Open Access Publications from the University of California

Studies of turbulence and flows in the DIII-D tokamak

Abstract

Understanding the turbulent transport of particles, momentum, and heat continues to be an important goal for magnetic confinement fusion energy research. The turbulence in tokamaks and other magnetic confinement devices is widely thought to arise due to linearly unstable gyroradius-scale modes. A long predicted characteristic of these linear instabilities is a critical gradient, where the modes are stable below a critical value related to the gradient providing free energy for the instability and unstable above it. In this dissertation, a critical gradient threshold for long wavelength ($k_{\theta} \rho_s \lesssim 0.4$) electron temperature fluctuations is reported, where the temperature fluctuations do not change, within uncertainties, below a threshold value in $L_{T_e}^{-1}=|\nabla T_e| / T_e$ and steadily increase above it. This principal result, the direct observation of a critical gradient for electron temperature fluctuations, is also the first observation of critical gradient behavior for \textit{any} locally measured turbulent quantity in the core of a high temperature plasma in a systematic experiment. The critical gradient was found to be $L_{T_e}^{-1}|_{crit}=2.8 \pm 0.4 \ \mathrm{m}^{-1}$. The experimental value for the critical gradient quantitatively disagrees with analytical predictions for its value. In the experiment, the local value of $L_{T_e}^{-1}$ was systematically varied by changing the deposition location of electron cyclotron heating gyrotrons in the DIII-D tokamak. The temperature fluctuation measurements were acquired with a correlation electron cyclotron emission radiometer. The dimensionless parameter $\eta_e=L_{n_e}/L_{T_e}$ is found to describe both the temperature fluctuation threshold and a threshold observed in linear gyrofluid growth rate calculations over the measured wave numbers, where a rapid increase at $\eta_e \approx 2$ is observed in both. Doppler backscattering (DBS) measurements of intermediate-scale density fluctuations also show a frequency-localized increase on the electron diamagnetic side of the measured spectrum that increases with $L_{T_e}^{-1}$. Measurements of the crossphase angle between long wavelength electron density and temperature fluctuations, as well as measurements of long wavelength density fluctuation levels were also acquired. Multiple aspects of the fluctuation measurements and calculations are individually consistent with the attribution of the critical gradient to the $\nabla T_e$-driven trapped electron mode. The accumulated evidence strongly enforces this conclusion. The threshold value for the temperature fluctuation measurements was also within uncertainties of a critical gradient for the electron thermal diffusivity found through heat pulse analysis, above which the electron heat flux and electron temperature profile stiffness rapidly increased. Toroidal rotation was also systematically varied with neutral beam injection, which had little effect on the temperature fluctuation measurements. The crossphase measurements indicated the presence of different instabilities below the critical gradient depending on the neutral beam configuration, which is supported by linear gyrofluid calculations.

In a second set of results reported in this dissertation, the geodesic acoustic mode is investigated in detail. Geodesic acoustic modes (GAMs) and zonal flows are nonlinearly driven, axisymmetric ($m=0,\ n=0$ potential) $E \times B$ flows, which are thought to play an important role in establishing the saturated level of turbulence in tokamaks. Zonal flows are linearly stable, but are driven to finite amplitude through nonlinear interaction with the turbulence. They are then thought to either shear apart the turbulent eddies or act as a catalyst to transfer energy to damped modes. Results are presented showing the GAM's observed spatial scales, temporal scales, and nonlinear interaction characteristics, which may have implications for the assumptions underpinning turbulence models towards the tokamak edge ($r/a \gtrsim 0.75$). Measurements in the DIII-D tokamak have been made with multichannel Doppler backscattering systems at toroidal locations separated by $180^{\circ}$; analysis reveals that the GAM is highly coherent between the toroidally separated systems ($\gamma > 0.8$) and that measurements are consistent with the expected $m=0,\ n=0$ structure. Observations show that the GAM in L-mode plasmas with $\sim 2.5-4.5$ MW auxiliary heating occurs as a radially coherent eigenmode, rather than as a continuum of frequencies as occurs in lower temperature discharges; this is consistent with theoretical expectations when finite ion Larmor radius effects are included. The intermittency of the GAM has been quantified, revealing that its autocorrelation time is fairly short, ranging from about 4 to about 15 GAM periods in cases examined, a difference that is accompanied by a modification to the probability distribution function of the $E \times B$ velocity at the GAM frequency. Conditionally-averaged bispectral analysis shows the strength of the nonlinear interaction of the GAM with broadband turbulence can vary with the magnitude of the GAM. Data also indicates a wave number dependence to the GAM's interaction with turbulence. Measurements also showed the existence of additional low frequency zonal flows (LFZF) at a few kilohertz in the core of DIII-D plasmas. These LFZF also correlated toroidally. The amplitude of both the GAM and LFZF were observed to depend on toroidal rotation, with both types of flows barely detectable in counter-injected plasmas.

In a third set of results the development of diagnostic hardware, techniques used to acquire the above data, and related work is described. A novel multichannel Doppler backscattering system was developed. The five channel system operates in V-band (50-75 GHz) and has an array of 5 frequencies, separated by 350 MHz, which is tunable as a group. Laboratory tests of the hardware are presented. Doppler backscattering is a diagnostic technique for the radially localized measurement of intermediate-scale ($k_{\theta} \rho_s \sim 1$) density fluctuations and the laboratory frame propagation velocity of turbulent structures. Ray tracing, with experimental profiles and equilibria for inputs, is used to determine the scattering wave number and location. Full wave modeling, also with experimental inputs, is used for a synthetic Doppler backscattering diagnostic for nonlinear turbulence simulations. A number of non-ideal processes for DBS are also investigated; their impact on measurements in DIII-D are found, for the most part, to be small.