity i s v e r y s m a l l compared to t h e e r r o r o n t h e absolute wavelengths measure a n d i s therefore s e t to zero i n t h e simulations displayed.
A detailed comparison of t h e experimental spectra with theoretical atomic d a t a i s t h e object of a study t h a t will be published in the literature. Here we limit ourselves to a few remarks. In t h e simulation of fig. 2' a central electron temperature of 2.8 keV** w a s assumed to e t a better reproduction of t h e intensity ratios between t h e k a n d j dielectronic satellites a n % th e resonance line w.*** T h e concentrations of the'Li-, Be- a n d B-like ions needed to b e increased by factors of about 2 with respect to t h e values deduced by coronal equilibrium
*
148 lines a r e included in this simulation. The atomic d a t a describing t h e principal mechanisms of population of levels in the transitions considered have been expressely computed by F. Bely-Dubau, M. Cornille, J. Dubau, P. Faucher and A.H. Gabriel using the computer packages SUPERSTRUCTURE [ I l l of the University College of London and AUTOLSJ [I21 of the Meudon Observatory. These data will be published in the mentioned study.**
This value is consistent with the quoted electron temperature measurement within its estimated error.***
The alphabetic notation introduced by Gabriel [13] is used here and in fig. 2.JOURNAL DE PHYSIQUE
I I I I 1 I I I I I
45 46 47 48 49 50 51 52 53 54 55 Time (secs)
Fig. 4 Response of the central plasma rotation velocity v+(O) to co-injection of neutral beams of power P N ~ ~ . Positive values of v+ indicate toroidal rotation in the co- direction. The average electron density behaviour is also displayed.
calculatichs [14] in order to reproduce more accurately the spectral features depending upon the density of these ions.
Several possible reasons have been considered to explain the discrepancy between the coronal prediction and the experimentally observed concentrations of the lower ionization stages. They include possible errors in the temperature measurements, uncertainties in the ionization, recombination, excitation rates and radiative probabilities, effects due to possible hollow concentration profiles of the Ni impurity in the plasma column. For various reasons the mentioned effects are not believed to justif alone the observed discrepancies. These are also
h
ascribed to charge exchan e with neutral ydrogen and deuterium atoms present mainly in the cooler outer regions o f t h e plasma column, where the lower ionization stages a r e more abundant, and possibly to a transport of the impurities across the plasma stronger t h a n estimated from V U V spectroscopy results.'
The simulation of fig. 3 results from the integration along t h e instrument line of sight of t h e emissivities of 74 transitions of the one and two electron systems of Ni: in particular the magnetic dipole transition (2s 2SIl2 - 1s 2SIl2) i s blended with the Ly-a2 line contributing about ten percent to the intensity of the latter, whose value is calculated to be 0.5 times t h a t of the Ly-a1 in the actual plasma conditions [16-171.
*
Another study of the He-like Ni resonance spectrum, observed on the TFTR tokamak, including satellites from the three electron system has recently been published [15].4. VELOCITY MEASUREMENTS
Among the diagnostic applications of this instrument, besides the capital information about the ion temperature a t the plasma centre or that about electron temperature, impurity concentrations and ionization balances, i t i s important the information deduced from t h e measurement of the plasma toroidal angular velocity. During the purely ohmic heating phases of the discharges the plasma is seen to rotate in the direction opposite to t h a t of the toroidal plasma current ( t h e counter-direction) by a n amount (o+ = 4 to 8.103 s-1) just above o u r instrumental detectability. But when the neutral beams are fired they apply a net torque to the plasma so t h a t the latter i s observed to speed up on a time scale of about 1 s reaching a n g u l a r velocities up to oq, = 8.104 s-lin the co- or counter-direction according to the beams orientation relative to the plasma current (see for example fig. 4).
1 channel = 2.66 x 1
o4
rn S-'
Fig. 5 The frequency of M H D n = 1 oscillations vs the Doppler shift of the He-like Ni resonance line measured in channels at the detector. The line across the experimental data represents the equation 2 7 r f ~ ~ ~ = o+. (from ref. [191)
C1-290 JOURNAL DE PHYSIQUE
I t may be interesting to mention the strong correlation found a t J E T between the fluid velocity a s measured by our instrument a n d t h e n = 1 magnetohydrodynamic (MHD) oscillations. These phenomena are due to deformations of the axisymmetric structure of the plasma caused by MHD instabilities [18]. The n number is the integer ratio between the torus large circumference and the spatial period of these perturbations along t h a t circumference.
The frequency of the oscillations, a s measured by means of fixed probes, can be interpreted a s d u e to a rotation of the perturbation pattern. When such perturbations were detected with amplitude a t the plasma edge larger than 10-4 times the stationary field, the apparent rotation speed of the n = 1 modes both internally ( a s detected by the electron cyclotron emission measurements [9]) and externally located (measured with magnetic probes) was seen to be very close to the central fluid speed (see fig. 5). This coincidence was observed both in ohmic phases a t low rotation speed and a t large speed when the neutral beams were injected into the plasma [191.
Further evidence of the close coupling of the MHD modes with the fluid translation motion i s given a t t h e occurrence of the so-called mode lockings [201. U n d e r c e r t a i n circumstances, in fact, the mentioned modes are slowed down until t h e n = 1 magnetic perturbation locks to the external structure. These events also happen during the neutral beams injection. In these cases also the plasma toroidal velocity has been observed to slow down from values comparable to the fluid sound speed to zero on a time scale of 0.1 s , notwithstanding the persisting action of the beams [191.
CONCLUSIONS
High resolution X-ray spectroscopy has been applied to diagnose the plasmas produced by J E T tokamak. The spectroscopic equipment was routinely functioning during the fourth year of operation of t h a t device to monitor the central ion temperature and the plasma toroidal rotation. Its particular lay-out will allow to extend operation to the future D-T phase.
Emission spectra including lines from nickel ions H-like to B-like could be resolved for t h e f i r s t t i m e a n d compared with theoretical c o m p u t a t i o n s . A b s o l u t e w a v e l e n g t h measurements with relative uncertainties lower than 111500 have been performed.
Investigation of the relationship between data from this diagnostic methodology and other measurements allowed to describe remarkable phenomena a s well a s to judge the merits of the different operating scenarios.
ACKNOWLEDGEMENTS
The work of design, construction, exploitation and interpretation of data from the X-ray crystal spectrometer is d u e to the collaboration of t e a m s from t h e Frascati a n d J E T laboratories to whom all the author of this note i s indebted. In particular thanks a r e due to R. Bartiromo, K. Behringer, F. Bombarda, F. De Marco, W. Engelhardt, N. Foden, E. Kallne, G. Magyar, S. Mantovani, B. Oliver, L. Panaccione, G. Pizzicaroli, M. Shaw, J. Snipes, D. Stork, H. Summers, G.J. Tallents, G. Tonetti and B. Viaccoz.
REFERENCES
[ l ] M. Bitter e t al., Phys. Rev. Lett., 4 2 (1979) 304
[2] M. Bitter e t al., in Temperature, its Measurements a n d Control in Science a n d Industry, ed. by J.F. Schooley ( A I P , New York, 1982) 5,693
[3] R. Bartiromo e t al., Nucl. Instrum. and Methods, 225 (1984) 378
(41 P.H. Rebut and P.P. Lallia, Proc. 7th Int. Conf. on Plasma Phys. (Kiev, URSS, 6-10 April 1987)
[5] J. Jacquinot e t al., Proc. l l t h Int. Conf. on Plasma Phys. and Controlled Nucl. Fusion Research (Kyoto, J a p a n , 13-20 November 1986) MA-CN-47lF-1-1
(61 G. Duesing e t a]., Phil. Trans. R. Soc.London A322 (1987) 109
[7] J.P. Coad, K.H. Behringer and K.J. Dietz, J. Nucl. Mat. 145-147(1987) 747
[8] K. Behringer et al., Proc. l l t h Int. Conf. on Plasma Phys. and Controlled Nucl. Fusion Research (Kyoto, J a p a n , 13-20 November 1986) MA-CN-47/A-IV-1
191 D.V. Bartlett, E.A.M. Baker, D.J. Campbell, A.E. Costley, Rev. Sci. Instrum., 56 (1985) 940
[lo1 D. Veron, Int. School of Plasma Phys., Worksho on Dia nostics for Fusion Reactor Conditions (Varenna, Italy, 6-17 September 1982) l!ur 8 3 5 1 - % - ~ ~ p. 283
[ I l l W. Eissner, M. Jones, H. Nussbaumer, Comput. Phys. Commun. 8 (1974) 270 [I21 TFR Group, J. Dubau and M. Loulergue, J. Phys. B, 15 (1981) 1007
[I31 A.H. Gabriel, Mon. Not. R. Astr. Soc., 160 (1972) 99
[I41 IonizationRates from A. Burgess and M.C. Chidichimo, Mon. Not. R. Astr. Soc., 203 (1983) 1269
Radiative Recombination Rates from S. Von Goeler e t al., Nucl. Fusion, 15 (1975) 301 Dielearonic Recombination Rates from D.E. Post e t al., At. Data Nucl. Tables, 20 (1977) 397
Tabulated by A.E.P.M. Abel van Maanen, J E T Report JET-DN-T(85)29 [I51 H. Hsuan e t al., Phys. Rev. A, 35(1987)4280
1161 G.J. Tallents, J. Phys. B, 18 (1985) 3299
[I71 N.N. Ljepojevic, R.W.P. Mc Whirter, S. Volonte, J. Phys. B, 18 (1985) 3285 [I81 H.P. Furth, Phys. Fluids, 28 (1985) 1595
[I91 D. Stork e t al., 14th Europ. Conf. on Contr Fus. and Plasma Phys. (Madrid, Spain, 22-26 J u n e 1987) Vol. I l D , I, 306
(201 J.A. Snipes e t al., 14th Europ. Conf. on Contr Fus. and Plasma Phys. (Madrid, Spain, 22-26 J u n e 1987) Vol. 11D, I, 69
