The impact of plasma flow on ion temperature (Ti) measurement using retarding field analyzers (RFAs) was experimentally investigated. Experimental results indicated that relatively higher temperature was obtained when the RFA collector was oriented toward the upstream side of the plasma flow, while relatively low Ti was observed on the downstream side. This trend remained consistent regardless of the plasma flow direction to the magnetic field. A simple analytical model that assumes a drifted-Maxwellian distribution explained the results well, suggesting that RFAs overestimate or underestimate Ti due to the plasma flow. While this effect is generally undesirable, it also enables simultaneous evaluation of Ti and plasma flow velocity. In this work, we propose a practical method to determine ion temperature and plasma flow velocity from I–V characteristics that are deformed by plasma flow.
This paper reports X-ray crystal spectroscopy (XCS) on highly charged ions of tungsten impurity within the Large Helical Device (LHD) under plasma conditions with a central electron temperature around 4 keV. By combining measured spectra with Flexible Atomic Code (FAC) calculations and a collisional radiative model weighted by ADAS ionization and recombination rates, emission lines from W 43+ –W 47+ charge states were systematically identified in the 3.7–4.0 Å band. The dataset was constructed by integrating 10 discharges to cover the entire wavelength range, using tungsten introduced reproducibly via coaxial W pellet injection. Notably, the W 47+ line at 3.7691 Å extends the highest charge state observed to date in the LHD. Leveraging the high dispersion characteristics of XCS, three diagnostics directly relevant to ITER-class edge/pedestal plasmas with Te of several keV were demonstrated: (i) Simultaneous measurement of multiple charge states via EUV-X-ray co-observation. This enables monitoring the evolution of tungsten charge state distribution in response to electron temperature variations, supporting impurity transport studies and operational monitoring. (ii) Ion temperature estimation from the Doppler width of the W 46+ 3.8784 Å line showing a core peak, including error propagation via determination of instrument functions and cross-comparison with the charge exchange recombination spectroscopy. (iii) An electron temperature indicator based on the W 46+ 3.8784/3.8956 Å intensity ratio. This exhibits strong sensitivity to Te , and FAC calculations reproduced the observed Te dependence of the line intensity ratio in the Te range of 3–5 keV.
Tungsten spectroscopic studies have been conducted in the Large Helical Device with a pellet injection technique. Spatiotemporal profiles of tungsten density were evaluated using a space-resolved spectrometer, for plasmas with an electron temperature of below 1 keV and electron density of 10 19 –10 20 m −3 . Slice & Stack, a method for reconstructing emissivity, was applied to a line at 191.7 Å, which is a part of the unresolved transition array (UTA) spectrum at 90–250 Å. Tungsten density was obtained using photon emission coefficients of W 17+ –W 27+ , evaluated from collisional-radiative model. The tungsten pellet injected from outside the plasma was first ablated in the edge plasma and subsequently diffused throughout the plasma. This behavior is typical of pellet injection experiments. Furthermore, after an event triggered by neutral beam injection (NBI) breakdown, tungsten accumulated in the core plasma. The radiation power was estimated from the evaluated tungsten density profile and cooling factor dataset, and compared with bolometer measurement. This sequence of processes would be useful for validating atomic data of tungsten ions in low-to-intermediate charge states.