Abstract. In the next generation of fusion reactors, such as DEMO, neutral beam injectors (NBIs) of high energy (0.8-1 MeV) deuterium atoms with high wall-plug efficiency (>50%) will be required to reach burning plasma conditions and to provide a significant amount of current drive. The present NBI system for DEMO assumes that 50 MW is delivered to the plasma by 3 NBIs. In the Siphore NBI concept, negative deuterium ions are extracted from a long, thin ion source 3 m high and 15 cm wide, accelerated and subsequently photo-neutralized. This requires the development of a new generation of negative ion sources. At the Swiss Plasma Center, a novel radio frequency helicon plasma source, based on a resonant network antenna source delivering up to 10 kW at 13.56 MHz, has been developed and is presently under study on the Resonant Antenna Ion Device (RAID). RAID is a linear device (1.9 m total length, 0.4 m diameter) and is equipped with an extensive set of diagnostics for full plasma characterization. In this work, the principles of operation of resonant antennas as helicon sources are introduced. We present absolute spectroscopy, Langmuir probe, and interferometry measurements on helicon plasmas. We characterize the performance of the source in terms of hydrogen/deuterium dissociation and negative ion production as a function of the input power. Furthermore, first results with the helicon birdcage antenna installed on the Cybele negative ion source at CEA-IRFM are presented, as a first step towards the validation of the Siphore concept.

A new generation of Neutral Beam (NB) systems will be required in future fusion reactors, such as DEMO, able to deliver high power, in total up to 50 MW, with high, 800 keV or higher, neutral energy. Only negative ion beams may be able to attain this performance, engendering strong research focus on negative ion production from both surface and volumetric plasma sources. A novel helicon plasma source, based on the resonant birdcage network antenna configuration, is currently under study at the Swiss Plasma Center (SPC), before installation on the Cybele negative ion source at CEA-IRFM. This source is driven by up to 10 kW at 13.56 MHz, and is being tested on a linear Resonant Antenna Ion Device (RAID). Passive spectroscopic measurements of the first three Balmer lines α, β and γ and of the Fulcher-α bands were performed with an f/2 spectrometer, for both hydrogen and deuterium. Multiple viewing lines and an absolute intensity calibration were used to determine the plasma radiance profile, with a spatial resolution < 3 mm. A minimum Fisher regularisation algorithm is applied to obtain the absolute emissivity profile for each emission line, for cylindrical symmetry that was experimentally confirmed. An uncertainty estimate of the inverted profiles was performed using a Monte Carlo approach. Finally, an RFcompensated Langmuir probe was inserted to measured the electron temperature and density profiles. The absolute line emissivities are interpreted using the collisional-radiative code YACORA which estimates the dissociation degree and the distribution of the atomic and molecular species, that includes the negative ion density. This paper reports the results of a power scan up to 5 kW in conditions satisfying Cybele requirements for the plasma source, namely a low neutral pressure, p ≤ 0.3 Pa and magnetic field B ≤ 150 G.

Negative ions are characterized in the helicon plasma source RAID (Resonant Antenna Ion Device) at the Swiss Plasma Center by means of Cavity Ring-Down Spectroscopy (CRDS) and Langmuir Probe (LP)-assisted laser photodetachment. A high density and axially homogeneous plasma column is produced via a RF antenna able to sustain the propagation of helicon waves in a steady state regime. An electron density n e ∼ = 2.0 × 10 18 m −3 in H 2 plasma at 0.3 Pa and 3 kW of input power is measured in the center of the plasma column by LP and microwave interferometry. The electron temperature profile is peaked on axis reaching T e ≈ 5 eV and decreasing to 1.5 eV at r = 0.05 m. Thus, a hot core region forms where H 2 molecules are rovibrationally excited (H 2 (ν)), and a cold edge, where low energy electrons can attach to H 2 (ν) and produce H − ions by dissociative attachment. In this work we use LP-assisted laser photodetachment and CRDS diagnostics to measure H − and D − radial density profiles and how they depend on source parameters. We show that negative ions are distributed on a shell of 0.06 m radius with a peak value of ∼ 2.0 × 10 16 m −3 in H 2 plasma. These results suggest that, although substantial technical development is needed, helicon plasmas could be considered as a possible candidate as sources of negative ions for future NBIs.

The negative-ion based neutral beam injector for heating and current drive of the ITER plasma (ITER HNB) is under development, at present focusing on the optimization of the full-scale plasma source in the SPIDER test stand. The production of H- or D- ions in the ion source is based on the low work function surfaces obtained by caesium evaporation. This paper describes the caesium conditioning procedure and the corresponding beam performances during the first operation of SPIDER with caesium. Technical solutions to overcome present limitations of the test stand are described. The influence of source parameters on the caesium effectiveness was investigated in short beam pulse operation; with total RF power of 400 kW and filling pressure below 0.4 Pa, and a limited number of extraction apertures, a negative ion current density of about 200 A/m2 was extracted in hydrogen, with beam energy lower than 60 keV. Beam optics and beam uniformity were assessed thanks to the acceleration of isolated ion beamlets. A possible procedure to accelerate a uniform beam was demonstrated at low RF power. The results obtained in this first investigation provided key indications on the operation of one of the largest existing sources of accelerated negative hydrogen-like ions.

In parallel to the developments dedicated to the ITER neutral beam (NB) system, CEA-IRFM with laboratories in France and Switzerland are studying the feasibility of a new generation of NB system able to provide heating and current drive for the future DEMOnstration fusion reactor. For the steadystate scenario, the NB system will have to provide a high NB power level with a high wall-plug efficiency (η∼60%). Neutralization of the energetic negative ions by photodetachment (so called photoneutralization), if feasible, appears to be the ideal solution to meet these performances, in the sense that it could offer a high beam neutralization rate (>80%) and a wall-plug efficiency higher than 60%. The main challenge of this new injector concept is the achievement of a very high power photon flux which could be provided by 3 MW Fabry-Perot optical cavities implanted along the 1 MeV D − beam in the neutralizer stage. The beamline topology is tall and narrow to provide laminar ion beam sheets, which will be entirely illuminated by the intra-cavity photon beams propagating along the vertical axis. The paper describes the present R&D (experiments and modelling) addressing the development of a new ion source concept (Cybele source) which is based on a magnetized plasma column. Parametric studies of the source are performed using Langmuir probes in order to characterize and compare the plasma parameters in the source column with different plasma generators, such as filamented cathodes, radio-frequency driver and a helicon antenna specifically developed at SPC-EPFL satisfying the requirements for the Cybele (axial magnetic field of 10 mT, source operating pressure: 0.3 Pa in hydrogen or deuterium). The paper compares the performances of the three plasma generators. It is shown that the helicon plasma generator is a very promising candidate to provide an intense and uniform negative ion beam sheet.

No abstract

No abstract

The note presents the first plasma density measurements collected by a novel microwave interferometer in a compact Electron Cyclotron Resonance Ion Sources (ECRIS). The developed K-band (18.5 ÷ 26.5 GHz) microwave interferometry, based on the Frequency-Modulated Continuous-Wave method, has been able to discriminate the plasma signal from the spurious components due to the reflections at the plasma chamber walls, when working in the extreme unfavorable condition λ ≃ L ≃ L (λ, L, and L being the probing signal wavelength, the plasma dimension and the plasma chamber length, respectively). The note describes the experimental procedure when probing a high density plasma (n > 1 ⋅ 10 cm) produced by an ECRIS prototype operating at 3.75 GHz.