Session UO4: ITER I

ITER Current Channel Control Under Disturbances and Disruptions with Implications from DIII-D Experiments

Control of the ITER plasma shape and position is challenging due to demanding performance requirements and limitations on superconducting poloidal field (PF) coil capabilities. For example, robust vertical stability in the presence of disturbances such as H-L transitions or ELM's, and following disruption thermal quenches, requires in-vessel Cu coils to augment the PF coils [1]. We analyze the most recent ITER in-vessel coil design for control robustness to expected disturbances and noise, and for control of post-disruption runaway electron (RE) current channels. We report on DIII-D experiments assessing position control for rampdown plasmas and RE channels, and control of RE current amplitude following disruptions using positive and negative applied loop voltage.\par \vskip8pt \noindent [1] Humphreys, D.A., et al., Nucl.\ Fusion {\bf 49} (2009) 115003.   

ASDEX Upgrade contribution to disruption studies for ITER

Plasma disruptions represent a hazard for the structural integrity of ITER. The contribution of the existing tokamaks to this project consists in refining the characterization of the disruption loads and their extrapolation, on the basis of physical models, and in learning to predict and mitigate disruptions. The ASDEX Upgrade research program covers these specific topics and this contribution reports on significant progress made in these areas. (I) The formation and evolution of the halo region is analyzed with MHD-transport codes and extrapolation to ITER is discussed. (II) Discriminant analysis is applied to each disruption group in order to determine the most significant plasma variables, which allow for classification, and is being used to discern between stable and pre-disruptive plasma states. (III) Progress has been made with MGI in attaining an effective electron density equivalent to 24 {\%} of the critical one (necessary for the collisional suppression of runaway electrons) and in studying the redistribution of the injected gas in the plasma.   

Wall forces produced during ITER disruptions

Nonlinear simulations with the M3D code [1] are performed of disruptions produced by a vertical displacement event (VDE) and a kink mode. The toroidally symmetric and asymmetric wall forces produced during a disruption are calculated in ITER. Expressions are derived for the wall force, including the sideways force, using a thin conducting wall model. The dependence of wall force on the kink growth rate and the wall penetration time is obtained. The largest force occurs when the growth rate equals the inverse wall penetration time. A theory is developed of the wall force produced by kink modes. The theory is in qualitative agreement with the simulations and JET experiments. The variation of the horizontal force with wall resistivity offers an important opportunity to ameliorate the sideways force of disruptions. If the wall can be made more conducting, it is possible to reduce the wall force by a large factor. The JET experiment operates in a regime with large halo currents [2] relative to eddy currents, which may overestimate the forces on a better conducting wall.\\[4pt] [1] W. Park {\it et al.}, Phys. Plasmas \textbf{6}, 1796 (1999). \\[0pt] [2] V. Riccardo {\it et al.}, Plasma Phys. Control. Fusion \textbf{46}, 925 (2004)   

Extrapolating the kinetic effects of energetic particles on resistive MHD stability to ITER

The effects energetic particles have on MHD instabilities is a key issue in the physics of burning plasma experiments such as ITER. Recent results indicate kinetic effects of energetic particles can play a crucial role in the stability of the m/n=2/1 tearing mode, especially in ITER where $\beta _{frac}=\beta _{h}$/$\beta $ is high ($\beta _{h}$ is energetic particle $\beta )$. Using realistic equilibria based on experimental reconstructions, the non-ideal MHD stability of the n=1 and 2 modes is calculated at a series of q$_{min}$, $\beta $, $\beta _{frac}$, and S=$\tau _{R}$/$\tau _{A}$, including the $\delta $f kinetic-MHD model in the 3-D extended MHD code NIMROD. Eigenvalue based computations using PEST-III and DCON give context to these results, and provide a basis for extrapolation. It is observed that for high q$_{min}$ the particles have significant stabilizing effects, while at low q$_{min} \ga $1 the interaction of the particles with the non-resonant response on axis causes destabilization of resistive modes. The requirements for directly computing energetic particle effects on resistive MHD modes in the burning plasma parameter regime are discussed.   

Microtearing instability in ITER*

Microtearing modes are found to be unstable in some regions of a simulated ITER H-mode plasma [1] with the GS2 code [2]. Modes with k$\rho _{s}>$1 are in the interior (r/a$\sim $0.65-0.85) while longer wavelength modes are in the pedestal region. This instability may keep the pedestal within the peeling-ballooning stability boundary [3]. Microtearing modes can produce stochastic magnetic field similar to RMP coils; they may have similar effects on ELMs by increasing the pedestal width. The possibility of using this technique for ELM mitigation in ITER is explored. We propose to use a deuterium gas jet to control the microtearing instability and the Chirikov parameter at the edge. Preliminary evaluation of its effectiveness will be presented and the limitations of the GS2 code will be discussed based on our understanding from NSTX [4]. *This work is supported by USDoE contract DE-AC02-09CH11466. \\[4pt] [1] R. V. Budny, Nucl. Fusion (2009)\\[0pt] [2] W. Dorland et al., Phys. Rev. Lett. (2000).\\[0pt] [3] P. B. Snyder et al.,Nucl. Fusion (2009).\\[0pt] [4] K. L. Wong et al., Phys. Rev. Lett. (2007).   

Equilibrium and Braking of Fully Avalanched Runaway Electron Currents: a New Disruption Mitigation Strategy for ITER

A first-of-kind solution of the Grad-Shafranov equation is described for post-thermal quench plasma discharges where the current is dominated by a fully avalanched runaway electron (RE) current plateau. The RE current profile tends to be much more centrally peaked relative to the initial plasma current and yet stable to internal kinks, in qualitative agreement with experiments. The current profile displays an interesting ``resilience,'' depending only weakly on the spatial distribution of the initial RE ``seed'' population. The RE current equilibrium has a slow secular time dependence that is traced to the presence of collisional dissipation on cold plasma electrons and the associated inductive energy decay. It is predicted that massive RE currents in ITER could be completely braked in $\sim 200$ ms by a combination of modest, 20$\times$ electron density increase, and reversed surface loop voltage, with $\rm{E_{sur}}\sim -1$ V/m. The model is being used to interpret recent DIII-D experiments where this erosion effect was recently observed.   

Generation and Stability of Runaway Electrons During Rapid-Shutdown in DIII-D

We present results of runaway electron (RE) experiments with argon killer pellet induced rapid shutdowns. RE generation observed in x-ray emission before the current quench (CQ) presents a paradox where generation occurs before any loop voltage from the CQ. To explore this paradox, we conducted new JFIT analysis of magnetic data during rapid-shutdowns revealing an inductance drop during the thermal quench (TQ) before the CQ begins. In this loop voltage analysis, the often-neglected inductance drop term exceeds the current drop term and occurs earlier. RE generation via the Dreicer effect using this term is large, where prior neglect of this term predicted small RE generation. Many of these RE escape before avalanche in diverted shape, but less so in a limited shape. The RE plateau later terminates with a drop in edge q and toroidally asymmetric hard x-ray emission, both signatures of a kink instability, suggesting ideal MHD-like stability criteria for runaways.   

Effect of plasma elongation on disruption runaway electrons

Studies of runaway electron (RE) populations during disruptions on a number of different tokamaks have shown two distinctly different types of behavior: (a) some machines tend to observe RE's during a significant number of current quenches, and (b) some machines rarely observe RE's during the disruption current quench. Of those that do see runaways, a general trait is that they run circular or low elongation and/or limited discharges (FTU, Tore-Supra, TEXTOR, JT-60U), and conversely, those that don't see runaways tend to run elongated, diverted discharges (C-Mod, DIII-D, ASDEX-U). This suggests that elongation might play a role in RE confinement during disruptions, and recent experiments on DIII-D support this hypothesis. An experiment to test this on Alcator C-Mod uses lower hybrid current drive to generate a strong RE population, and gas jet injection to trigger reproducible disruptions. Behavior of runaways during disruptions in both low elongation and high elongation equilibria are compared. Experimental findings will be presented and compared to NIMROD modeling predictions. Implications for using this technique to enable an extensive research thrust on RE physics and mitigation will be discussed, as will implications for ITER.   

L-H Transition Studies on DIII-D to Determine \hbox{H-mode} Access for Operational Scenarios in ITER

A comprehensive set of L-H transition experiments has recently been performed on DIII-D to determine the requirements for access to H-mode plasmas in ITER's first (non-nuclear) operational phase with H and He plasmas, and second (activated) operational phase with D plasmas. The results from these experiments have revealed that the H-mode power threshold, $P_{\rm TH}$: (a) increases with the applied torque for all 3 main ion species (H, He, D); (b) increases with the application of the I-coil current, which is normally used to induce $n=3$ resonant magnetic perturbations required for edge localized mode (ELM) suppression; (c) can be significantly reduced by changing the plasma geometry; (d) exhibits a weak dependence on the edge electron and ion temperatures, but shows a strong dependence on the edge toroidal rotation; and (e) is not significantly affected by the application of magnetic error fields expected from test blanket module assemblies in ITER.   

L-H Threshold Studies in NSTX

L-H transition experiments in NSTX have been run in support of the high priority ITER and ITPA issue of access to the H-mode. Experiments revealed that the L-H threshold power for helium was 20 to 40{\%} greater than that for deuterium. There was a $\sim $35{\%} reduction in the threshold power for discharges using lithium evaporation. Application of n=3 fields at the plasma edge, potentially critical for suppression of ELMs in ITER, resulted in a 65{\%} increase in threshold power with no change in plasma rotation. Threshold powers were almost a factor of two greater at 1 MA than at 0.7 kA, consistent with calculations from XGC0 showing a deeper E$_{r}$ well and stronger E$_{r}$ shear near the edge at lower current. Low triangularity discharges required lower heating powers to transition into the H-mode, also consistent with XGC0. No systematic differences in T$_{e}$, n$_{e}$, p$_{e}$, T$_{i}$, v$_{\phi}$ or their derivatives between purely L-mode and pre-transition H-mode plasmas were found.   

First real-time detection on surface dust detection in tokamaks

Dust generated from plasma surface interactions has important consequences for the operation and safety of next-step devices and local measurements of dust are part of the ITER dust strategy. The first real-time measurements of surface dust in the NSTX vessel have been successfully made using an novel electrostatic surface dust detector. Impinging dust particles create a temporary short circuit on fine grid of interlocking circuit traces that is biased to 50 v and the resulting current pulse is recorded by counting electronics. Techniques used to increase the sensitivity to match NSTX dust levels while suppressing electrical pickup will be presented. The detector has been calibrated with both carbon and lithium particles. In a separate experiment a probe with ITER scale castellation gaps was filled with dust particles and exposed to an intentional disruption in NSTX. Results on the mobilization of dust from the castellations will be reported.