The standard one-dimensional formulation of impurity transport has proved to be highly questionable for impurities with medium/high Z, and is unsuitable for tungsten. This allows to reveal an inconsistent interpretation of the coupling between particle transport and ionization/ recombination processes. A discrete two-dimensional (2D) Fokker-Planck-Kolmogorov equation with charge and radius variables is proposed to study impurity equilibrium and transport in tokamak and stellarator plasmas. The impurity behaviour is consistently considered in terms of a 2D Markovian stochastic process, which has a complex structure according to probabilistic laws that include particle transport as part of a more general impurity charge state transport. It is found that a self-consistent 2D Markovian impurity equilibrium has the consequence that particle transport corresponding to the stationary state is governed by ionization and recombination processes, which are statistically independent of the particle motion. These generalised 2D coronal equilibrium conditions, usually observed for heavy impurities in experiments, turn out to be the stationary solution of the discrete 2D equation introduced and are justified in a discrete 2D grid model of impurity equilibrium. A complete analysis of the simplest discrete 2D equilibrium cell provides the basis for a general solution, while the grid model is reduced to a set of such local equilibrium cells using a pseudo-state technique.Modelling of the density profiles of carbon, argon and tungsten impurities is carried out by TICS (transport of impurity charge states) code developed using available experimental data. It is found that the introduced equilibrium function, which is a discrete profile of the ratios of reduced ionization and recombination rates, systematically predicts the central accumulation, equilibrium and radial transport of impurities in H-and L-mode and ion transport barrier plasmas, in agreement with the experimental data.

First experimental results of tungsten transport investigation in OH and ECRH plasmas in the T-10 tokamak with W-limiter and movable Li-limiter are presented. It is shown that tungsten tends to accumulate (a joint process of cumulation and peaking) near the plasma axis in ohmic regimes. The cumulation of W is enhanced in discharges with high values of the parameter that coincides with accumulation conditions of light and medium impurities in T-10 plasmas. Experiments with Li-limiter show the immeasurable level of Li3+ (0.3–0.5% of ne) of T-10 CXRS diagnostics because of the low inflow of Li with respect to other light impurities. Nevertheless, the strong influence of lithium on inflow of light and tungsten impurities is observed. In discharges with lithized walls, vanishing of light impurities occurs and values of are obtained. It is also shown that the tungsten density in the plasma center decreases by 15 to 20 times while the W inflow reduces only by 2 to 4 times. In lithized discharges with high γ, the flattening of the tungsten density profile occurs and its central concentration decreases up to 10 times during the on-axis ECRH. This effect is observed together with the increase of the W inflow by 3 to 4 times at the ECRH stage.

No abstract

The problems related to the spectral line-shape formation in the scrape of layer (SOL) in fusion reactor plasma for typical observation chords are considered. The SOL plasma is characterized by the relatively low electron density (10 12-10 13 cm −3) and high temperature (from 10 eV up to 1 keV). The main effects responsible for the line-shape formation in the SOL are Doppler and Zeeman effects. The main problem is a correct modeling of the neutral atom velocity distribution function (VDF). The VDF is determined by a number of atomic processes, namely: molecular dissociation, ionization and charge exchange of neutral atoms on plasma ions, electron excitation accompanied by the charge exchange from atomic excited states, and atom reflection from the wall. All the processes take place step by step during atom motion from the wall to the plasma core. In practice, the largest contribution to the neutral atom radiation emission comes from a thin layer near the wall with typical size 10-20 cm, which is small as compared with the minor radius of modern devices including international test experimental reactor ITER (radius 2 m). The important problem is a strongly non-uniform distribution of plasma parameters (electron and ion densities and temperatures). The distributions vary for different observation chords and ITER operation regimes. In the present report, most attention is paid to the problem of the VDF calculations. The most correct method for solving the problem is an application of

No abstract