Search

In this work the onset of tearing modes in the termination phase of plasma pulses on JET is investigated. It is shown that the broadening or the shrinking of the current density profile, as a consequence of a core hollowing or an edge cooling of the electron temperature profile, strongly increases the probability of destabilizing a 2/1 tearing mode also in absence of an external trigger (e.g. a sawtooth crash). Two parameters are defined to highlight changes in the shape of the temperature profile that can lead to MHD instabilities and an empirical stability diagram is introduced into the space of the two new parameters. A large data-set of pulses carried out in the high-current scenario at JET with ITER-like wall is analyzed and criteria for the development of disruption alerts based on the two risk indicators for MHD instabilities are discussed, taking into account the different dynamics of the observed phenomena leading to the onset of 2/1 tearing modes.

The ITER scenarios and the project of DEMO involve stable operation above the Greenwald density, which justifies efforts to understand and overcome the density limit, this last observed as a disruptive termination of tokamak discharges and a thermal crash (with no disruption) of stellarator and reversed-field pinch (RFP) ones. Both in the tokamak and the RFP, new findings show that the high density limit is not governed by a unique, theoretically well-determined physical phenomenon, but by a combination of complex mechanisms involving two-fluid effects, electrostatic plasma response to magnetic islands and plasma-wall interaction. In this paper we will show new evidence challenging the traditional picture of the 'Greenwald limit', in particular with reference to the role of thermal instabilities and the edge radial electric field Er in the development of this limit. © 2015 EURATOM.

Neon injection in FTU can cause a spontaneous increase of the line-average density by a factor 2. The recent experiments were devoted to characterize the plasma response to the neon injection at different densities and plasma currents. A qualitative estimate from UV spectroscopy measurements indicates that the density behaviour cannot be attributed simply to the stripped electrons from the puffed impurity, but a modification of particle transport should be invoked in order to explain the spontaneous rise and the higher peaking. JETTO transport and GENE gyrokinetic codes analyses, as well as a calculation of the electron diffusion coefficients D and pinch velocity U, contribute to feature the peaking effect. © 2015 EURATOM.

The Frascati Tokamak Upgrade (FTU) is a high field and high density device which allows one to study confinement and transport at plasma parameters generally not accessible to other tokamaks and is relevant to next generation experiments (ITER). In this paper we study the effect of different density, temperature and q profiles on confinement and transport in various types of plasmas including ohmic, externally heated and pellet-fuelled discharges. Peaking of the electron density profile in pellet-fuelled discharges has been found to enhance the energy confinement time on FTU to values as high as 120 ms at densities of the order of those expected in ITER. Experiments designed to further improve confinement by injecting pellets at higher densities show the existence of a second threshold for saturation of confinement when the local electron heat conductivity corresponds to the same order of the local ion neoclassical conductivity in the region of maximum temperature gradient. Reflectometry measurements show turbulence suppression in pellet-injected discharges (as predicted by microstability analysis) and give insights to the change in turbulence in electron internal transport barrier plasmas. Finally, particle transport has been studied in experiments with full LH current drive.

Perturbative and steady-state heat transport of FTU tokamak in current ramp-up discharges are investigated by means of modulated electron cyclotron heating (MECH). Perturbative and steady-state transport experiments are coherent with an electron heat transport which switches from low to high values when electron temperature gradient length reaches a threshold value 1/L_Tc. The threshold value 1/L_Tc is shown to be proportional to the ratio s/q. The experimental findings are compared to predictions of an empirical model based on the assumption of a threshold gradient length, L_Tc (1/L_T = |grad Te/Te|), in the electron temperature Te below which electron thermal diffusivity, chi_e, switches from low to high values. Plasma responses to steady state and MECH are modelled assuming the electron diffusivity as chi_PB = chi_0 + alpha Te^(3/2) (1/L_T - 1/L_Tc)^(1/2); here Te^(3/2) reflects the gyro-Bohm assumption, chi_0 represents the heat transport for 1/L_T < 1/LT_c and the term (1/L_T - 1/L_Tc)^(1/2), which sets in for 1/L_T > 1/L_Tc , mimics an extra transport possibly due to electron temperature gradient (ETG) modes. In agreement with ETG threshold 1/L_Tc is shown to be correlated with the magnetic shear s.

Experiments on real time control of magneto-hydrodynamic (MHD) instabilities using injection of electron cyclotron waves (ECW) are being performed with a control system based on only three real time key items: an equilibrium estimator based on a statistical regression, a MHD instability marker (SVDH) using a three-dimensional array of pick-up coils and a fast ECW launcher able to poloidally steer the EC absorption volume with dρ/dt = 0.1/30 ms maximum radial speed. The MHD instability, usually a tearing mode with poloidal mode number m and toroidal mode number n such that m/n = 2/1 or 3/2 is deliberately induced either by neon gas injection or by a density ramp hitting the density limit. No diagnostics providing the radial localization of the instabilities have been used. The sensitivity of the used MHD marker allows to close the control loop solely on the effect of the actuator's action with little elaboration. The nature of the instability triggering mechanism in these plasma prevents that the stabilization lasts longer than the ECW pulse. However when the ECW power is switched on, the instability amplitude shows a marked sensitivity to the position of the absorption volume with an increase or decrease of its growth rate. Moreover the suppression of the dominant mode by ECRH performed at high plasma density even at relatively low power level facilitates the development of a secondary mode. This minimized set of control tools aim to explore some of the difficulties which can be expected in a fusion reactor where reduced diagnostic capabilities and reduced actuator flexibility can be expected. © 2015 EURATOM.

Modification of the sawtooth period through ion cyclotron resonance frequency (ICRF) heating and current drive has been demonstrated in a number of experiments. The effect has been seen to depend critically on the location of the ICRF absorption region with respect to the q = 1 flux surface. Consequently, for ICRF to be a viable tool for sawtooth control, one must be able to control the ICRF absorption location in real time so as to follow variations in the location of the q = 1 surface. To achieve this, the JET ICRF system has been modified to allow the JET real time central controller to control the frequency of the ICRF generators. An algorithm for real time determination of the sawtooth period has been developed and a closed loop controller, which modifies the frequency of the ICRF generators to bring the measured sawtooth period to the desired reference value, has been implemented. This paper shows the first experimental demonstration of closed loop sawtooth period control by real time variation of the ICRF wave frequency.

The assumption that the tokamak magnetic configuration at equilibrium is associated with the stationary magnetic entropy and with minimum plasma thermal energy leads to the prediction of definite profiles for the induced current density and for the pressure. These profiles should characterize the states spontaneously assumed by the tokamak under normal conditions (to be defined in the text).The theoretical results are compared with the observations collected from different tokamaks, plasma regimes and heating methods, taking into account the possible existence of non-inductive currents. The comparison allows the identification of the limits of validity of the model and confirms the robustness, within these limits, of the magnetic entropy concept for the description of the ‘privileged’ tokamak states.