Nicola Pilan

The neutral beam heating systems for ITER will be based on radio-frequency driven sources, accelerating a negative ion beam current of tens of amperes for up to one hour. In its second operation phase SO-2, the full-scale prototype source SPIDER was operated for the first time in multi-beamlet configuration at ∼70 kW driver ⁻¹ , extracting a H ^— current density of 215 A m ⁻² over ¼ of the extraction area at 0.3 Pa source filling pressure, providing a total accelerated beam power of ∼700 kW. Scaling of the results confirms the specifications of the on-going RF generators upgrade, indicating that target current density of 330 A m ⁻² in hydrogen is achievable by increasing the RF power to 100 kW driver ⁻¹ . Beam uniformity was assessed at the beamlet-group scale under moderate underperveance conditions: at high power row-averaged beamlet currents were within ±10% for fifteen out of sixteen rows. A lower homogeneity was observed at the single-beamlet level. The influence of the magnetic filter field on beam uniformity was also characterized, confirming the beneficial effect of multi-driver operation on homogeneity at the extraction region. A caesium conditioning procedure was developed and validated for the partial-source configuration. An anomalous amplification of the acceleration current peculiar of the vacuum-insulated source design specific to ITER sources, attributed to stray backstreaming charges and surface emission effects, was identified and mitigated by electrostatic shielding. Together with the ongoing installation of new solid-state RF generators and an improved pumping system, the results presented in this work open the way to full-source operation in the next experimental campaign and yield findings of direct relevance to the ITER injector design.

This article summarizes 15 years of experimental activity at the High-Voltage Padova Test Facilities in support of the design and commissioning of the ITER neutral beam injector (NBI). High-voltage DC tests up to 800 kV were carried out in high and medium vacuum on electrode gaps ranging from a few millimeters to 150 mm. Systematic experiments on stainless steel electrodes in high vacuum validated a probabilistic model for predicting the maximum hold-off voltage of generic multielectrode systems and confirmed the presence of a significant total-voltage effect for gap lengths above a few centimeters. The spatial distribution of power dissipation during microdischarges in large vacuum gaps suggests charge accumulation points and mutual charged-particle exchange between electrodes to interpret the experimental observations. Time-resolved X-ray measurements indicate that, in a significant fraction of vacuum breakdown events, bursts of electrons impacting the anode act as precursors, with electron-stimulated desorption playing a key role in discharge development and voltage collapse. In medium vacuum, tests in Ar, H₂, and N₂ characterized the dependence of the hold-off voltage with pressure up to the left branch of the Paschen curve and clarified the influence of static magnetic fields on Townsend discharge. Long-duration tests under stationary electric fields up to 60 kV/mm revealed irreversible switch-on phenomena with microampere-level currents on short-gap configurations. Finally, acceptance tests on solid insulators, optical fibers, and spark gaps are reported, and the main lessons learned and open issues are discussed.

The High Voltage Padova Test Facility (HVPTF) is an experimental device for investigating high voltage (HV) Direct Current insulation in a vacuum, in support of the development of the prototype of a neutral beam injector for ITER, named MITICA. Inside a high vacuum environment, two stainless steel electrodes, spaced by a few centimeters gap, can achieve a voltage difference of up to 800 kV. During the electrode conditioning process, small current bursts, known as micro-discharges (MDs) occur, accompanied by the emission of X-rays, and of an overall increase in gas emission measurement. This study focuses on the fine dynamics of the MD phenomenon measured by the current signal and by the X-ray detectors, in an attempt to provide a possible physical interpretation of this mechanism. In particular, it appears that the MD fine dynamics are related to the growing asymmetry between the MD current values collected by the anode and cathode, observed during the electrode conditioning. The role of the vacuum chamber, acting as a third electrode, is investigated by an updated toy model, which evaluates the different current contributions collected from the two polarized electrodes, particularly those generated by secondary electron (SE) emission. The ratio of anode to cathode currents computed by the toy model closely matches the experimental one. Further experimental observations supporting the hypothesis of anode gas emission and expansion are discussed. Finally, a first study about the statistical distribution of the time intervals between successive MD suggests the occurrence of almost two different trigger mechanisms, evolving during the electrode conditioning process.

The electrical insulation of the MITICA Beam Source at 1 MV is a challenging issue, which has not been fully addressed so far on the basis of experimental results and of theoretical models available in literature. Being MITICA the full-size prototype of the Heating Neutral Beam Injector for the ITER fusion experiment, its electrical insulation is constituted just by vacuum gaps and alumina insulators, since other insulating materials such as SF6 gas or fibreglass-reinforced plastic (FRP) would be quickly degraded by the expected neutron flux produced by fusion reaction. Extrapolations based on HV tests on reduced-scale models have recently indicated the risk of electrical breakdowns in the vacuum gap between electrodes nominally operating at -1 MV and the vacuum vessel (at ground potential). The risk of electrical breakdown can be mitigated by introducing an intermediate Electrostatic Shield (ES), which essentially is an equipotential (metallic) enclosure surrounding the HV electrode, so as to divide the vacuum gap in two independent insulating gaps of 400 kV and 600 kV respectively. However, for optimal negative ion production, the ion source shall operate in H₂ or D₂ at a pressure of ∼ 0.3 Pa and unavoidably produces a flow of gas leaking out in the surrounding vacuum. Thus, the presence of an intermediate shield can substantially increase the background gas pressure in the vacuum gaps, and, due to the large gap length (0.6 m), exacerbate the risk of breakdown when the pressure approaches the conditions of Paschen-type discharges. In addition to this, RF-induced breakdowns were found on the rear side of the ion source during the operation of the prototype source SPIDER, which were somewhat correlated to a relatively high hydrogen pressure in that area. For these reasons, a structure capable of constituting a full equipotential barrier all around the BS and, at the same time, having sufficient gas conductivity (breathability) to allow efficient pumping of background gas, has been designed. In the first part of the paper, the requirements and design optimization of a breathable module of the intermediate ES are described. Then, an experimental campaign for the validation of the electrode implementation the test configurations and the experimental procedure is discussed.

The switch-on effect of the current emission has been known for long time. This effect consists of the existence of a switch-on voltage, above which current emission from the cathode makes a sudden transition to higher value (one–two orders of magnitude). After the switch-on event, the high current emission status remains even at lower voltages. Our experiments have shown that this switch-on event occurs not only when the voltage (i.e., the cathode electric field) exceeds a certain value, but also occurs applying a relatively low electric field ( $\approx$ 40 MV/m) for a long time (tens–hundreds of hours). These transitions are interpreted as changes in the structure of the metal–insulator (MI) potential barrier present, for example, in stainless steel electrodes. This article presents a first characterization of this effect in terms of transition time, current, and voltage level for electrodes made of different materials and/or with different surface treatments. Considerations are finally exposed to explain this switch-on effect as a consequence of the accumulation of electric charge at the MI cathode interface.

In the context of magnetic confinement fusion, the Megavolt ITER Injector and Concept Advancement (MITICA) project is supported by another experiment called High-Voltage Padova Test Facility (HVPTF): a device designed to improve the knowledge of high voltage holding in a vacuum. HVPTF includes a vacuum chamber containing two electrodes separated by an insulator or a vacuum gap, across which electrical discharges can develop. The electrodes are powered by two independent Cockcroft–Walton power supplies, allowing for a total voltage difference up to 800 kV. The current, voltage, and pressure inside the chamber are monitored at a sampling rate of 100 Hz. Studying the physics behind electrical discharges is fundamental to prevent their development in MITICA, where they can lead to structural damage to the experimental components. For this reason, in April 2022, a gas electron multiplier (GEM) detector was installed at HVPTF, allowing to perform X-ray measurements resolved in time, space, and energy at high rates ( $ {>} 1~\text{MHz}$ ) without incurring severe pile-up problems. This work describes the analysis of data coming from current, voltage, and pressure sensors, and the X-ray GEM (XR-GEM) detector at HVPTF during two 2022 experimental campaigns. Correlations between sensor signals with the X-ray measurements have been highlighted. Experimental results indicate the presence of characteristic X-ray emissions from anode atoms during electrical discharge events in the needle–plane electrode configuration. An algorithm was developed for generating synthetic detector data based on user-defined X-ray emissions within the vacuum chamber and performing tomographic reconstructions using actual or synthetic detector data. Preliminary results indicate possible gas emissions from the anode’s surface during electrical discharge events.

The article outlines the development of a diagnostic system designed to pinpoint the origins of X-ray emissions in high-voltage direct current (HVdc) equipment isolated by large vacuum gaps. Electrons extracted from cathodic surfaces through field emission (FE) processes are accelerated to energies below 1 MeV, generating bremsstrahlung radiation upon impact with anodic surfaces. This radiation constitutes one of the main evidence that high-voltage conditioning is taking place on a long gap isolated system and with voltages above 50–100 kV. The development of MITICA, the prototype for ITER’s neutral beam injector (NBI), necessitates diagnostic systems capable of monitoring the progression of high-voltage conditioning in the electrostatic accelerator. Detectors can be placed in atmospheric air to monitor the evolution of the high-voltage conditioning of the electrostatic accelerator through the metallic vacuum vessel. The adoption of an X-ray collimator embedding a cerium-doped lutetium-based scintillation crystal has been proposed to cope with this target. A prototype of a single-pixel collimator (SPC) with a lead shield and variable geometry has been produced and tested at the high-voltage Padova test facility (HVPTF). High-voltage assessments, reaching up to 780 kVdc, were conducted to refine the collimator design. A multipixel variable-geometry system, utilizing the same technology, is recommended for monitoring the vacuum high-voltage conditioning of the MITICA electrostatic accelerator.

This paper presents a first characterization of direct current Switch-On effect, for high voltage, high vacuum electrodes. Considerations are exposed to explain this Switch-On effect as a consequence of the accumulation of electric charge at the MI cathode interface.