Session QO2: Magnetic Confinement Configurations, Turbulence

Optimized Generation of Helical Current in a NTM by Modulated ECCD in ASDEX Upgrade

The control of MHD instabilities is of great interest to expand operational space and hence the performance of present day and future tokamaks, such as ITER. Electron Cyclotron Current Drive (ECCD) with its highly flexible, localised deposition is an ideal tool for this purpose. Experiments with optimised deposition, maximising $\eta_NTM =j_ECCD/j_bs$ have been continued for NTM stabilization. They verify the relevance of this figure of merit, which has been adopted by ITER for the NTM stabilization and have increased the $\beta_N$ range in which NTMs can be stabilized. Another crucial point for ITER is the requirement of phased injection in the island's O-point. Experiments on (3,2) NTM stabilization with an broadened deposition mimic the situation in ITER (marginal island size $<$ the deposition width in the final stabilization). The experiments show an advantage of the modulated ECCD and are accompanied by modelling using a fully nonlinear resistive MHD code. An analysis of NTM stabilization based on the modified Rutherford equation has been carried out to obtain fit coefficients for predictions for ITER. We will discuss the next steps undertaken on ASDEX Upgrade to verify feedback controlled MHD mode control. A central part of this is the on-line determination of both mode position and ECCD deposition and its use to poloidally steer the mirrors.   

Neoclassical Tearing Modes and Fast Ions Confinement in ASDEX Upgrade

We present new ASDEX Upgrade (AUG) results on fast ion losses (FIL) caused by NTMs. A new detector provides energy and pitch-angle resolved FIL measurements (1 MHz bandwidth). Lost particles, of NBI origin , are mostly passing, and are lost basically with their birth energy (100 keV). Their transit frequency (approx. 200 kHz) is higher than the mode frequency (typically less than 20 kHz). We observe a good coincidence between the frequency and phase of the mode and those of the losses and a strong correlation between the NTM amplitude and the amount of FIL. This is discussed as a result of orbit stochasticity: small magnetic perturbations, which do not cause magnetic field lines ergodicity, can nevertheless result in drift islands in the fast particle’s phase space. Simulations of this mechanism with the Hamiltonian guiding center code ORBIT give results consistent with experimental data.   

Experimental Study of the Effects of Lithium Coated Plasma Facing Components on Energy Confinement Time in the CDX-U Device

The first ever measurements of energy confinement time for spherical tokamak plasmas in the presence of lithium coated plasma facing components (PFC's) have been made in the CDX-U device. The energy confinement time, as derived from power balance considerations using parameters calculated from plasma equilibria, was as large as $6$ milliseconds for Ohmic plasmas in the presence of both solid and liquid lithium PFC's. This represents a significant improvement over baseline plasmas, which typically had energy confinement times of $1$ millisecond or less. The energy confinement for plasmas with lithium PFC's also showed an improvement over that expected from the ITER98(y,1) confinement scaling. The improvement in confinement over this scaling correlates with the observed increase in density ``pump-out'', which is indicative of low wall-recycling. Plasma equilibria were calculated using a modified version of the Equilibrium and Stability Code (ESC), and were constrained by measurements made from a collection of magnetic field diagnostics. The ESC was modified to incorporate the first ever implementation of a novel response function technique for in-situ magnetic field diagnostic calibration that is insensitive to toroidal asymmetries and vessel wall currents.   

Non-inductive production of tokamak-like ST plasmas with washer gun sources on the Pegasus Toroidal Experiment

Washer gun plasma sources in the lower divertor region of the Pegasus ST can act as DC helicity injection sources for purely non-inductive startup or as pre-ionization sources for ohmic induction at very low toroidal field. In both applications, use of these edge current sources results in considerably altered current profiles and suppression of the strong internal tearing modes that in the past constrained ohmic plasmas to I$_{P} \quad \le $ I$_{TF}$ and relatively low I$_{N}$. At low applied vertical field, the gun-driven I$_{\phi }$ is large enough to cause a poloidal flux reversal, and the plasma relaxes into a tokamak-like configuration. Discharges of I$_{\phi }$ $>$ 50 kA are produced by $\sim $ 4 kA of injected current, with $>$ 80{\%} of I$_{\phi }$ persisting after the guns shut off. These discharges exhibit features indicative of tokamak plasmas, including strong vacuum field deformation, increased current decay times ($>$ 0.7 ms), and core heating. They also have very high field utilization factor, I$_{P}$ / I$_{TF} \quad \le $ 2.3, I$_{N} \quad >$ 12, and very low 2/1 mode activity.   

Properties of the FRX-L Field Reversed Configuration (FRC) inferred from density profile measurements assuming MHD equilibrium

A laser interferometer probes the line integrated time history of plasma density along eight chords of the FRX-L high-density ($\sim 10^{17} $ cm$^{-3}$) FRC. The data is Abel and tomographically inverted to provide density profiles. The FRC is roughly in an axisymmetric rotational MHD equilibrium for a $3$ $\mu $s interval between the initial implosion that produces the FRC and an $n=2$ rotational instability that terminates confinement. $B_{z}(r,t)$ may then be inferred, given an external magnetic field measurement. The period during which its area integral\ approximates an independent axial flux measurement self-consistently identifies the equilibrium interval. Basic FRC properties such as temperature, poloidal flux, and $\alpha $ (rotational to ion diamagnetic drift frequency ratio) are then inferred. Results indicate that poloidal flux estimates based on magnetic and axial flux measurements alone are conservative, and that the critical $\alpha $ for $n=2$ instability (estimated two ways) is roughly between 0.5 and 1.0.   

Visible-wavelength emission by Hydrogen in the Princeton FRC

The Princeton Field-Reversed-Configuration (FRC) experiment is investigating prolate FRCs heated by odd-parity rotating magnetic fields. Using visible wavelength spectroscopy with both an iCCD-based 0.5m visible light spectrometer and a 0.5m monochromator, we have investigated the role of collisions with neutrals on plasma performance. Both atomic and molecular emission were recorded, with a time resolution of 0.05 ms. Analysis of the Fulcher alpha H$_{2}$ molecular band yields excited population rotational temperatures of 0.025-0.05eV, implying ground state rotational temperatures of 0.05-0.1eV. The molecular gas is largely dissociated in the first millisecond of the pulse.   

X-ray Measurements in the Princeton FRC

Two Si-PIN X-ray detectors with electronics for pulse height analysis have been installed on the RMF$_o$-heated Princeton FRC (PFRC) to measure X-ray emission from about 800 eV to several keV. The lower limit is partially set by 0.5 or 1 mil Be windows that maintain vacuum between the detectors and the plasma. The energy dispersion of the detector systems have been calibrated using known X-ray fluorescence lines of several elements excited by a radioactive Fe-55 source. Gating of the multichannel analyzer allows time-resolved spectra to be obtained, normally with a resolution of 0.5 ms during a typical 2-3 ms shot. Spectra obtained from the PFRC will be presented with a discussion of the characteristics of the electron energy distribution and a comparison of the predicted electron energy distribution based on RMF$_o$ heating.   

Formation of collisionless high-beta plasmas by application of odd-parity rotating magnetic fields

Odd-parity rotating magnetic fields (RMF$_o$) have been applied to mirror-configuration plasmas in a device of 4-cm radius, defined by discrete co-axial copper-ring flux conservers, and with divertor chambers beyond both mirrors. At an applied RF power of 10 kW and a bias field of 50 G, line-averaged electron density of 1e12 cm$^{-3}$ and excluded flux of 0.005 mVs have been sustained for over 0.5 ms, corresponding to a Coulomb collisionality of $<$ 0.02. The divertors allow reduction of the electron-neutral collision frequency to similarly low levels. Under such conditions, measurements of RMF$_o$ coupling indicate full penetration of the RMF$_o$ to the major axis.   

Mass Dependence in the Turbulent Critical Energy using BCS Theory

The existence of a lambda-like behavior in turbulent transport coefficients suggests that there may be a second order (continuous) phase transition as systems evolve from a non-turbulent to a turbulent state. There are quantitative implications from the use of the Ginzburg-Landau (GL) approach for this phase change. When the BCS theory is applied using the G-L equations to turbulence, the value of the critical turbulent energy, U$_{c, }$is derived directly from the force constraint (and intermolecular constants), the role of the electron can be replaced by the constituent atoms (or molecules) and the free energy per unit volume is proportional to m, the mass of the constituent atoms (or molecules). This gives a turbulent `isotope' effect from a derivation of the isotope effect in superconductivity.   

Mass Dependence on Turbulence in a Laser-Induced Plasma

Single pulse plasmas are created when a focused pulsed Nd-Yag laser is fired on various gas targets at various pressures inside of a sealed chamber. The optical emissions of selected atomic and singly ionized lines were recorded with the oscilloscope at a rate of 10 GS/s. Standard calculations of turbulent parameters such as the characteristic frequency, the spectral index, the chaotic dimension, and the turbulent fluctuation energy were performed for each spectrum. From these calculations, a critical turbulent energy is determined for each combination of pressure and composition of the target gases in the sealed chamber. The evolution of the turbulent parameters as well as evolution of the electron temperature is calculated for various pressures and compositions of the target gas in the chamber. The turbulent parameters of different species are compared in order to determine a mass dependence of these turbulent phenomena. These relationships are explored using a second-order phase transformation explanation of a transition to turbulence from a non-turbulent to a turbulent state.   

Survey of Accelerator-Region Density on CTIX using Laser Deflectometery

Recent experiments have demonstrated that accelerator-region gas puffing increases the plasma density of compact toroids (CTs) created by the Compact Toroid Injection Experiment (CTIX). CT velocity can be determined from time of arrival measurements of optical and magnetic signals. To determine total CT mass a plasma density profile must be assumed or measured using either heterodyne interferometers or, especially in higher density plasmas, by the newer and simpler technique of laser deflectometry. In high density plasma, axial deflectometry provides absolute plasma density without an assumption of a radial profile. CT kinetic energy can then be calculated from measurements of plasma density and velocity. We will use this new system to determine CT kinetic energy as a function of accelerator input energy.   

Observation of turbulent MHD flow around mechanical obstacles in the target chamber of CTIX

The Compact Toroid Injection eXperiment injects high velocity (200 km/s) magnetized plasma (B $\sim$ 0.5 to 1 Tesla) into a target chamber with large windows (0.5 m by 0.2 m) that allow transverse imaging of the plasma with fast digital cameras at sub-microsecond exposures. Helium is added to the hydrogen plasma in the acceleration region to enhance plasma brightness. It has been observed that after passing through a wire array the plasma develops turbulent flow patterns that are most clearly detected at the plasma edge. Fourier analysis of the images gives information about the k-spectrum of the turbulent fluctuations, while a consideration of the ratio of exposure time to plasma time of flight provides further constraints on a model of the plasma velocity field in this system. Studies of the MHD interaction between the high velocity plasma and targets such as in this system are intended to address fundamental issues in the physics of applications such as Magnetized Inertial Fusion (MIF) and CT fueling of tokamaks.   

Rotation induced L to H mode transition of a cylindrical plasma column

The outer region of the plasma column of the LAPD is rotated in a controlled fashion by biasing a section of the vacuum chamber wall positive with respect to the cathode. The plasma column of the LAPD device at UCLA is 17.5 m in length and 60 cm in diameter. A uniform, 400 Gauss axial magnetic field is used in these experiments. Cross-field ion current due to ion-neutral collisions provides the torque to spin up the plasma. In the non-rotating plasma column, cross-field particle transport is measured to proceed at the Bohm diffusion rate. Rotation, above a threshold voltage, suppresses cross-field transport from Bohm to classical rates, leading to steeper radial density gradients. Suppression of radial particle transport is global and not isolated to the region of flow shear.