Robert Andre

Comparison of poloidal velocity measurements to neoclassical theory on the National Spherical Torus Experimenta)

Knowledge of poloidal velocity is necessary for the determination of the radial electric field, which along with its gradient is linked to turbulence suppression and transport barrier formation. Recent measurements of poloidal flow on conventional tokamaks have been reported to be an order of magnitude larger than expected from neoclassical theory. In contrast, poloidal velocity measurements on the NSTX spherical torus [Kaye et al., Phys. Plasmas 8, 1977 (2001)] are near or below neoclassical estimates. A novel charge exchange recombination spectroscopy diagnostic is used, which features active and passive sets of up/down symmetric views to produce line-integrated poloidal velocity measurements that do not need atomic physics corrections. Inversions are used to extract local profiles from line-integrated active and background measurements. Poloidal velocity measurements are compared with neoclassical values computed with the codes NCLASS [Houlberg et al., Phys. Plasmas 4, 3230 (1997)] and GTC-NEO [Wang et...

Core transport of lithium and carbon in ELM-free discharges with lithium wall conditioning in NSTX

Core transport of intrinsic carbon and lithium impurities is analysed in H-mode discharges in NSTX. The application of lithium coatings on graphite plasma-facing components led to high-performance H-mode discharges with edge localized mode (ELM) suppression and resulted in core carbon accumulation. Lithium ions did not accumulate and had densities less than 1% of carbon densities. Core transport codes NCLASS, NEO and MIST are used to assess the impact of lithium evaporative coatings on impurity transport. The disappearance of ELMs, due to changes in the electron pressure profiles, together with modifications in neoclassical transport, due to changes in main ion temperature and density profiles, explains the core carbon accumulation in discharges with lithium coatings. Residual anomalous transport in the pedestal region is needed to explain the experimental carbon density profile shape and evolution. The enhancement in neoclassical lithium particle diffusivities due to the high carbon concentration is partially responsible for the low lithium core concentration.

MHD-induced energetic ion loss during H-mode discharges in the National Spherical Torus Experiment

Magnetohydrodynamic (MHD) induced energetic ion loss in neutral beam heated H-mode discharges in the National Spherical Torus Experiment (NSTX) is discussed. After H-mode onset, the neutral particle analyser (NPA) spectrum usually exhibits a significant loss of energetic ions mainly for E > Eb/2 where Eb is the beam injection energy, although this loss occasionally extends to lower energy. The magnitude of the energetic ion loss diminishes with increasing tangency radius of the NPA sightline, increasing the toroidal field and the neutral beam injection energy. Modelling suggests that MHD-induced ion loss is enhanced during H-mode operation due to an evolution of the q and beam deposition profiles that feeds both passing and trapped ions into the region of the plasma affected by the low-n MHD activity. It must be emphasized that this loss mechanism is a pressure profile effect that can sometimes occur in unusually high density L-mode discharges, but almost always is observed in H-mode discharges because of the intrinsic broad electron density profile. Analysis of the particle interaction with a model magnetic perturbation supports the energy selectivity of the observed MHD-induced loss. Transport analysis using a fast-ion diffusion model to emulate MHD-induced energetic ion loss shows significant modifications of the heating produced by the neutral beam which changes the inferred power balance, thermal diffusivities and the thermal energy confinement time. In the case cited herein, MHD-induced energetic ion loss increased the thermal energy confinement, τE, by ~15% and reduced the toroidal beta, βT, by ~7%. An accounting of energetic ion loss is therefore important for proper analysis of power balance and transport in plasmas exhibiting MHD-induced energetic ion loss.

Microturbulence and flow shear in high-performance JET ITB plasma

The transport, flow shear, and linear growth rates of microturbulence are studied for a JET plasma with high central q in which an internal transport barrier (ITB) forms and grows to a large radius. The linear microturbulence growth rates of the fastest growing (most unstable) toroidal modes with high toroidal mode number are calculated using the GS2 and FULL gyrokinetic codes. These linear growth rates, γlin, are large, but the flow-shearing rates, γE×B (dominated by the toroidal rotation contribution), are also comparably large when and where the ITB exists.

Experimental and modeling uncertainties in the validation of lower hybrid current drive

This work discusses sources of uncertainty in the validation of lower hybrid wave current drive simulations against experiments, by evolving self-consistently the magnetic equilibrium and the heating and current drive profiles, calculated with a combined toroidal ray tracing code and 3D Fokker–Planck solver. The simulations indicate a complex interplay of elements, where uncertainties in the input plasma parameters, in the models and in the transport solver combine and—in some cases—compensate each other. It is concluded that ray-tracing calculations should include a realistic representation of the density and temperature in the region between the confined plasma and the wall, which is especially important in regimes where the LH waves are weakly damped and undergo multiple reflections from the plasma boundary. Uncertainties introduced in the processing of diagnostic data as well as uncertainties introduced by model approximations are assessed. It is shown that, by comparing the evolution of the plasma parameters in self-consistent simulations with available data, inconsistencies can be identified and limitations in the models or in the experimental data assessed.

Simulations towards the achievement of non-inductive current ramp-up and sustainment in the National Spherical Torus Experiment Upgrade

One of the goals of the National Spherical Torus Experiment Upgrade (NSTX-U) (Menard et al 2012 Nucl. Fusion 52 083015) is the demonstration of fully non-inductive start-up, current ramp-up and sustainment. This work discusses predictive simulations where the available heating and current drive systems are combined to maximize the non-inductive current and minimize the solenoidal contribution. Radio-frequency waves at harmonics higher than the ion cyclotron resonance (high-harmonic fast waves (HHFW)) and neutral beam injection are used to ramp the plasma current non-inductively starting from an initial Ohmic plasma. An interesting synergy is observed in the simulations between the HHFW and electron cyclotron (EC) wave heating. Furthermore, time-dependent simulations indicate that, depending on the phasing of the HHFW antenna, EC wave heating can significantly increase the effectiveness of the radio-frequency power, by heating the electrons and increasing the current drive efficiency, thus relaxing the requirements on the level of HHFW power that needs to be absorbed in the core plasma to drive the same amount of fast-wave current.

Central safety factor and β N control on NSTX-U via beam power and plasma boundary shape modification, using TRANSP for closed loop simulations

The high-performance operational goals of NSTX-U will require development of advanced feedback control algorithms, including control of βN and the safety factor profile. In this work, a novel approach to simultaneously controlling βN and the value of the safety factor on the magnetic axis, q0, through manipulation of the plasma boundary shape and total beam power, is proposed. Simulations of the proposed scheme show promising results and motivate future experimental implementation and eventual integration into a more complex current profile control scheme planned to include actuation of individual beam powers, density, and loop voltage. As part of this work, a flexible framework for closed loop simulations within the high-fidelity code TRANSP was developed. The framework, used here to identify control-design-oriented models and to tune and test the proposed controller, exploits many of the predictive capabilities of TRANSP and provides a means for performing control calculations based on user-supplied data (controller matrices, target waveforms, etc). The flexible framework should enable high-fidelity testing of a variety of control algorithms, thereby reducing the amount of expensive experimental time needed to implement new control algorithms on NSTX-U and other devices.