NNBI for ITER: status of long pulses in deuterium at the test facilities BATMAN Upgrade and ELISE

Abstract

For the D-T operational phase of the international fusion experiment ITER, starting in 2035, pulses are planned with Q > 10 and a duration of 400 s and pulses with Q = 5 for up to 3600 s. The two neutral beam injection (NBI) systems will deliver 33.3 MW of heating power to the plasma and will also be an important source for non-inductive current drive. An essential part of the NBI systems is a large RF driven source for negative hydrogen or deuterium ions. In order to fulfil the requirements imposed to the NBI system, the ion source has to deliver an intense, stable and homogeneous large negative ion beam over pulse lengths of 400 s or 3600 s. Till now, long pulses in deuterium at the extracted negative ion current density required for ITER could not be demonstrated. The achievable performance is limited by a high current and a pronounced vertical asymmetry of the co-extracted electrons, both strongly increasing with time, whereas the stability of the negative deuterium ions is not an issue. For reduced performance, i.e. reduced negative ion current, lower and more stable co-extracted electron currents are observed, making possible pulses up to 3600 s (using pulsed extraction). One main aim of ongoing investigations at the two test facilities BATMAN Upgrade (Bavarian test machine for negative ions) and ELISE (extraction from a large ion source experiment) is the development of scenarios stabilizing and symmetrizing the co-extracted electron current in deuterium. These investigations are discussed and the current status of hardware upgrades ongoing at BATMAN Upgrade and ELISE towards the ITER scenario of one-hour steady state extraction is presented.

Ion source developments at IPP: On the road towards achieving the ITER-NBI targets and preparing concepts for DEMO

“…The test facility ELISE is equipped with a half-size ITER source in order to bridge the gap towards the full-size ITER source. The size-scaling route is, for example, illustrated in [8]. The source consists of four RF-drivers two of each connected in series to one RF generator.…”

“…Caesium is evaporated by ovens: one at the back plate of the ion source at BUG, two at the side walls (left and right) at ELISE and three (at the backplate) for the ITER source. The evaporation rate per oven is typically several milligrams per hour [8] and caesium deposits on surfaces from which it is re-distributed in the source by the plasma. To avoid cold spots and thus caesium accumulation in the source the surfaces are temperature controlled; typically the source walls around 40°C and the PG around 150°C.…”

“…20 s plasma with 10 s beam extraction. Over 60 % could be demonstrated in long pulses up to one hour [8,11]. Extending the pulse length at high performance is limited by the higher co-extracted electron current than in hydrogen accompanied by a stronger inhomogeneity in vertical direction and stronger increase with time.…”

“…in addition to the filter field created by a current of several kA flowing vertically through the PG (as in the ITER source) permanent magnets are mounted at the side walls to influence the field topology. Further reduction of the co-extracted electrons is achieved by implementing metallic rods in vertical direction to influence the particle fluxes by biasing these potential rods [8]. In the standard configuration (BUG, ELISE, ITER), a metallic plate (6 mm thick) parallel to the PG with openings for the beamlet groups is mounted in the source 7 mm above the PG being electrically connected to the source walls.…”

Towards the optimization of the Cs evaporation configuration for long pulse operation of negative ion sources

“…In these sources negative ions are mostly produced by surface conversion of H/D atoms onto a metallic converter surface: the conversion efficiency is increased by reducing the work function of the surface through Cs evaporation [5]. The main limiting factor for the achievement of the required performances is represented by the amount of co-extracted electrons and their instability during long pulses, especially in deuterium [6]. This increase of co-extracted electrons during a pulse is attributed to a degradation of the work function of the plasma grid, i.e.…”