Benchmark of single beamlet analysis to predict operational parameter for ITER in Japan - Italy joint experiments

The negative ion accelerators for the ITER neutral beam system has been designed by using the conventional beam codes for the positive ion accelerators because mechanism of meniscus formation of negative ions has not been clarified well. In this way, the operational window of the negative ion accelerators can be simply predicted, but the grid heat loading by peripheral beam (beam halo) could not be simulated. To understand the applicable range of the conventional codes to the negative ion accelerators, and then to understand the negative ion meniscus consequently, the benchmark tests between the conventional codes (BEAMORBT (BO) [1] and SLACCAD (SC)[2]) and the experiments have been performed under a joint experiment by RFX and QST on the Negative Ion Test Stand in Naka. Figure 1 shows one numerical results as an example with the grid configuration. The extraction voltage (Vext) is applied between the plasma grid (PG) and the extraction grid (EXG). The acceleration voltage (Vacc) is applied between EXG and the grounded grid (GG). Figure 2 shows the beam divergence angle as a function of Vext at Vacc of 20 kV. Vext with the minimum divergence angle are shown around 4.0 kV both in the numerical codes and the experimental result. The operational window can be predicted in the numerical codes. However, the absolute values of the divergence angle of the experiment was larger than those of the numerical result. One of reasons will be that the numerical code cannot represent the beam halo. Even in the numerical results, the divergence angle in the BO is slightly lower than that of the SC. To understand the meniscus effect to the beam divergence, the relationship between meniscus depth ?zmenis and beam divergence in the numerical results was compared as shown in Figure 3. The details are reported in the paper.