Inboard/outboard asymmetry of poloidal flow observed in the Compact Helical System

Abstract

The poloidal flow velocity and the density of fully ionized carbon were measured in the Compact Helical System ͓K. Ida et al., Phys. Rev. Lett. 67, 58 ͑1991͔͒ using the charge exchange spectroscopy with the bidirectional viewing. The poloidal asymmetry of the poloidal flow velocity observed in the outward shifted plasma is due to the electrostatic potential being constant on the magnetic flux surfaces. The poloidal asymmetry of the ion density can be explained by the conservation of the poloidal ion flux under the strong toroidal viscosity. © 2000 American Institute of Physics. ͓S1070-664X͑00͒04002-7͔The spectroscopic measurements of plasma rotations are widely used to evaluate radial electric fields in magnetically confined toroidal plasmas. The rotation velocity is connected to the radial electric field and the ion pressure gradient by the radial force balance. In general, radial electric fields, ion pressure gradients, and rotation velocities are not constant on magnetic flux surfaces, since electrostatic potentials and ion pressures are considered to be constant on the magnetic flux surfaces. When the plasmas have toroidal curvatures ͑toroi-dal effects͒, the conservation of the poloidal plasma flux causes the difference of the poloidal rotation velocities between the inboard region of the magnetic axis (RϽR ax , where R ax is the major radius of the magnetic axis͒ and the outboard region (RϾR ax ). The standard neoclassical theory 1,2 predicts an inboard/outboard asymmetric parallel flow ͑a ''Pfirsch-Schlueter-like'' toroidal flow͒ to compensate for the unbalance of the perpendicular fluxes on the inboard and the outboard regions and to make the flow incompressible ("•vϭ0) in a steady state. In heliotron/ torsatron, 3-7 however, the strong toroidal viscosity due to helical ripples 3 damps this toroidal flow, as well as the net toroidal flows due to the radial electric field, the ion pressure gradient, and the toroidal momentum injection by neutral beams. This damping is strong in the peripheral region where the helical ripple is large. The poloidal rotation of impurity ions is also large in this peripheral region, since it is mainly driven by the radial electric field determined by the ambipolar condition of ripple diffusion fluxes.4-6 Therefore, the flow of impurity ions may not be incompressible as assumed in the neoclassical theory.The difference of poloidal rotation velocities and densities between the inboard and the outboard regions has not been studied in detail in either tokamaks or helical systems. In tokamaks, the poloidal rotation measurements are restricted in the outboard regions [8][9][10][11] because of the difficulty of installing chords viewing vertically the inboard regions where there are only very narrow spaces between toroidal and poloidal coils. In order to study the inboard/outboard asymmetry of poloidal flow, very precise measurements of Doppler shifts of impurity lines are required. The inboard/ outboard asymmetry has never been reported for lack of the accuracy of the absolute Doppl...

A Simultaneous Spectroscopic Measurement of the Global and Edge Local Structures in the Ion Temperature and Plasma Rotation Profiles in the Compact Helical System

“…Recently, from various edge plasma measurements in ETB plasmas in the CHS, it is suggested that the edge pedestal region, where the ion pressure gradients and a corresponding ion flow are expected, is very narrow (for e.g., ∆r/a ≈ 0.05) if it exists [6]. Radial profiles of the flow velocities and the ion temperatures in such a narrow region could not be detected with the previous CXS geometry at a vertically elongated section [11] with the chord spacing ∆R = 7.5 mm in a major radial direction corresponding to a spatial resolution of ∆r/a ≈ 0.06. Even though a vertical view at horizontally elongated sections subtends a limited radial range, it has superior spatial resolution near the edge region.…”

“…Especially in studies of radial electric fields, note that we do not measure directly the dominant ion component of the plasmas. We can know not only the flow velocity and the temperature of the impurity ion, but also the impurity density profile from the intensity of the charge exchange excited lines [11]. In the study of the transition phenomenon in the ETB plasmas, however, we must consider the density profile of the proton and thus the impurity density profile is out of scope of this paper.…”

“…The location of the vertically viewing chords at the vertically elongated cross section is illustrated in Refs. [11] and [14]. At this location, the beam radius (half-width at half-maximum∼6 cm in the vertical direction) can be neglected compared with the plasma size in the vertical direction (minor radius of ∼30 cm).…”

“…Therefore beam radius is negligible also at this horizontally elongated section measurement. The Doppler shifts at all of these chords are calculated as deviations from the "standard" wavelength (without Doppler shifts) obtained by the bidirectional viewing of the beam [11]. Therefore, total number of fiber channels that one Czerny-Turner type spectrometer (with a focal length of 1 m and a 2400 l/mm grating) has to detect simultaneously is 90.…”

“…[1], which assumed applications to large tokamaks with ion temperatures of T i ≥ 2 keV and toroidal rotations of V t ≈ 100 km/s. The best approach to measure correctly the Doppler shifts by canceling the mechanical wavelength offset is with bidirectional viewing (simultaneous views along opposite viewing directions) [10,11,13]. Therefore, in the multi-channel CXS systems to measure the plasma rotation profiles, the required number of observing channels includes not only that for observing the beam from one direction, but also that for the other direction to detect the absolute value of the Doppler shifts, as well as channels for measuring the background radiation [2].…”