Efficiency of RF plasma generation for fusion relevant ion sources

In view of a growing global energy demand, harnessing nuclear fusion as an energy source is an appealing prospect. The fusion experiment ITER aims to demonstrate the technological and scientific feasibility of a positive power balance via the thermonuclear fusion within a magnetically confined plasma. It is intended as an intermediate step towards a fusion power plant and will be followed by the DEMOnstration power plant DEMO, envisaged to be the first fusion device to actually generate net electricity. To achieve the high plasma temperatures (> 10${}^8$ K) required to sustain the fusion process in these devices, powerful heating systems such as the neutral beam injection based on negative ions (NNBI) are required. This system works as follows: in an ion source operating a low pressure deuterium plasma, negative deuterium ions are created. The negative ions are extracted, accelerated, subsequently neutralized and then injected into the fusion plasma, providing both heating power and the capability to externally drive a plasma current. For ITER, beams with a particle energy of 1MeV and an accelerated ion current of 40 A have to be provided for pulse lengths of up to one hour. In view of DEMO, demands corresponding to or even exceeding the ITER parameters are expected. In particular, a higher pulse length of at least two hours and an increased wall-plug efficiency of the NNBI system up to 60% will be required. In addition, a high reliability of the reactor and all its sub-systems is a key issue, since DEMO is intended as a prototype for a fusion power plant. Hence, the NNBI systems have to rely on reliable sources for negative ions which are capable to meet these demands. At NNBI ion sources for ITER, inductively coupled plasmas (ICPs) in D${}_2$ at a pressure of 0.3 Pa will be operated within so-called drivers, where electron densities of the order of 10${}^{18}$ m${}^{-3}$ and electron temperatures around 10 eV are reached. One source will comprise eight drivers, each supplied with an RF power up to 100 kW at an excitation frequency of 1 MHz in order to achieve the required parameters. Such a high power level poses strong demands on the RF generators and circuits and may thus limit the long term reliability of the ion source. In view of DEMO, it is therefore highly beneficial to reduce the RF power consumption, while retaining the plasma parameters relevant for an efficient negative ion production. The latter mainly comprise a high density of both positive ions and atoms, which serve as reactants in the surface conversion process utilized to generate negative ions. In addition, a high RF power transfer efficiency is a prerequisite, which implies that a high fraction of the electrically provided RF power is coupled to the plasma itself, while the heat loads within the RF system due to ohmic losses - potentially limiting the reliability of the source - are low. In this context, alternatives to ICPs driven at 1MHz which are considered promising in view of an efficient plasma generation at ion sources are fundamentally assessed. Apart from deuterium, these investigations are also conducted in hydrogen, since ion source test beds are frequently operated in H${}_2$ as well, mainly for diagnostic purposes. The main framework of the followed approach is that the general concept of the ITER ion source may not be fundamentally altered. This ensures an envisaged high return of experience from ITER NNBI to DEMO. Two alternative plasma generation methods are specifically considered: the first one are ICPs driven at a higher frequency. Investigations conducted in noble gases have shown that the excitation frequency can strongly affect the power absorption by the plasma. However, the driving frequency of the inductively heated ion source H${}_2$/D${}_2$ plasma has never been specifically optimized regarding an efficient production of negative ions. The second assessed concept are helicon discharges operated at low magnetic fields ($\le$ 12 mT), which rely on plasma heating via propagating electromagnetic waves. Due to their capability to reach high plasma densities with relatively moderate RF power at low pressures, they have raised interest regarding an application at ion sources for fusion among several groups. However, at conditions relevant for RF driven ion sources for NNBI, no investigations have been reported up to now which were capable to demonstrate the efficiency of H${}_2$/D${}_2$ helicons in direct and quantitative comparison to ICPs. Performing such a fundamental assessment directly at a large ion source experiment is not feasible due to the lack of flexibility and diagnostic access, though. The investigations within this work are therefore conducted at a versatile, smallscale laboratory setup which is operated at lower RF powers than ion sources. It can be equipped with a helical coil for inductive heating or with a Nagoya type-III antenna for low-field helicon operation, without changing the general discharge geometry. RF generated plasmas applying excitation frequencies between 1 and 4 MHz and RF powers up to 2 kW in H${}_2$ and D${}_2$ at operational conditions relevant for ion sources are considered. Due to its comprehensive diagnostic setup, the RF power transfer efficiency and discharge parameters most crucial for an efficient and reliable plasma generation in view of ion sources for NNBI are accessible. This allows for a direct comparison of the proposed alternative plasma generation techniques to the baseline case of ICPs driven at 1 MHz and thus, for an assessment of their potential to reduce the RF power consumption. The final aim of this fundamental approach is to identify whether any of the proposed alternative methods is promising enough to qualify for a further assessment and application at a larger scale. In order to ensure the resilience of such an evaluation, the fundamentals of the different heating mechanisms in the light molecular gases H${}_2$/D${}_2$ need to be analysed thoroughly. Such has never been conducted up to now, since the basic characteristics of low pressure low temperature plasma heating are typically investigated in noble gas discharges like argon. In course of these novel studies, the dependences of the main discharge parameters on operational parameters like pressure, RF power and external magnetic field are regarded, with a particular focus on the isotopic differences between hydrogen and deuterium with respect to the plasma heating mechanisms.