Summary of Progress in U.S. Heavy Ion Fusion Science Research
Abstract not provided
Engineering topics
Publications and source records attributed to Grisham, L R.
Abstract not provided
Vacuum insulation on a large size negative ion accelerator with multiple extraction apertures and acceleration grids for fusion application was experimentally examined and designed. In the experiment, vacuum insulation characteristics were investigated in the JT-60 negative ion source with >1000 apertures on the grid with the surface area of {approx}2 m{sup 2}. The sustainable voltages varied with a square root of the gap lengths between the grids, and decreased with number of the apertures and with the surface area of the grids. Based on the obtained results, the JT-60SA (super advanced) negative ion source is designed to produce 22 A, 500 keV D{sup -} ion beams for 100 s.
In a multi-aperture multi-grid accelerator of the ITER neutral beam injector, the beamlets are deflected due to space charge repulsion between beamlets and beam groups, and also due to magnetic field. Moreover, the beamlet deflection is influenced by electric field distortion generated by grid support structure. Such complicated beamlet deflections and the compensations have been examined utilizing a three-dimensional beam analysis. The space charge repulsion and the influence by the grid support structure were studied in a 1/4 model of the accelerator including 320 beamlets. Beamlet deflection due to the magnetic field was studied by a single beamlet model. As the results, compensation methods of the beamlet deflection were designed, so as to utilize a metal bar (so-called field shaping plate) of 1 mm thick beneath the electron suppression grid (ESG), and an aperture offset of 1 mm in the ESG.
Acceleration of a 500 keV beam up to 2.8 A has been achieved on a JT-60U negative ion source with a three-stage accelerator by overcoming low voltage holding which is one of the critical issues for realization of the JT-60SA ion source. In order to improve the voltage holding, preliminary voltage holding tests with small-size grids with uniform and locally intense electric fields were carried out, and suggested that the voltage holding was degraded by both the size and local electric field effects. Therefore, the local electric field was reduced by tuning gap lengths between the large size grids and grid support structures of the accelerator. Moreover, a beam radiation shield which limited extension of the minimum gap length was also optimized so as to reduce the local electric field while maintaining the shielding effect. These modifications were based on the experiment results, and significantly increased the voltage holding from <150 kV/stage for the original configuration to 200 kV/stage. These techniques for improvement of voltage holding should also be applicable to other large ion sources accelerators such as those for ITER.
The voltage gradient which can be sustained between electrodes without electrical breakdowns is usually one of the most important parameters in determining the performance which can be obtained in an electrostatic accelerator. We have recently proposed a technique which might permit reliable operation of electrostatic accelerators at higher electric field gradients, perhaps also with less time required for the conditioning process in such accelerators. The idea is to run an electric current through each accelerator stage so as to produce a magnetic field which envelopes each electrode and its electrically conducting support structures. Having the magnetic field everywhere parallel to the conducting surfaces in the accelerator should impede the emission of electrons, and inhibit their ability to acquire energy from the electric field, thus reducing the chance that local electron emission will initiate an arc. A relatively simple experiment to assess this technique is being planned. If successful, this technique might eventually find applicability in electrostatic accelerators for fusion and other applications.
In JAEA, research and developments to realize high power accelerator (1 MeV, 40 AD{sup -} ion beams for 3600 s) for ITER have been carried out experimentally and numerically utilizing a five stage MAMuG (Multiaperture, Multigrid) accelerator. In this paper, the extension of the gap length, which is required to improve the voltage holding capability, is examined in two dimensional beam optics analyses and also from view point of stripping loss of ions. In order to suppress excess power loadings due to the direct interception of negative ions, which is issued in long pulse tests, the beamlet deflection is analyzed in three dimensional multibeamlet analyses. The necessary modifications shown above are applied to the MAMuG accelerator for coming long pulse tests in JAEA and ITER.
Developments of the large negative ion source have been progressed in the high-energy, high-power, and long-pulse neutral beam injector for JT-60 Super Advanced. Countermeasures have been studied and tested for critical issues of grid heat load and voltage holding capability. As for the heat load of the acceleration grids, direct interception of D{sup -} ions was reduced by adjusting the beamlet steering. As a result, the heat load was reduced below an allowable level for long-pulse injections. As for the voltage holding capability, local electric field was mitigated by tuning gap lengths between large-area acceleration grids in the accelerator. As a result, the voltage holding capability was improved up to the rated value of 500 kV. To investigate the voltage holding capability during beam acceleration, the beam acceleration test is ongoing with new extended gap.
Voltage holding capability of JT-60 negative ion source that has a large electrostatic negative ion accelerator with 45 cmx1.1 m acceleration grids was experimentally examined and improved to realize 500 keV, 22 A, and 100 s D{sup -} ion beams for JT-60 Super Advanced. The gap lengths in the acceleration stages were extended to reduce electric fields in a gap between the large grids and at the corner of the support flanges from the original 4-5 to 3-4 kV/mm. As a result, the voltage holding capability without beam acceleration has been successfully improved from 400 to 500 kV. The pulse duration to hold 500 kV reached 40 s of the power supply limitation.
Voltage holding in high voltage electrostatic accelerators is a longstanding problem which limits the practically obtainable performance of most systems. This paper proposes an idea for improving the voltage holding in electrostatic accelerators by suppressing breakdowns between successive stages of an accelerator. The idea consists of flowing electric currents along the accelerator electrodes to produce magnetic fields which envelope each electrode and its support structure, so as to prevent very low energy electrons from leaving the surfaces of electrodes and subsequently picking up energy from the electric field. In order to be useful in some applications, this magnetic insulation would only need to produce modest gains in voltage holding capability, and its utility can be easily tested experimentally as described herein.
Fusion neutral beam systems have conventionally used gas cells of the beam isotope as neutralizers. In the design of negative ion beam systems for ITER, and likely also for future reactors, the large gas efflux arising from these cells can lead to many problems. We discuss the possibility of decreasing the gas load by using a lithium jet neutralizer, which would also afford higher neutralization efficiency.
Excess heat loads to accelerator grids limit extension of pulse length in operation of the large negative ion sources with multi aperture accelerator. Part of the heat loads is caused by interception of deflected beamlets due to their space charge repulsion. In this paper, a beamlet steering technique using aperture offset was examined for compensation of the beamlet deflections utilizing a three dimensional beam analysis simulating the D{sup -} negative ion source of JT-60 U. The beamlet deflection was analyzed in detail using fifty beamlets, which were extracted from apertures arranged in a lattice pattern of 10x5. The simulation showed successful compensation of the beamlet deflection by aperture offsets defined according to the thin lens theory. Even if the beam energy was changed, the necessary aperture offset would not be changed maintaining the perveance and a ratio of extraction and acceleration voltage. In JT-60 U, it was shown that the aperture offset of less than 1.0 mm would be enough to compensate the repulsion of all beamlets. When the magnetic field was applied for suppression of co-extracted electrons, necessary aperture offset was estimated to be {+-}0.5 mm for 500 keV D{sup -} ion beam in JT-60 U, in addition to the offset for the space charge repulsion. This result showed good agreements with the previous experimental results and design study of the JT-60 U N-NBI.