Search

A recent extensive experimental study of impurity production and penetration for various density regimes is described. Deuterium and carbon emissions near a neutraliser plate (NP) of the Tore Supra Ergodic Divertor (ED) has been measured with an absolutely calibrated visible endoscope, for high- and low-density plasma regimes. From these radiation measurements, we have deduced an effective carbon flux, and at an order-of-magnitude estimate of the NP erosion: 50 m for the fall 1999 experimental ED campaign (≈500 shots). Combining the measured carbon and deuterium fluxes, we deduce a global experimental carbon sputtering yield for the NP in the range 2×10 −2 –3×10 −1 for ohmic pulses, showing evidence of the importance of carbon self-sputtering.

Highly radiative edge layer experiments in Tore Supra ergodic divertor (ED) plasmas are reported. In these experiments, up to 88% of the injected power is radiated in steady state attached plasma conditions. The observed radiated power is higher than the prediction of Matthews' multi-machine scaling. Such behaviour is consistent with the fact that the ratio of the ergodic divertor volume to the main plasma volume is higher for an ergodic divertor than for axisymmetric divertors, or with the enhanced impurity screening observed in Tore Supra. Indeed, a new scaling is derived for impurity radiation in a region of parallel ion transport. It shows that impurity screening is the dominant mechanism controlling divertor radiation, impurity transport along the open field lines having only little influence on P rad .

Sophisticated X-ray spectroscopic instruments have been developed and used for confinement studies in tokamaks at JET and Culham Laboratory. The Bragg rotor survey spectrometer, when programmed to operate in monochromatic or limited spectral range (LSR) modes, has been used to derive transport coefficients from the transient spectral signature of injected, non-recycling, test ions. The coefficients appropriate to specific confinement regimes have been investigated with respect to ion charge state and tokamak machine parameters. The LSR mode allows discrimination against background X-:ray light, a particularly serious problem at high electron density, ne ≈ 1020 m−3, in H mode where there is evidence, in COMPASS-D, for modest accumulation of impurities. Impurity ion accumulation is observed most clearly in passive spectroscopic studies of the 1s2 − 1s2p line emission from intrinsic impurity ions such as CI(XVI) using a high resolution Johann configuration crystal spectrometer. Line shapes and line ratios within the limited spectral range 4.45-4.50 A are interpreted in terms of ion and electron temperature and ion fluid velocity. In particular, the ‘q’/‘w’ line intensity ratios are interpreted in terms of the relative abundance of the CI(XV)/CI(XVI) ions which are modelled in terms of ion transport. These continuous, passive measurements of ion confinement are compared with results from the transient injection method. Absolute levels of the X-ray continuum, adjacent to the Cl(XVI) lines, are used to derive Zeff. Both the Bragg and Johann X-ray spectrometers have been calibrated for absolute photon sensitivity. The efficiencies of the diffractors in these instruments can be measured using a newly developed, two-axis goniometer.

This paper presents a spectroscopic study of deuterium and impurity radiation in the MkI divertor of JET in different experimental regimes. Radiated power measurements using absolutely calibrated VUV spectroscopy are consistent with bolometer measurements in a variety of impurity scenarios. These observations, combined with measurements of visible emission, are compared with code predictions of divertor radiation distributions, highlighting uncertainties in the description of the background plasma and of the carbon source, which affect detailed modelling of divertor impurity transport.

The JET neutral beam injection system has proved to be both effective and reliable as a plasma heating device. The ion heating and plasma fuelling characteristics of the system are ideally suited to the production of high fusion performance plasmas while the flexibility in the choice of beam species (H, D, T, 3 He or 4 He) and the ability to inject into almost any JET plasma configuration allows a wide variety of related physics experiments to be carried out. The capability to inject (for the first time) tritium beams was essential to the successful execution of the first tritium experiments in which 1.7 MW of power from DT fusion reactions was generated.

Beryllium films (150 to 300 A) have been evaporated on all JET in-vessel components (carbon limiters, nickel screen, inconel walls). Immediate and lasting improvements of the plasma have been observed in all plasma regimes. We concentrate on the results obtained with ICRH and NBI on the density limit and on H-mode plasmas which are particularly sensitive to impurity release. Specific effects produced by RF waves are reduced to a negligible level when the following 3 aspects are combined - beryllium evaporation, dipole phasing, correct screen angle. The density limit obtained with ICRF is now identical to NBI heated cases and substantially higher than the values obtained in the pre-beryllium phase. Long H-modes (1.3 s) have been obtained with RF alone. The H mode confinement is similar to that of NBI cases. In limiter plasmas Zeff decreased to 1.2–1.4 at high density and for 10 MW input power. The results are related to oxygen gettering, suppression of nickel sputtering from the antenna screens and to the use of dipole phasing.

A high-throughput high-resolution curved crystal spectrometer for measurements in the soft X-ray range on fusion magnetic devices is described. The spectrometer uses a large useful area (8 × 5 cm 2 ) quartz lamina bent at a curvature radius 385 cm long as the dispersing element. The detector is a one-dimensional space resolving multiwire proportional counter. The spectral resolving power is R = 18 000, permitting to obtain adequate ion temperature measurements, down to 700 eV, from the Doppler broadening of resonance lines of highly ionized metallic impurities.

Uncertainties in deducing central ion temperatures in the JET tokamak from the line width of the helium-like nickel resonance line emitted by impurity nickel ions in the plasma centre are examined. The statistical errors in fitting a Voigt function to the experimental line profile data points produce estimated ion-temperature errors of typically

We suggest measuring the energy deposition profile in additionally heated plasmas by modulating, at an appropriate frequency, a fraction of the supplementary power. If the frequency is high enough the temperature oscillations occur, with appreciable amplitude, only where the energy is directly deposited. Operating at lower frequency, plasma transport properties can be studied, through the measure of the phase delay of the temperature oscillations.

A new regime of particle recycling has been observed in high density discharges ( n e ≤ 4.5 x 14 cm -3 ) in the Frascati torus (FT). This new regime is characterized by a sudden enhancement of D a emission as measured at different toroidal locations. This enhancement correlates with other observations such as a drop in the impurity line emission, narrowing of electron temperature profiles and strong cooling of the plasma edge. All these observations point to the possible formation of a cold gas-plasma blanket as foreseen theoretically when high values of na (the product of average density and plasma radius) are achieved, like on FT.

We have measured the spectrum of plasma emission in the Frascati tokamak FT around the resonance line 1s2p → 1s 2 of the helium-like ion Cr XXIII. The presence of dielectronic satellites has been detected and their wavelengths determined. The intensity ratio of these satellites to the resonance line has been calculated from the corresponding lines in Fe XXV with a simple extrapolation. Good agreement with experimental points has been obtained in high electron density discharges. Line ratios systematically lower than the predicted values are obtained at low density and this has been related to the presence of a non-thermal population of electrons as indicated also by other diagnostics.

H-like argon line spectra have been recorded by means of a high-resolution crystal spectrometer, following puff injection during a tokamak discharge, in a small range of temperature and density plasma conditions (Te ≅ 1.2 keV, ne ≅ 1014 cm−3). Accurate wavelengths determination shows good agreement with theoretical predictions. Second order diffraction of He-like iron, occurring at the same Bragg angle, permits an interesting wavelength comparison. Argon line intensities have been computed, in order to determine both the Lyα 2 Lyα 1 and the dielectronic satellite to resonance line ratios, the former showing a small increase over the usual 0.5 value towards the end of the discharge and the latter allowing the calculation of a plasma electron temperature in fairly good agreement with Thomson scattering measurements. Finally, the plasma ion temperature is deduced by Doppler broadening of the resonance lines.