Physics basis and future trends for negative ion sources.

The explanation of hydrogen negative ion formation in the hydrogen plasma volume is based on the hypothesis that the precursors of H(-) ions are rovibrationally excited molecules. It will be shown that this hypothesis is confirmed by the agreement between predicted and measured H(-) densities. The reason for the existence of an optimum pressure for the negative ion density in volume sources is discussed. The role of the magnetic filter in contemporary negative ion sources is analyzed. Experiments indicating the effect of the ion source surfaces (plasma electrode, collar, walls) will be described. It is concluded that vibrationally excited molecules, produced by recombinative desorption from surfaces, play a significant role. It is shown that a high H(-) ion emission from a high work function surface is very doubtful at the present state of knowledge. Therefore it is considered that the success of the cesium-free accelerator ion sources operating in the 100 mA/cm(2) range is due to production of vibrationally excited molecules on the plasma electrode or collar surface. In the collar case, the negative ion production is enhanced due to the absence of plasma in the production region; thus mutual neutralization loss is canceled. The physics of surface production of H(-) ions and some unexplained features related to it will be discussed.

Plasma perturbation induced by laser photodetachment.

The plasma dynamics arising from laser photodetachment is discussed herein theoretically and experimentally. The hybrid fluid-kinetic model, where the positive ions and electrons are treated by the fluid theory and the negative ions are treated within the ballistic approximation, is extended and applied to the analysis of densities perturbed by laser photodetachment. The agreement between the theory and measured data confirms the validity of the considered plasma dynamics model. This model, including the positive ion perturbation, shows a good agreement with the time evolution and the spatial distribution of perturbed electron densities which are measured by a Langmuir probe inside and outside the laser beam. From the overshoot in the time evolution of perturbed electron current in the center of the laser beam, the positive ion temperature was found to be in the range 0.1-0.25 eV, while the electron temperature changes from 0.3 to 3.2 eV.

Measurement of H- density in plasma by photodetachment.

Techniques have been developed for measurement of the density of H- in a plasma by photodetachment. Photodetachment is detected by the increase in electron density with no change in positive ion density after a light pulse from a ruby laser. The authenticity of photodetachment signals can be assured by their comparison with known cross sections for photodetachment of H-. Interpretations of photodetachment data are less ambiguous than probe interpretations because photodetachment is not affected by the mass of positive ions and is not limited in usefulness by the Debye distance. Photodetachment measurements with time resolution and spatial resolution are straightforward.

Effect of fast positive ions incident on caesiated plasma grid of negative ion source.

This paper describes the effect on negative ion formation on a caesiated surface of the backscattering of positive ions approaching it with energy of a few tens of eV. For a positive ion energy of 45 eV, the surface produced negative ion current density due to these fast positive ions is 12 times larger than that due to thermal atoms, thus dominating the negative ion surface production instead of the thermal atoms, as considered until now.

Analysis of H atoms in a negative ion source plasma with the non-equilibrium electron energy distribution function.

In negative ion sources for the neutral beam injection, it is important to calculate H atom flux onto the plasma grid (PG) surface for the evaluation of H(-) production on the PG surface. We have developed a neutral (H(2) molecules and H atoms) transport code. In the present study, the neutral transport code is applied to the analysis of the H(2) and H transport in a NIFS-R&D ion source in order to calculate the flux onto the PG surface. Taking into account non-equilibrium feature of the electron energy distribution function (EEDF), i.e., the fast electron component, we have done the neutral transport simulation. The results suggest that the precise evaluation of the EEDF, especially in the energy range 15 eV < E < 30 eV is important for the dissociation rate of H(2) molecules by the electron impact collision and the resultant H atom flux on the PG.

Analysis of the H- ion emissive surface in the extraction region of negative ion sources.

To understand the plasma characteristics in the extraction region of negative H(-) sources is very important for the optimization of H(-) extraction from the sources. The profile of plasma density and electrostatic potential in the extraction region with and without extraction grid voltage are analyzed with a 2D particle in cell modeling of the NIFS-RD H(-) sources. The simulation results make clear the physical process forming a double ion plasma layer (which consists only of positive H(+) and negative H(-) ions) recently observed in the Cs-seeded experiments of the NIFS-R&D source in the vicinity of the extraction hole and the plasma grid. The results also give a useful insight into the formation mechanism of the plasma meniscus and the H(-) extraction process for such double ion plasma.

Influence of the electron cross-field diffusion in negative ion sources with the transverse magnetic field and the plasma-electrode bias.

The physical mechanisms involved in the extraction of H(-) ions from the negative ion source are studied with a PIC 2D3V code. The effect of a weak magnetic field transverse to the extraction direction is taken into account, along with a variable bias voltage applied on the plasma electrode (PE). In addition to previous modeling works, the electron diffusion across the magnetic field is taken into account as a simple one-dimensional random-walk process. The results show that without PE bias, the value of the diffusion coefficient has a significant influence upon the value of the extracted H(-) current. However, the value of this coefficient does not affect qualitatively the mechanism leading to the peak of extracted H(-) ion current observed for an optimum value of the PE bias.