Search

The ITER tokamak requires two heating Neutral Beam Injectors (NBIs), based on the neutralisation of 40A of negative hydrogen/deuterium ions accelerated to 1MeV with a pulse length of one hour. As these simultaneous parameters were never experimentally achieved, in order to optimise source operation and to assess beam properties, a specific test facility ("PRIMA") was established in Padova, comprising two experiments: SPIDER (full-size negative ion source with 40A beam, 100keV particle energy) and MITICA, the prototype of ITER NBIs. A key NBI component is the particle accelerator, where electrostatic and magnetic fields produce the beam optics. The design of SPIDER and MITICA accelerators was based on the most advanced numerical codes available for the investigation of the expected beam properties and on the most up-to-date experience of various research groups. Several diagnostic systems will characterise source and beam during the experiments. In view of the operation, effort is presently devoted (within bilateral collaboration frameworks between Italian and Japanese agencies) to the preparation for the synergistic employment of simulations and experiments. Specifically, a characterisation of the beam properties is carried out in the NIFS-R&D multi-cusp negative ion source (NIFS-RNIS) by a sophisticated set of source and beam diagnostic systems. Simulations are mainly performed at Consorzio RFX, by a suite of numerical tools to compute or investigate various aspects of beam physics. Preliminary comparisons showed very good agreement between simulations and measured data regarding the basic features of the accelerator. The present contribution describes the synergy between numerical codes and diagnostic techniques. Both are applied to real-time characterisation of the beam, with the main scopes of providing interpretation of the measured results and of supplying data for successive choices during the operation. These activities allow also extensive validation of numerical models and the choice of the most useful set of codes to be used in future experiments like DEMO.

Development of the high performance negative hydrogen ion source is a fundamental demand in realizing fusion reactor. In order to clarify the extraction mechanism of H´, temporal and spatial variations of the negative ions and electrons in the extraction region are intensively surveyed at NIFS. In addition, the beam acceleration experiments have been performed by changing the accelerator configuration in order to improve the voltage holding capability and to study the negative ion beam optics. In a caesiated hydrogen plasma, it was observed that the negative hydrogen ion density (nH´) becomes one order higher in magnitude than the electron density (n e) in the vicinity of the plasma grid (PG). The response of negative-ion rich plasmas to the extraction field was investigated by measuring the plasma potential (V p) profiles in the axis perpendicular to the PG before and during beam extraction. The V p increases with applying the extraction field, and the influence of the extraction field on the V p was observed at 30 mm from the PG. As for the nH´, it was also observed that the extraction field affects on the nH´ at 30 mm from the PG, where the nH´ decreases simultaneously with the beam extraction. These observations indicate that the extraction field affects the particle dynamics in the wide region extending over 30 mm from the PG. This feature is completely different from that of electron-ion plasmas. We also found that the negative ion production efficiency becomes twice higher by changing the shape of grounded grid holes without any modification on the plasma chamber. We assumed that the behaviour of the back-streaming ion was affected by changing the GG. The back-streaming ion trajectory was analyzed with the beam trajectory simulation, and it was found that the back-streaming ion distributes in the larger area on the back plate with the slot GG. This implies that some part of condensed Cs on the back plate was more effectively evaporated by the back-streaming ion with the slot GG and that Cs flux onto the PG surface increased, which resulted in the enhancement of the negative ion production. This result suggests that the accelerator configuration is one of the key factors to determine the negative ion production efficiency and that the Cs consumption can be reduced by the Cs recycling from the wall of the ion source chamber.

Negative ions are fundamental to produce high energy neutral beams used to he at the plasma in fusion experiments . The negative ions so urces for reactor class fusion devices represent a huge extrapolation from existing device, hence the use of modelling tools i s mandatory. In particular, the design of the neutral beam injector of the ITER experiment was strongly based on the results of n umerical codes. In this report we c ompare the models with the results of real experiments, carried out during dedicated campaigns at the R&D negative ion source (RNIS) hosted at the national institute for fusion science (NIFS), in Japan.

Neutral beam injector (NBI) is an indispensable device to produce high-density and hot plasmas in fusion development machine. We have developed negative-ion-based NBI (N- NBI) in LHD, that have been operated continuously since 1998. Three N-NBIs are arranged tangentially, and was equipped with a negative hydrogen ion (H-) source which enabled high energy and high power beam injection. The nominal specication of beam energy and beam power for each N-NBI was 180 keV and 5 MW, respectively. We had achieved 16 MW beam injection by three N-NBIs in LHD experiment. Moreover, we installed two positive-ion-based neutral beam injector (P-NBI) arranged perpendicularly to the LHD. High-power (12 MW) with lower energy (40~60 keV) beam injection allowed effective ion heating and profile measurement of ion temperature by charge exchange recombination spectroscopy. A hydrogen negative ion source is a key component in N-NBI system, that will also use future NBI in ITER. Understanding of H- behavior from cesiated metal production surface to extraction aperture is an important issue to maintain high power negative ion beam. We have researched distribution of H- ions and its extraction behavior by cavity ring-down spectroscopy and a spectrally-selective imaging system, respectively, in the 1/3 scaled ion source (NIFS-RNIS). We found that H- ions widely distributed in the extraction region, and these ions constituted H- beam. Recently, we have a lot of domestic and international collaboration program for H- source. We applied a spectrally-selective imaging system in RF hydrogen negative on source in ELISE test facility in Max-Planck-Institut fur Plasmaphysik (IPP) Garching (Germany). We found clear signal response on H? emission during beam extraction. This observation will contribute to obtain behavior of H- ions in RF source which is variable output for production high power beam and optimization of beam shape in ITER source. We also observed an extracted H- beam shape in NIFS-RNIS used the mini-STRIKE system which consists of a carbon- fibre-carbon-composite (CFC) tiles and an infrared camera collaborated with Consorzio RFX (Italy). We found the clear dependence between the beam shape and operating parameters. These findings have contributed to the high performance and safety NBI operation in the LHD.

At the National Institute for Fusion Science (NIFS, Japan), neutral beams with the energy up to 15 MW are routinely injected to the hydrogen plasma confined in the Large Helical Device (LHD) using three negative ion beam-lines[1]. Together with the heating beam-lines, a test stand beamline is also operated at NIFS, to gain a better understanding of the physics regulating the formation, extraction and acceleration of negative ions, and their conversion to neutrals. The test stand is based on a 1/3-scaled negative ion source of the same arc-driven source adopted for the LHD beamline, and its acceleration system is a standard triode. The electrode grids are divided into 2 segments having slightly different geometries of the intermediate steering grid (SG) and a common multi slotted grounded grid. In this paper the beamlet optics of both systems was studied by means of an instrumented calorimeter [2] made of carbon fibre composite (CFC) tiles; the tiles were constructed making the propagation of thermal energy along the beam direction largely dominant with respect to the other directions, so that the thermal pattern on the rear plane of the tile has a strong correlation with the power profile deposited by the beamlets on the beamfacing plane of the tile. Infrared image of this thermal pattern is analysed to study the beamlet characteristics obtained in different source conditions [3]. The experimental measurements are then compared with 3D numerical simulations, in order to validate the codes, and to assess their degree of reliability. The simulations show a satisfactory agreement with the experimental results; some minor discrepancies are highlighted and their possible causes are addressed in the manuscript.

The thermo - mechanical analysis and the mechanical design of the accelerator of MITICA (i.e. the full size protot ype of the ITER neutral beam injector under contruction at RFX) are based on the calculation of the power deposition induced by particle impacts. This calculation is performed by EAMCC, a relativistic particle tracking code based on the Monte - Carlo method for describing collisions inside the accelerator, under prescribed electric and magnetic fields. The magnetic field maps are produced by 3D codes, while the electric field maps come from the 2D axisimmetric code SLACCAD. In order to perform a multi - beamlet analysis, which allows to take into account the beamlet - beamlet repulsion, and to consider other effects neglected under the hypotesis of axisimmetric geometry, a fully 3D version of the code is required. Moreover, an accurate description of the variation of the two fields inside the accelerator requires an electric and a magnetic fields maps with a sufficiently fine mesh, but the size (in terms of required memory) of these 3D maps does not allow to extend the domain of an accurate simulation over a single beamlet. In this paper a modified version of EAMCC, fully 3D, capable of modifying the mesh of the 3D maps and of dealing with uneven meshes is presented. A finer mesh is used just in the regions where a more detailed description of the fields is required, for a more realistic simulation. A comparison between the original code and the modified version is presented at first, as a validation of the modifications introduced in the latter. Subsequenlty, the main results of a single - beamlet analysis per formed with the two versions of the code are shown and the differences between the 2D and the 3D simulations discussed. The last part is dedicated to the multi - beamlet simulation of the accelerator.

Beamlet - monitoring experiment with a combination of the NIFS teststand and "mini - STRIKE", which is a test device of the beam monitor to be installed at "SPIDER", has been carried out as an international collaboration program between NIFS and Consorzio RFX . The monitor plates of the mini - STRIKE are made of one - directional CFC tiles, whose fib e res are directed to the beam axis and the beamlet footprints exposed on the plates are observed with IR camera on the opposite side of the exposed surface as shown in Fig. 1. A negative ion source with the a half size of the LHD source is applied for this experiment, and the monitor plates are placed at 0.76 m apart from the exit of the grounded grid of the ion source. The beamlets are extracted from 5 (H) x 3 (V) apertures of the plasma grids by masked with Mo plates. Each beamlet distribution is clearly identified as a thermal image on the monitor plate, and beamlet tilting has been observed. The titling is affected with electron deflection field and the tilting directions flip alternatively row by row. In this talk, we compare the tilting characteristic s with the beamlet - monitoring results obtained at the LHD beamline [1] as well as simulation results of the beamlets.