Open Access

Adrian Heiler

• Modern high-power negative hydrogen ion sources rely predominantly on the surface production of negative hydrogen ions. For this, a low work function of the extraction electrode surface is required to effectively convert impinging atoms and positive ions from low pressure low temperature hydrogen plasmas into negative ions. The state-of-the-art technique for the generation of low surface work functions is in situ evaporation of the alkali metal Cs, which exhibits a bulk work function of 2.0–2.1 eV. However, the achievement of a temporally stable and reliable low work function coating is challenging in an ion source environment due to the influence of the reactive hydrogen particles (atoms and positive ions) and energetic UV/VUV photons (up to 15 eV) during plasma phases as well as residual gases during vacuum phases. At RF driven ion sources for neutral beam injection systems for nuclear fusion where in the case of the ITER experiment pulse lengths up to one hour are required, temporalModern high-power negative hydrogen ion sources rely predominantly on the surface production of negative hydrogen ions. For this, a low work function of the extraction electrode surface is required to effectively convert impinging atoms and positive ions from low pressure low temperature hydrogen plasmas into negative ions. The state-of-the-art technique for the generation of low surface work functions is in situ evaporation of the alkali metal Cs, which exhibits a bulk work function of 2.0–2.1 eV. However, the achievement of a temporally stable and reliable low work function coating is challenging in an ion source environment due to the influence of the reactive hydrogen particles (atoms and positive ions) and energetic UV/VUV photons (up to 15 eV) during plasma phases as well as residual gases during vacuum phases. At RF driven ion sources for neutral beam injection systems for nuclear fusion where in the case of the ITER experiment pulse lengths up to one hour are required, temporal instabilities of the source performance are a major issue and especially problematic in deuterium operation. Since work function measurements at negative ion sources are challenging and not routinely applied, investigations on the work function dynamics of caesiated surfaces upon exposure to ion source relevant particle and photon fluxes are gained at a dedicated laboratory experiment. The experiment is equipped with a finely adjustable Cs oven allowing in situ caesiation of surfaces installed at a sample holder. A comprehensive set of diagnostics is available for the determination of fluxes to which the surface is exposed in vacuum and plasma phases, and the surface work function is measured by the exploitation of the photoelectric effect. Within this work, the accuracy of the work function diagnostic is significantly improved by lowering the detection limit for photocurrents by several orders of magnitude. Furthermore, the experimental setup is extended by an external plasma source with which the surface under investigation can be exposed to fluxes of specific hydrogen plasma species. The application of the improved work function diagnostic reveals that ultra-low work functions in the range of 1.2–1.3 eV can be reproducibly generated under the typically given non-ultra-high vacuum conditions where water is the dominant residual gas. The work function is found to be dependent on the flux ratio of Cs to water onto the surface, and the achieved work function of significantly below that of bulk Cs is attributed to the formation of Cs oxide layers due to reactions between Cs and water at the surface. The work function behavior upon re-caesiation after operational breaks, upon heat as well as upon exposure to various gas species is studied in detail, allowing the identification of beneficial and detrimental influences on the caesiation process. In order to investigate the role of the different plasma species in the plasma-surface interaction, caesiated surfaces are selectively exposed to VUV photons, hydrogen atoms and positive hydrogen ions using the external plasma source. It is shown that each species can affect the surface separately, demonstrating for the first time that photonic, atomic as well as ionic interactions must be considered in the plasma-surface interaction. The full plasma-surface interaction is investigated by the ignition of plasmas in front of the caesiated surface, where scenarios with plasma pulses ranging from a few seconds up to several hours with and without Cs evaporation both in hydrogen and deuterium are applied. It is shown that ultra-low work function layers can be reliably generated but are not stable upon plasma exposure exceeding one minute, which explains the observed deterioration of the ion source performance in long pulses. During steady-state plasma operation, a minimum work function of 1.8 eV is found to be stable for sufficiently high Cs fluxes onto the surface both in hydrogen and deuterium and independent of the initial work function in the vacuum phase.…