München, Techn. University, Diss., 2003.

Düsseldorf, Universiẗat, Diss., 2003.

Greifswald, University, Diss., 2002.

Bochum, Universiẗat, Diss., 2001.

No description available.

Dans un tokamak, les interactions entre le plasma et la paroi conduisent à l'érosion des composants face au plasma. Celle-ci peut nuire au fonctionnement et à la sûreté du tokamak. Afin de respecter les limites imposées pour la sûreté du projet ITER, il est donc nécessaire de contrôler la quantité de matière érodée. La spectroscopie optique d’émission dans le domaine visible est traditionnellement utilisée pour mesurer des flux de particules de la paroi vers le plasma. Ces mesures sont permises grâce à un modèle collisionnel-radiatif s’appuyant sur des données de physique atomique. Cependant, ces données ne prennent pas en compte la géométrie d’observation du diagnostic de spectroscopie, et présentent des incertitudes relativement importantes. D'autre part, les phénomènes de transport, de déposition et de ré-érosion, ainsi que la modification de la transmission ou de la réflexion des optiques peuvent conduire à une mesure erronée de la quantité de matière effectivement érodée. Une méthode de calibration in situ, qui consiste à injecter par laser une source connue de particules de carbone dans la ligne de visée du diagnostic de spectroscopie pendant les décharges plasma, est proposée. L’étude expérimentale de l’ablation laser du carbone a permis de déterminer les conditions optimales à la constitution de cette source, et de caractériser les espèces ablatées. Ces expériences sont complétées par une modélisation du spectre d’émission du plasma induit par laser, afin d’obtenir des informations sur son degré d’ionisation. Enfin, les résultats des premières expériences de validation réalisées sur le tokamak allemand TEXTOR sont présentés et discutés. / In a tokamak, plasma-wall interactions lead to the erosion of plasma facing components, which can be detrimental to plasma operation and to the safety of the tokamak. In order to fulfil the safety requirements imposed to the ITER project, it is necessary to monitor the amount of eroded material. Optical emission spectroscopy in the visible range is traditionally used to measure particle fluxes from the wall to the plasma. These measurements are done thanks to a collisionnal-radiative model based on atomic physics data. However, these data don’t take into account the observation geometry of the spectroscopic diagnostic, and suffer from relatively large uncertainties. Furthermore, transport, deposition and re-erosion phenomena, as well as the evolution of the transmission or the reflection of optical components can lead to an incorrect estimation of the amount of effectively eroded material. An in situ calibration technique, which consists in injecting by laser a known carbon particle source in the line of sight of the spectroscopic diagnostic during plasma operation, is proposed. The experimental study of laser ablation of carbon allowed to determine the optimal conditions for the constitution of this source, and to characterise the ablated species. These experiments are completed by a modelling of the emission spectrum of the laser induced plasma, in order to obtain information on its ionisation degree. Finally, results of the first validation experiments realised in the German TEXTOR tokamak are presented and discussed.

Les nombreux avantages que présenteraient la fusion thermonucléaire, en particulier la configuration tokamak, en font un candidat idéal en vue de la transition énergétique. Cependant, un certain nombre de difficultés technologiques et physiques restent à résoudre avant que l'étape d'une centrale électrique à fusion puisse voir le jour. La production de poussières est l'une des principales difficultés rencontrées dans les tokamaks. Ces petites particules composées de matériaux présents dans les parois de la machine sont créées par l'érosion de ces parois par le plasma dans lequel les réactions de fusion doivent avoir lieu. Les poussières peuvent être transportées dans le plasma et y libérer de grandes quantités d'impuretés, ce qui a pour conséquence de baisser les performances de la machine (en augmentant les pertes radiatives et en créant des instabilités), et qui peut mettre en danger les composants face au plasma. Dans le but de comprendre le transport de ces poussières, des expériences d'injection sont réalisées sur le tokamak coréen \KSTAR. Les trajectoires des poussières dans le plasma sont observées par des caméras rapides et sont extraites des films à l'aide de routines de traitement d'images. Un code numérique implémentant les derniers modèles d'interactions plasma-poussières est développé, et des comparaisons avec les données expérimentales sont faites, confirmant la tendance générale de ces modèles à la sous-estimation de la longueur des trajectoires des poussières. Des pistes d'amélioration sont présentées. Concernant les sources et puits de poussières, l'accent est porté sur l'adhésion et remise en suspension de particules sur les parois de la machine. / Thermonuclear fusion could play an important role amongst the numerous alternative energy sources, especially though the tokamak configuration. It could be a prime candidate for the energy transition, owing to its significant advantages (fuel abundance, low amount of wastes generated, low risks of accidents). However, a certain amount of technological and physical challenges require solving before any fusion power plant can be built. Dust production is one of the major difficulties encountered in tokamaks. These small particles, made out of wall material, are created by erosion of the plasma-facing components by the plasma, where the fusion reactions occur. Dust particles can be transported in the plasma, thereby unleashing large amounts of impurities, which in turn reduces the plasma performances (by raising radiative losses and generating instabilities) and can even jeopardize plasma-facing components. Aiming to understand dust transport, injection experiments are performed on the Korean tokamak \KSTAR. Trajectories are recorded on film via fast cameras and are extracted by image processing routines. A numerical tool implementing the latest models for dust-plasma interactions is developed, and comparisons with experimental data is made, confirming the overall tendency of these models to underestimate the trajectory lengths. Leads of improvements are presented. Concerning dust sources and sinks, the focus is made on dust adhesion and resuspension of dust on the machine walls.

Diamagnetic measurements of poloidal beta have been performed in the STOR-M tokamak by a flux loop placed exterior to the vacuum chamber. Poloidal beta is defined as the ratio of plasma kinetic pressure to poloidal magentic field pressure. Compensation for the vacuum toroidal field has been performed using a non-enclosing co-planar coil, and vibrational compensation from auxiliary coils. It was found that in STOR-M conditions (20\% toroidal magnetic field decay over discharge) there is significant influence on the diamagnetic flux measurements from strong residual signals, presumably from image currents being induced by the toroidal field coils, requiring further compensation. A blank (non-plasma) shot is used specifically to eliminate the residual component which is not proportional to the toroidal magnetic field. Data from normal ohmic discharge operation is presented and calculations of poloidal beta from coil data ($eta_ heta simeq 0.5$) is found to be in reasonable agreement with the values of poloidal beta obtained from measurements of electron density and Spitzer temperature with neoclassical corrections for trapped electrons. Contributions present in the blank shot (residual) signal and the limitations of this method are discussed. A pulse with Compact Toroid Injection was examined and compared to a normal ohmic discharge, and one where the Compact Toroid Injector was used to supply the tokamak with neutral gas. Soft X-Ray (SXR) measurements were taken and compared. There is a strong agreement between the profiles of the poloidal beta and the SXR measurements. The bulk plasma thermal energy was measured and found to increase by 5.6 J following the injection of a CT. The diamagnetic measurements appear to be affected by image currents induced in the chamber walls by the plasma current, and also by plasma position fluctuations. Future work outlining the possibilty of compensating these currents and improving the measurements is presented.

Central fuelling is a fundamental issue in the neat generation tokamak ITER (International Thermonuclear Experimental Reactor). It is essential for optimization of the bootstrap current which is proportional to the pressure gradient of trapped particles. The conventional tokamak fuelling techniques, such as gas puffing and cryogenic pellet injection, are considered to be inadequate to fulfill this goal due to premature ionization caused by high plasma temperature and density. Fuelling by injecting a compact torus (CT) may be the only viable method suitable for a reactor-grade tokamak. CTs can be injected at different angles with respect to the tokamak toroidal magnetic field, either horizontally or vertically. In vertical injection, deeper CT penetration is expected due to the absence of the gradient of tokamak toroidal magnetic field in that direction. This thesis contributes to experimental investigation of vertical compact torus injection into the STOR-M tokamak. <p>To perform vertical injection, the original injector- USCTI (University of Saskatchewan Compact Torus Injector) was modified by attaching a segment of 90˚ curved drift tube to bend the CT trajectory from horizontal to vertical. Bench tests have shown that a CT injected horizontally can be deflected effectively to the vertical direction. The velocity of 130 kms^{-1}has been achieved while the CT passes through the 90˚ curved drift tube. It was found that the CT magnetic field structure kept intact as a typical structure of compact torus plasma. By further optimization of the USCTI configuration, the velocity has been increased to 270 kms^{-1}. Based on the encouraging bench test results, actual vertical CT injection experiments have been performed in the STOR-M tokamak. Experimental results demonstrated, for the first time, vertical CT injection into a tokamak. Prompt increases both in line averaged electron density and in soft X-ray emission (central cord) are observed following vertical injection. Some H-mode phenomena, characterized by suppression of the m =2 Mirnov oscillation level and drop in Hα radiation level, have also been observed following the vertical injection. Fuelling effects caused by vertical injection and by tangential injection are discussed. The experimental results suggest that vertical CT injection is a feasible tokamak fuelling technique.

Central fuelling is a fundamental issue in the neat generation tokamak ITER (International Thermonuclear Experimental Reactor). It is essential for optimization of the bootstrap current which is proportional to the pressure gradient of trapped particles. The conventional tokamak fuelling techniques, such as gas puffing and cryogenic pellet injection, are considered to be inadequate to fulfill this goal due to premature ionization caused by high plasma temperature and density. Fuelling by injecting a compact torus (CT) may be the only viable method suitable for a reactor-grade tokamak. CTs can be injected at different angles with respect to the tokamak toroidal magnetic field, either horizontally or vertically. In vertical injection, deeper CT penetration is expected due to the absence of the gradient of tokamak toroidal magnetic field in that direction. This thesis contributes to experimental investigation of vertical compact torus injection into the STOR-M tokamak. <p>To perform vertical injection, the original injector- USCTI (University of Saskatchewan Compact Torus Injector) was modified by attaching a segment of 90˚ curved drift tube to bend the CT trajectory from horizontal to vertical. Bench tests have shown that a CT injected horizontally can be deflected effectively to the vertical direction. The velocity of 130 kms^{-1}has been achieved while the CT passes through the 90˚ curved drift tube. It was found that the CT magnetic field structure kept intact as a typical structure of compact torus plasma. By further optimization of the USCTI configuration, the velocity has been increased to 270 kms^{-1}. Based on the encouraging bench test results, actual vertical CT injection experiments have been performed in the STOR-M tokamak. Experimental results demonstrated, for the first time, vertical CT injection into a tokamak. Prompt increases both in line averaged electron density and in soft X-ray emission (central cord) are observed following vertical injection. Some H-mode phenomena, characterized by suppression of the m =2 Mirnov oscillation level and drop in Hα radiation level, have also been observed following the vertical injection. Fuelling effects caused by vertical injection and by tangential injection are discussed. The experimental results suggest that vertical CT injection is a feasible tokamak fuelling technique.