R. Abgrall,1 M. H. Achard, J. Adam, G. Agarici, E. Agostini, M. Airaj, F. Albajar-Vinas, L. Allegretti, J. P. Allibert, J. C. Alliez, A. Allouche,2 J. Andreoletti, J. M. Ane, P. Angelino, T. Aniel, G. Antar, N. Arcis, A. Argouarch, C. Arnas,2 G. Arnoux, R. Arslanbekov, J. F. Artaud, E. Asp, S. Assas, G.Attuel, R.Aymar,A.Azeroual, S. Balme, O. Barana, B. Bareyt, V. Basiuk, M. Basko, P. Bayetti, L. Baylor,3 B. Beaumont, R. Becherer, A. Becoulet, M. Becoulet, L. Begrambekov,4 S. Benkadda,2 F. Benoit, V. Bergeaud, G. Berger-By, S. Berio, P. Bernascolle, N. Bernier, M. Berroukeche, B. Bertrand, D. Bessette, P. Beyer,2 P. Bibet, J. Bizzaro, P. Blanchard,5 J. Blum,6 S. Boddeker, D. Boilson,7 G. Bon Mardion, P. Bonnel, X. Bonnin, J. Boscary, G. Bosia, J. M. Bottereau, F. Bottiglioni, H. Bottollier-Curtet, C. Bouchand, G. Bouligand, F. Bouquey, C. Bourdelle, R. Bregeon, F. Bremond,8 S. Bremond, C. Breton, M. Breton, C. Brosset, R. Brugnetti, J. L. Bruneau, J. Bucalossi, R. V. Budny,9 Y. Buravand, C. Bush,3 M. N. Bussac,10 A. Cambe, H. Capes, J. J. Capitain, P. Cara, J. L. Carbonnier, S. Carpentier, J. Carrasco, A. Casati, O. Chaibi, C. Chamouard, M. Chantant, P. Chappuis, D. Chatain, E. Chatelier, M. Chatelier, J. H. Chatenet,10 X. P. Chen, L. Cherigier, G. Chevet, L. Chiarazzo, D. Ciazynski, G. Ciraolo, F. Cismondi, F. Clairet, J. Clary, C. Clement, L. Colas, N. Commaux, E. Corbel, J. J. Cordier, Y. Corre, L. Costanzo, A. Cote, J. P. Coulon, L. Courtois, X. Courtois, B. Couturier, J. P. Crenn, P. Cristofani, N. Crouseilles,11 O. Czarny, P. Da Silva Rosa, C. Darbos, G. Darmet, M. Davi, R. Daviot, H. De Esch, B. De Gentile, J. C. De Haas, E. De La Cal, C. De Michelis, C. Deck, J. Decker, P. Decool, P. Degond, R. Dejarnac, E. Delchambre, E. Delmas, L. Delpech, H. Demarthe, M. Dentan, G. Depret, P. Deschamps, C. Desgranges, P. Devynck, L. Doceul, N. Dolgetta, C. Doloc, Y. Dong,12 P. Dore, D. Douai, H. Dougnac, H. W. Drawin, J. Druaux, M. Druetta,13 F. Dubois, M. Dubois, N. Dubuit, J. L. Duchateau, T. Dudok de Wit, E. Dufour, R. Dumont, G. Dunand, L. Dupas, Y. Duran,14 A. Durocher, D. Edery, A. Ekedahl, D. Elbeze, L. G. Eriksson, D. Escande,2 A. Escarguel, F. Escourbiac, T. Evans,15, F. Faisse, G. Falchetto, T. Fall, M. Farge,16 J. L. Farjon, E. Faudot,17 P. Fazilleau, N. Fedorczak, C. Fenzi-Bonizec, J. R. Ferron,15 I. Fidone, C. Figarella, E. Fleurence, I. Fleury, M. Fois, C. Forrest,15 C. A. Foster,3 S. Fouquet, C. Fourment, D. Fraboulet, P. Francois, B. Franel, D. Frigione,18 P. Froissard, G. Fubiani, V. Fuchs,14 M. Fumelli, B. Gagey, V. Galindo, D. Gambier, L. Garampon, X. Garbet, R. Garbil, J. Garcia, J. L. Gardarein, L. Gargiulo, P. Garibaldi, P. Garin, E. Gauthier, A. Geraud, T. Gerbaud, F. Gervais,19 M. Geynet, P. Ghendrih, T. Gianakon, R. Giannella, C. Gil, J. P. Girard, G. Giruzzi, L. Godbert-Mouret,2 P. Gomez, M. Goniche, A. Gordeev,4 G. Granata, V. Grandgirard, R. Gravier, B. Gravil, M. Gregoire, S. Gregoire, P. Grelot, D. Gresillon,19 C. Grisolia, G. Gros, *Other affiliations are those at the time the collaborations began. 1INRIA-CNRS, Université Sciences et Technologies, Bordeaux, France 2Physique des Interactions Ioniques et Moléculaires ~PIIM !, Université de Provence, Centre Universitaire St Jérôme, 13397 Marseille Cedex 20, France 3Oak Ridge National Laboratory, Fusion Energy Division, P.O. Box 2009, Oak Ridge, Tennessee 37831-8070, USA 4Moscow Physics and Engineering Institute ~MEPhI!, 31 Karhirskoe Sh, 115409 Moscow, Russian Federation 5Centre de Recherche en Physique des Plasmas, Association EURATOM-Confédération Suisse, Ecole Polytechnique Fédérale, PPB-Ecublens, 1015 Lausanne, Suisse 6Université Joseph Fourier, Grenoble I, B.P. 53, 38041 Grenoble Cedex 9, France 7School of Physical Sciences, Dublin City University, Glasnevin, EI-Dublin 9, Ireland 8INRIA Sophia-Antipolis, 2004 Route des Lucioles, B.P. 93, 06902 Nice-Sophia-Antipolis, France 9Princeton Plasma Physics Laboratory, James Forrestal Campus, Princeton, New Jersey 08543, USA 10Centre de Physique Théorique, Ecole Polytechnique, 91128 Palaiseau, France 11IRMA, Université Louis Pasteur, Strasbourg, France 12Southwestern Institute of Physics, Chengdu 610041, China 13Laboratoire TSI, Université Jean Monnet, 42023 St-Etienne, France 14Association EURATOM-IPP.CR, Institute of Plasma Physics AS CR, Za Slovankou 3, 182 21 Praha 8, Czech Republic 15General Atomics, P.O. Box 85608, San Diego, California 921865608, USA 16LMD, Ecole Normale Supérieure, 75 Paris, France 17LPMIA, Université Henri Poincaré, Nancy 1, B.P. 239, 54506 Vandœuvre Cedex, France 18Associazione EURATOM-ENEAsulla Fusione, C.R. Frascati, Roma, Italy 19Laboratoire de Physique et Technologie des Plasmas ~LPTP!, Ecole Polytechnique, 91128 Palaiseau, France

The last experimental performances achieved on Tore Supra using the radio frequency heating and current drive systems are presented. In view to increase the injected power and the pulse length new developments are undertaken. Results concerning the 118 GHz ECRH system, a new guard limiter and an advanced LHCD antenna are reported.

Significant results on steady state control of the plasma parameters have been obtained with RF systems on Tore Supra. A sustained effort is under way to improve power transmission reliability. Developments of sources and antennas are in progress for future operation on Tore Supra with purse length in tile range of several hundreds of seconds.

Long pulse RF experiments at high power level have been carried out on Tore Supra by coupling lower hybrid (LH) waves to the plasma. Plasma pulses lasting two minutes have been obtained sucessfully, leading to a new injected energy record of 280 MJ. Full current drive discharges have been also sustained beyond 60 s using active feedback control of the LH power on the plasma current; power densities of up to 24 MW/m(2) have been achieved. Experiments on power coupling at distances of up to 15 cm from the plasma have provided new potentialities concerning far distance coupling in a reactor environment. Steady state conditions were also achieved in combined scenarios with 2 MW of LH power and 2 MW launched by the ion cyclotron resonance heating (ICRH) system, corresponding to a power density of 8 MW/m(2). With a view to improving long pulse performance, several ongoing projects relevant to ITER are being carried out for the LH system: mode converter, passive-active multijunction and new lateral protections. Progress in the understanding and modelling of LH power deposition profiles is reported, taking into account the toroidal magnetic field ripple in the wave dynamics, which reaches 7% a the Tore Supra plasma edge. Despite the rather modest amplitude of the focal ray perturbation, its global effect may be strong, as a consequence of the combined effects of toroidal and poloidal inhomogeneities. Simulations of LH experiments have been performed by coupling ray tracing calculations with a one dimensional relativistic Fokker-Planck code. In the 'few passes' regimes where the wave makes some, but not many, passes inside the plasma before it is absorbed, the effect of magnetic ripple usually leads to a broadening of the power deposition profile and a shift towards the plasma centre, a behaviour which may be explained by an increase of the overall ray stochasticity. A comparison between simulations and experimental observations is reported.

Recent Tore Supra results supporting the feasibility of high-confinement. long duration discharges are reported. Two-minute discharges with improved confinement have been obtained. This progress is largely due to improvements in the operating control system of Tore Supra. which now allows real-rime feedback control of global plasma parameters. A clear correlation between improved confinement and current profile shape has been established. Transport barriers for electron heat diffusion are observed in experiments where the magnetic shear is weak or negative in the central part of the plasma. These observations strongly support development of new current drive schemes for current profile control. Considerable progress concerning particle and heat exhaust has also been achieved. A new technique allowing conditioning in the presence of a toroidal magnetic field has been implemented. A vented limiter has been tested as a means of particle exhaust through collection of neutrals, and its performance is compared with that of a "classical", ion collecting throat limiter. The knowledge gained from extensive long pulse experimentation has been used to develop a new generation of plasma facing components. which will permit further development of the long discharge capability of Tore Supra.

Recent stationary improved confinement experiments with current density profile modifications in non-inductive Tore Supra operation are reported. Significant progress has been obtained by achieving long duration discharges using lower hybrid (LH) waves: (i) 2 min in the 4 T improved confinement LHEP (LH enhanced performance) regime, H-RLW = W-e/W-e-RLW = 1.6 at a current I-p = 0.8 MA, loop voltage V-1 approximate to 0.1 V, LH power P-lh = 2.5 MW (W-e, W-e-RLW are the electron energy content and Rebut-Lallia-Watkins L-mode prediction), (ii) 75 s long in a fully non-inductive LHEP regime (H-RLW = 1.4, V-1 = 0 V, I-p approximate to 0.62 MA, P-lh approximate to 3.0 MW) using a new plasma control scheme, (iii) 30 s at 1.7 MA (V-1 approximate to 0.25 V, P-lh = 2.5 MW) in the L-mode regime. MHD stability in full current drive operation and the role of weak-or reversed-central magnetic shear in the central LHEP electron temperature transition (reproduced in full current drive in a systematic manner and measured by independent electron temperature diagnostics) are discussed.Stationary high-beta(p) experiments were performed with fast-wave direct electron heating (FWEH) in a large range of operation: 42-76 MHz and B-t = 2-3.9 T. Maximum coupled power of 9.5 MW is obtained in the 48 MHz/2T configuration with good electron heating performance and improved confinement at high density (central density of approximate to 6 x 10(19) m(-3)). By increasing the magnetic shear in the gradient region, stationary improved confinement states (6 MW during 5 s, H-RLW approximate to 2.2) are reached in a reproducible manner with poloidal beta, beta(p), approaching 1 and 40% of bootstrap current.The high-bootstrap FWEH experiments have been recently combined with LHCD at reduced loop-voltage (less than or equal to 0.1 V and up to 70% of non-inductive current). A total injected power up to 7.2 MW (LH: 3.6 MW, FW: 3.6 MW) has produced stationary improved confinement with a peaked electron temperature profile (central value up to 8.0 keV) at a central density of 3.9 x 10(19) m(-3).

In view of high-power, long-pulse steady-state operation, Tore Supra has incorporated in its design the active control of heat and particles in a realistic environment. In the early experimental phase of Tore Supra, the first generation of plasma-facing components was tested, and these tests provided much physics and technological information and illuminated various operational difficulties. In particular, these experiments revealed the weakness of the graphite-to-metal brazing process originally adopted for actively cooled high-heat-flux components. Consequently, a new inner-wall technology was developed in 1994 and is to be tested in 1995-1996 with a totally rebuilt 40-deg toroidal sector. A carbon-fiber-reinforced carbon-metal compound is based on the newest brazing technology and rigorous quality control. Components such as the toroidal pump limiter and the guard limiters of plasma-heating antennas are being developed in the same way. For structures where brazing is difficult, boron carbide-coated components have been developed and installed in Tore Supra. For lower heat fluxes, a bolted concept has been designed and tested. The influence of inner-first-wall misalignment in Tore Supra on the power exhaust limitation of brazed components has been studied.Results from the technological development for the different power exhaust systems and the associated experimental knowledge obtained during plasma operation in Tore Supra are presented.

The combination of r.f. waves in the lower hybrid (LH) and ion cyclotron frequency ranges offers a versatile and efficient way of heating tokamak plasmas while controlling their transport properties and magnetohydrodynamic stability through the control of the current density profile. Experimental and theoretical studies on the applications of such plasma waves have been carried out on Tore Supra during recent years and are reported here.The LH system coupled up to 6.5 MW during 2 s at 3.7 GHz through two multijunction launchers. In the longest plasma shot, the total injected LH energy reaches a record value of 170 MJ during a 62 s LH pulse, at a power level of 2.8 MW, corresponding to an average power density of 17 MW m(-2). The ion cyclotron resonant frequency (ICRF) system (35-80 MHz) is composed of three resonant double-loop antennae. Up to 4 MW have been coupled with a single antenna, allowing a record power density through the Faraday screen of 16 MW m(-2) to be reached. 30 s steady-state r.f. pulses have been obtained with up to 54 MJ delivered to the plasma. One of the major observations has been the transition to the so-called ''stationary lower hybrid enhanced performance (LHEP) regime'' (I-p = 0.8 MA; n(e0) = 2.8 x 10(19) m(-3); P-LH = 3.2 MW) in which the (flat) central current density (q(0), approximate to 2) and (peaked) electron temperature profiles (T-e0 approximate to 6-8 keV) are fully decoupled. This regime exhibits a significant improvement of the global confinement (40%) owing to the increase in l(i), i.e. in the magnetic shear in the outer half of the discharge, supplemented by a large reduction in the electron thermal diffusivity in the central zone where the magnetic shear vanishes because of the slight off-axis character of the LH power and current deposition. TRANSP analyses show that LHEP plasmas provide access to the second ballooning stability regime. At higher current and density (I-p = 1.5 MA; n(e0) = 6 x 10(19) m(-3)), ICRH stabilization of sawteeth (4 MW) combined with lower hybrid current drive (LHCD) current profile modifications has allowed to extend the stabilized phase for up to 1 a with 3.4 MW of LH power, the duration of sawtooth-free periods increasing with increasing LH power. The dynamical properties of fast electrons during LHCD have been investigated recently on Tore Supra through power modulation experiments. It is shown that slowing down always predominates and, from the long-time evolution of the hard X-ray emission, the radial diffusion rate of the fast electrons is estimated to be 0.1-0.3 m(2) s(-1).Theoretical developments have focused on the modelling of LHCD and also on fast wave current drive (FWCD) and fast wave heating. The effect of intrinsic stochasticity on the propagation of LH waves is discussed and a fully developed statistical theory of stochastic wave diffusion and multipass absorption, with applications to Tore Supra through a wave diffusion-Fokker-Planck (WDFP) numerical code, is briefly presented. This model provides a simple explanation for the temperature dependence of the LHCD efficiency in small tokamaks. The ion cyclotron resonant heating (ICRH) full-wave code ALCYON has been upgraded to compute the power and current deposition profiles from direct electron absorption of the fast wave (electron Landau damping-transit time magnetic pumping). The code has been used to study FWCD in Tore Supra, the Joint European Torus and the International Thermonuclear Experimental Reactor.Finally a new concept of an efficiently cooled reflector antenna for LHCD applications in a steady-state reactor is briefly described.

The link between the current profile and the confinement is studied, involving various regimes: high power minority ion cyclotron resonant heating, high power lower hybrid current drive, fast wave direct electron heating and current drive and pellet enhanced performance. It is shown that the electron heat diffusivity decreases when the magnetic shear increases in the confinement zone and/or when it decreases in the plasma center.

Seven superconducting focusing magnets have been constructed for vertical gyrotrons devoted to the plasma heating of the tokomak Tore Supra. The performances in magnetic field strength, profile and homogeneity are spread over a large range so as to suit gyrotrons of microwave frequencies extending from 110 GHz to 150 GHz. The cryostats have a low consumption in cryogenic fluids which insure a one week autonomy

During the 92 TORE SUPRA experimental campaign, 48% of the plasma shots has been devoted to lower hybrid experiments. 75 shots with a mean pulse duration of 27 s (10 to 60 s1) have been performed using the LH system with a mean RF power of 2.8 MW corresponding to an incident power density at the launcher mouth of 17 MW/m2. More than 40 RF shots—1 to 5 s long—have been also achieved with injected power between 0.6 to 0.94 of the maximum available power (6.8 MW). With a typical launcher‐limiter distance of 3 cm, the thermal load on the carbon guard of the launcher is lowered to less than 1% of the total injected power (3 MW).On test bed facility, modules of antenna made of Dispersion Strengthened Copper (D.S.C.) have been successfully tested in collaboration with the CCFM (Canada) and JAERI (Japan). Super long shots up to 100 mn have been achieved with transmitted power density of 50 MW/m2.For next step device, as TORE SUPRA CONTINU, D.S.C. material, and RF power density of 25 MW/m2 are relevant options for high temperature quasi continuous operating conditions.

Technical aspects of the installation of the ECRH transmission lines and antennas on Tore Supra areoutlined. Six corrugated circular transmission lines will transport each 500 kW in HE11 mode at 110 or118 GHz. Special attention is given to vacuum operation, CW operation, cooling, alignment and supportstructures, and mode purity measurements.

The lower hybrid system of Tore Supra (LHCD) is mainly devoted to current drive experiments allowing long pulse (> 30 s) discharges in the range of 1-2 MA and for line-averaged density around 3.1019m−3(ref. 1). Up to now about 30 plasma shots have been performed with pulse duration larger than 30 s. Record shot of 60 s plateau 1 MA 2.10−19m−3averaged density was successfully achieved thanks to a 62 s 2.8 MW LH power injection corresponding to an energy of 170 MJ. High power LH injection exceeding 6 MW and corresponding to more than 90 % of the maximum available power and to power density of 40 MW/m2was also performed for pulse duration of 1 to 3 s. The conditioning is mainly based on baking and repetitive RF 10 ms long pulses in vacuo between plasma shots. After long opening time the mid maximum power is quickly obtained with some tens plasma shots, while more than 100 plasma shots are needed to reach power exceeding 5 MW. For ITER like device (i.e. quasi continuous and reliable operation) at frequency close to 5 GHz, RF working power density of 20-30 MW/m2seems to be a relevant value for the design of the antennae.

In large tokamak such as JET, JT60, Tore Supra, the lower hybrid frequency antennae are composed of hundreds of waveguides and are fed by tens of klystrons. The N// spectrum which is then radiated depends on the geometry and on the wave amplitude and phase. It has to be checked when the total system is linked to the machine. Here four methods using the RF system are described which could be relevant for ITER: The first method is the standard one. It uses measurements in laboratory to have the antennae electric lengths and measurement on the real system after its calibration. The incident wave is driven by the RF system for each klystron and the incident phase is then measured at the input of the antenna, computed at the antenna mouth and compared to the ones of neighbouring klystron. The second one uses only the real system. The incident phase in the waveguide i fed by the klystron i is measured at the antenna mouth with the use of a magnetic loop. The other waveguides fed by the same klystron are matched with good loads. The calibration of the RF measurement system is not necessary here. The third method uses the observation of the amplitude of the reflected signal of the klystron i and i+1 when they are fed together and the incident phase of i is constant whereas the one of i+1 is pulsed. The experiment has to be done on plasma. In this method only the measurement of amplitude is necessary, the components of the antenna must be identical from the reflection point of view. The fourth one consist in the measurement of the scattering matrix at the input of the antenna on plasma. For N klystrons, N cases of feeding linearly independent must be done in such a way to have a linear system of N × N equations obtained from the measurement of the electric field at the input of the antenna (192 RF signals on Tore Supra). Then the S parameters can be obtained and by using the property of symmetry of the matrix, the measurement system calibration can be checked.

With Lower Hybrid Current Drive (LHCD), very long pulses can be obtained in TORE SUPRA. Volt-second saving studies by LHW assisted current ramp-up, current drive and electron heating are first discussed. A 12 seconds stationary discharge with improved energy confinement time (40 %) by peaking the current profile is presented with a central electron temperature up to Te(o) approximately 10 keV . A 1 minute flat top time duration discharge has been achieved with a plasma current I(p) = 1 MA : 80 % of this current is driven by LHW. During this discharge the core density is fed continuously by deuterium gas puff indicating that the inner first wall is still pumping all along the shot. Finally results concerning the interaction of LHW with high density plasmas (peak density approximately 10(20) m-3) and 3 MeV fusion protons are given.

The main characteristics of the ECRH current drive system to be installed on Tore Supra are developed. The system is built around specially developed gyrotrons expected to deliver 500 kW CW at 110 GHz. New features such as the use of cryogenic windows are included in order to achieve the very long pulses (up to 210 s) required by the Tore Supra program.

The TORE SUPRA lower hybrid current drive experiments (8 MW/3.7 GHz) use large phased waveguide arrays, four rows of 32 active waveguides and two passive waveguides for each of the two grills, to couple the waves to the plasma. These launchers are based on the 'multijunction' principle which allows them to be quite compact and is therefore attractive for the design of efficient multi-megawatt antennas in NET/ITER. Extensive coupling measurements have been performed in order to study the radiofrequency (RF) characteristics of the plasma loaded antennas. Measurements of the plasma scattering coefficients of the antennas show good agreement with those obtained from the linear coupling theory (SWAN code). Global reflection coefficients of a few per cent have been measured in a large range of edge plasma densities (0.3 × 1018 m-3 ⩽ neg ⩽ 1.4 × 1018 m-3) or antenna positions (0.02-0.05 m from the plasma edge) and up to a maximum injected RF power density of 45 MW/m2. When the plasma is pushed against the inner wall of the chamber, the reflection coefficient is found to remain low up to distances of the order of 0.10 m