Diffusion of boron and phosphorus in dry 02, wet At, and wet 02 ambients was investigated to ootain a formula describing the oxidation enhanced diffusion. At low impurity concentration, the diffusion coefficient is expressed as a function of oxide growth rate, depth, and temperaturewhere dXo/dt is in ~m/hr and X is in ~m. F value is 6.0 X 10 -2 cmS/hr for boron and 5.2 • 10-~ cm2/hr for phosphorus. This equation can be explained by a dual mechanism involving both vacancy and interstitial silicon atoms. A depth dependence of the diffusion coefficient is explained by a model that excess interstitial silicon atoms are captured by vacancies migration into bulk silicon. At high impurity concentration, less enhanced diffusion was observed.

The ITER Neutral Beam Test Facility (NBTF), called PRIMA (Padova Research on ITER Megavolt Accelerator), is hosted in Padova, Italy and includes two experiments: MITICA, the full-scale prototype of the ITER heating neutral beam injector, and SPIDER, the full-size radio frequency negative-ions source. The NBTF realization and the exploitation of SPIDER and MITICA have been recognized as necessary to make the future operation of the ITER heating neutral beam injectors efficient and reliable, fundamental to the achievement of thermonuclear-relevant plasma parameters in ITER. This paper reports on design and R&D carried out to construct PRIMA, SPIDER and MITICA, and highlights the huge progress made in just a few years, from the signature of the agreement for the NBTF realization in 2011, up to now-when the buildings and relevant infrastructures have been completed, SPIDER is entering the integrated commissioning phase and the procurements of several MITICA components are at a well advanced stage.

The physics design of the accelerator for the heating neutral beamline on ITER is now finished and this paper describes the considerations and choices which constitute the basis of this design. Equal acceleration gaps of 88 mm have been chosen to improve the voltage holding capability while keeping the beam divergence low. Kerbs (metallic plates around groups of apertures, attached to the downstream surface of the grids) are used to compensate for the beamlet–beamlet interaction and to point the beamlets in the right direction. A novel magnetic configuration is employed to compensate for the beamlet deflection caused by the electron suppression magnets in the extraction grid. A combination of long-range and short-range magnetic fields is used to reduce electron leakage between the grids and limit the transmitted electron power to below 800 kW.

The ITER Neutral Beam Test Facility (PRIMA) is presently under construction at Consorzio RFX (Padova, Italy). PRIMA includes two experimental devices: an ITER-size ion source with low voltage extraction, called SPIDER, and the full prototype of the whole ITER Heating Neutral Beams (HNBs), called MITICA. The purpose of MITICA is to demonstrate that all operational parameters of the ITER HNB accelerator can be experimentally achieved, thus establishing a large step forward in the performances of neutral beam injectors in comparison with the present experimental devices. The design of the MITICA extractor and accelerator grids, here described in detail, was developed using an integrated approach, taking into consideration at the same time all the relevant physics and engineering aspects. Particular care was taken also to support and validate the design on the basis of the expertise and experimental data made available by the collaborating neutral beam laboratories of CEA, IPP, CCFE, NIFS and JAEA. Considering the operational requirements and the other physics constraints of the ITER HNBs, the whole design has been thoroughly optimized and improved. Furthermore, specific innovative concepts have been introduced.

Hydrogen negative ion beams of 507 keV, 1 A and 486 keV, 2.8 A have been successfully produced in the JT-60U negative ion source with a three-stage accelerator by overcoming a poor voltage holding of the accelerator with large-size grids of ∼2 m2. This is the first result of H− beam acceleration up to 500 keV at a high current of over 1 A. In order to improve the voltage holding capability, the breakdown voltages of the large-size grids and small-size electrodes with uniform and locally strong electric fields were examined by changing the gap length. It was found that the voltage holding of the large-size grids was below half of that of the small-size electrodes with a uniform electric field which was used in the design of the accelerator. This degradation was found to be caused by the local electric field concentrations in addition to the size. Based on the results of the voltage holding tests and beam optics calculations, the gap lengths of the large-size grids were tuned to have a capability to sustain 600 kV. As a result, the gap tuning realized stable voltage holding during beam accelerations without significant degradations of the beam optics and stripping loss. These results indicated that stable 500 keV beam accelerations required for JT-60SA are feasible and this gap tuning is also applicable for the design of ITER accelerator.