The new molecular assisted recombination [MAR] of hydrogen plasmas supported by ammonia is investigated and presented throughout this work. The plasma detachment scheme has been considered to be a practical approach to drastically reduce the harsh heat flux to prevent severe damage onto the divertor plates. In the detached divertor plasma operation, the plasma from the fusion core loses its energy by the radiation loss in the upstream divertor region, and is recombined before it touches the divertor plate in the downstream region, with help of impurity gas seeding. Nitrogen is the strong candidate of cooling gas species, and forms compounds with hydrogen i.e. ammonia. This hydronitrogen-enhanced MAR [HN-MAR] process would occur throughout two steps: 1) charge or ion exchange reactions between hydrogen ions and ammonia molecules, 2) recombination reactions between NH3+ or NH4+ ions and electrons. In this work, the HN-MAR process including those 2 steps is studied combining experiments and rate equation model. The PISCES-E plasma device is used to make plasmas (ne ~ 10^16-10^18 m-3, Te ~ 3-4 eV) with D2, N2, ND3, and Ar gas feeding. As the 1st step of HN-MAR, the formation processes of ND3+ and ND4+ are investigated in low density D-N plasmas (ne ~ 10^16 m-3). The model shows that dominant neutralization channel of D+, D2+, and D3+ in the volume is the creation process of ND3+ and ND4+ throughout the charge or ion exchange reactions with ND3. The 2nd step of HN-MAR is studied using high density D-N-Ar plasmas (ne 10^17=10^18 m-3), in which the recombination process of ND3+ and ND4+ cannot be negligible comparing to the wall loss. The result shows that, as the plasma density is increased to these larger values, the ND4+ density fraction undergoes a drastic decrease in good agreement with a 0-d chemical kinetics model. These results provide evidence for the dissociative recombination reactions in the HN-MAR process. Finally, it is strongly suggested that the HN-MAR process is major chemical reactions in our D-N plasmas.

We report the observation of a net inward, up-gradient turbulent particle flux which occurs when a collisional drift waves generate a sufficiently strong radially sheared azimuthal zonal flow in a cylindrical magnetized plasma. At low magnetic fields (B≤1.0 kG), particle transport is outward at all radii. As the magnetic field is further increased to 1200G, an up-gradient inward particle flux develops between the peak of the velocity shear and the maximum density gradient. The mean density gradient is also observed to steepen in response to this inward flux. Time-domain and bispectral Fourier domain analysis shows that at the peak of the velocity shear, where the particle flux is outward, the turbulent Reynolds stress acts to reinforce the shear flow. In contrast, in the region of the inward particle flux, the zonal flow drives the fluctuations, and a transient increase in the shearing rate is occurs prior to an increase in the magnitude of the inward flux. The results suggest a hypothesis in which the shear flow is responsible for the up-gradient particle flux and the corresponding steepening in the mean density gradient. However, a linear instability analyses using experimentally measured density and E×B flow profiles in a linear, modified Hasegawa-Wakatani theory model with the coupled potential and density fluctuations failed to reproduce the essential elements of our experimental observations, suggesting some other mechanism is responsible for the inward flux. We summarize recent new experimental results which point towards the possible role of finite ion temperature gradient effects, possibly combined with parallel flow shear, in driving up-gradient particle flux

The turbulence-driven shear flows in both transverse and parallel direction have been investigated in a tokamak (HL-2A) and a linear device (CSDX). The first part of this study is devoted to the evolution of zonal flows and their turbulent drive near Greenwald limit in HL-2A tokamak. As the normalized line-averaged density $ \bar{n}_e/n_G $ is raised, the shearing rate of the poloidal $ {\bf E \times B} $ and turbulent phase velocities drop. Meanwhile, the turbulent drive for the low-frequency zonal flow (the Reynolds power) collapses at higher $ \bar{n}_e/n_G $ values. Correspondingly, turbulent particle transport increases in high collisionality plasmas. Besides damping the edge shear flows through increased collisional dissipation, the increased collisionality can also be connected to an increased non-adiabatic electron response at high densities, which raises the cross-correlation between density and potential perturbations. The second part is focused on the generation of intrinsic parallel flow in a linear magnetized device--CSDX. The axial flow shearing rate $V_{z}^{\prime}$ and the axial Reynolds power $\mathcal{P}_{z}^{Re}$ track the rising density gradient as the B field is raised. The axial Reynolds stress is the dominant source driving the axial flow as the density gradient grows during the magnetic field scan. The joint probability density function of radial and axial velocity fluctuations, $\mathsf{P}\left(\tilde{v}_{r},\tilde{v}_{z}\right)$, is highly tilted and anisotropic at higher magnetic field, indicating enhanced imbalance of the spectral correlator $\langle k_{\theta}k_{z} \rangle $ which can give rise to nonzero residual stress according to the dynamical symmetry breaking model. Furthermore, the axial velocity shear $V_{z}^{\prime}$ and its turbulent drive $\mathcal{P}_{z}^{Re}$ rise with increasing azimuthal velocity shear $V_{\theta}^{\prime}$ when $B<800$ G; after 800 G, $V_{\theta}^{\prime}$ and its turbulent drive collapse. The direct energy exchange between axial and azimuthal flows is low, since the $ \mathcal{P}_{z}^{Re} $ is about 10 times smaller than $ \mathcal{P}_{\theta}^{Re} $.

The Controlled Shear Decorrelation Experiment (CSDX) linear plasma device provides a unique platform for investigating the underlying physics of self-regulating drift-wave turbulence/zonal flow dynamics. A 3D drift-reduced nonlocal cold ion fluid model which evolves density, vorticity, and electron temperature fluctuations, with proper sheath boundary conditions, is used to simulate dynamics of the turbulence in CSDX. The simulations show density gradient-driven drift-waves are the dominant instability in CSDX. However, the choice of insulating or conducting end plate boundary conditions affects the linear growth rates and energy balance of the system due to the absence or addition of Kelvin-Helmholtz modes generated by the sheath-driven equilibrium E×B shear and sheath-driven temperature gradient instability.

The physics study is then followed by an extensive quantitative validation study. For the direct comparison of nonlinear simulation results to experiment, a synthetic Langmuir probe diagnostic is used to generate a set of synthetic observables which are in turn used to construct the validation metrics. A significant improvement of model-experiment agreement relative to the previous 2D simulations is observed. An essential component of this improved agreement is found to be the effect of electron temperature fluctuations on floating potential measurements, which introduces clear amplitude and phase shifts relative to the plasma potential fluctuations in synthetically measured quantities. Moreover, systematic simulation scans show that the self-generated E×B zonal flows profile is very sensitive to the steepening of density equilibrium profile. To minimize the number of simulations required for uncertainty quantification, a more advanced seed sampling methods for efficient sampling of the input parameter space, and a rapidly converging non-intrusive Probabilistic Collocation method to map out a probabilistic model response is implemented. This approach is shown to yield significantly better constrained uncertainty estimates than conventional uniform sampling methods for practical numbers of nonlinear simulations.

Understanding the erosion and redeposition behavior of tokamak plasma facing materials is important to ensure component lifetime. The goal of this thesis was to characterize the erosion and redeposition of aluminum (Al) when exposed to a tokamak divertor plasma. Al was chosen for its similarities to beryllium (Be), which is intended for use as the first wall material in ITER but is toxic and restricted in DIII-D and most other tokamaks. The Divertor Material Evaluation Station (DiMES) was used to expose a set of Al-coated samples to low-density L-mode plasma discharges in the DIII D divertor. Different plasma conditions were used for each sample (including He and D plasmas), and samples with both ideal (smooth) and practical (rough) surfaces were exposed. Measurements of quantitative emission spectroscopy and film thickness change were compared to simulations of sputtering, ionization, and redeposition of Al to determine the gross erosion rate, redeposition fraction, and spectroscopic emission efficiency.

We present the first quantitative spectroscopic measurements of neutral emission anisotropy due to sputtering erosion in a tokamak divertor plasma. We present an ionization-emission model that reproduced the anisotropy by assuming full angular sputtering yield distributions predicted by grazing angle sputtering simulations and ion beam sputtering measurements. Grazing angle ions were expected due to the disappearance of the classical Debye sheath in favor of a thicker magnetic pre-sheath (MPS) at small magnetic field surface inclination angles. The direction of presumed sputtering anisotropy and E×B drift of ions within the MPS was consistent with the direction of deposition patterns found on the samples and within individual pores of the rough surfaces. A model of the erosion-redeposition cycle including re-erosion and material mixing reproduced observed film thickness change measurements. The characteristic Al redeposition length was on the order of its ionization length, 1-3mm at the 1×10^13cm^-3 plasma densities analyzed. Total redeposition fractions ranged from 30% to 76%, increasing with higher plasma temperature due to the higher sheath electric field strength. On rough surfaces on the order of 50% of redepositing material was retained in hidden or shadowed regions, while on smooth surfaces this effect was negligible.

Tritium accumulation in nuclear fusion reactor materials is a major concern for practical and safe fusion energy. This work examines hydrogen isotope exchange as a tritium removal technique, analyzes the effects of neutron damage using high energy copper ion beams, and introduces a diffusion coefficient that is a function of the concentration of trapped atoms.

Tungsten samples were irradiated with high energy (0.5 - 5 MeV) copper ions for controlled levels of damage - 10^{-3} to 10^{-1} displacements per atom (dpa) - at room temperature. Samples were then exposed to deuterium plasma at constant temperature (~ 380 K) to a high fluence of 10^{24} ions/m^2, where retention is at is maximized (i.e. saturated). By then subsequently exposing these samples to fractions of this fluence with hydrogen plasma, isotope exchange rates were observed. The resulting deuterium still trapped in the tungsten is then measured post mortem. Nuclear reaction analysis (NRA) gives the depth resolved deuterium retention profile with the 3He(D,p)4He reaction, and thermal desorption spectroscopy (TDS) gives the total amount of deuterium trapped in the tungsten by heating a sample in vacuum up to 1200 K and measuring the evaporated gas molecules with a residual gas analyzer.

Isotope exchange data show that hydrogen atoms can displace trapped deuterium atoms efficiently only up to the first few microns, but does not affect the atoms trapped at greater depths. In ion damaged tungsten, measurements showed a significant increase in retention in the damage region proportional to dpa^{0.66}, which results in a significant spike in total retention, and isotope exchange in damaged samples is still ineffective at depths greater than a few microns. Thus, isotope exchange is not an affective tritium removal technique; however, these experiments have shown that trapping in material defects greatly affects diffusion.

These experiments lead to a simplified diffusion model with defect densities as the only free parameter. After examining the rate limiting processes, it is observed that trapped and solute atoms reach equilibrium concentrations before atoms diffuse further. This ultimately leads to the derivation of a diffusion coefficient that has a non-linear dependence on the concentration of trapped atoms, and this new coefficient can resolve discrepancies of diffusivity measurements in the literature.

We report a DIII-D database study of the H-mode power threshold over a wide range of plasma conditions and in the presence of resonant magnetic perturbations (RMPs). This database consisting of global (i.e. averaged) quantities is first compared to the 2008 ITPA database and the resulting L-H power scaling [38]. Since ELM control is critical for ITER, and applied 3D fields will likely be present prior to the transition into the H-mode, this study is important for assessing the impact of RMP ELM suppression on the L-H power threshold. The L-H transition has been studied extensively and is dependent on the physical and magnetic divertor geometry, shear flows, and drifts, among other parameters, some of which are altered by RMP fields. In order to understand the effects of RMPs on the L-H threshold, we attempt to make a robust empirical model, using only DIII-D data, that includes magnitudes and the toroidal modes of various resonant and non-resonant 3D fields. In addition, we assess the validity of previous assumptions about fast ion losses, as well as the usefulness of 0-D database regressions for extrapolation to ITER. Results from this database study show the standard 0-D parameters to be insufficient for capturing the complex L-H transition physics at a level high enough to provide an extrapolation to ITER with reasonable certainty.

Tokamaks can excite kink modes that can lock or nearly lock to the vacuum vessel wall, and whose rotation frequencies and growth rates vary in time but are generally inversely proportional to the magnetic flux diffusion time of the vacuum vessel wall. This magnetohydrodynamic (MHD) instability is pressure limiting in tokamaks and is called the Resistive Wall Mode (RWM). Future tokamaks that are expected to operate as fusion reactors will be required to maximize plasma pressure in order to maximize fusion performance. The DIII-D tokamak is equipped with electromagnetic control coils, both inside and outside of its vacuum vessel, which create magnetic fields that are small by comparison to the machine’s equilibrium field but are able to dynamically counteract the RWM. Presently for RWM feedback, DIII-D uses its interior control coils using a classical proportional gain only controller to achieve high plasma pressure. Future advanced tokamak designs will not likely have the luxury of interior control coils and a proportional gain algorithm is not expected to be effective with external control coils. The computer code VALEN was designed to calculate the performance of an MHD feedback control system in an arbitrary geometry. VALEN models the perturbed magnetic field from a single MHD instability and its interaction with surrounding conducting structures using a finite element approach. A linear quadratic gaussian (LQG) control, or ℋ₂ optimal control, algorithm based on the VALEN model for RWM feedback was developed for use with DIII-D’s external control coil set. The algorithm is implemented on a platform that combines a graphics processing unit (GPU) for real-time control computation with low latency digital input/output control hardware and operates in parallel with the DIII-D Plasma Control System (PCS). Simulations and experiments showed that modern control techniques performed better, using 77% less current, than classical techniques when using coils external to the vacuum vessel for RWM feedback. RWM feedback based on VALEN outperformed a classical control algorithm using external coils to suppress the normalized plasma response to a rotating n=1 perturbation applied by internal coils over a range of frequencies. This study describes the design, development and testing of the GPU based control hardware and algorithm along with its performance during experiment and simulation.

For magnetic confinement fusion reactors, physical sputtering of the plasma-facing materials (PFMs) can influence the lifetime of PFMs and the migration of impurities. During plasma exposure, the surface morphology of a PFM can be modified, resulting in the formation of complex submicron-scale structures. This dissertation demonstrates how such modifications can influence physical sputtering of PFM, the subsequent transport and migration of the physically sputtered impurities. In Chapter 3, the microscopic cone structures that can spontaneously form on a Cr surface physically sputtered by He plasma are studied in the PISCES-A linear plasma device with hyperspectral imaging measurements and in simulations couple SDTrimSP and GITR codes. Due to line-of-sight redeposition of sputtered Cr atoms onto cone structures, the effective sputtering yield is reduced, and the angular distributions at smaller angles with respect to the macroscopic surface normal is enhanced. In Chapter 4, the influence of fuzz structures on physical sputtering of Mo by He plasma is studied using the similar methods. While sputtering yield reduction due to the line-of-sight redeposition in the fuzz layer is confirmed, fuzz structures do not influence the angular distributions of sputtered Mo significantly. According to the depth profile of the differential sputtering yield from the simulations, a simple analytical model that characterizes the fuzz layer by a single collision mean free path is proposed, which successfully explains the effect of fuzz on physical sputtering. In Chapter 5, how angular distributions of physically sputtered W influence migration of W in the ITER divertor region is conceptually studied using ERO 2.0 simulations. When using simple idealized unidirectional distributions, the line-of-sight redeposition locations of W atoms clearly vary with the direction of the initial velocity. However, despite the differences between the angular distributions of W sputtered on a flat surface and on a fuzzy surface, the corresponding W deposition profiles look similar, since the suppression of neutral W atom population by ionization dominates the profiles, enhancing atomic deposition at locations closer to the strike points. From the results of this dissertation, the influence of surface morphology should be considered when studying physically sputtered impurities.

In this dissertation, we present advances in turbulence modeling for magnetically confined plasmas. We investigate the ecology of microscopic drift wave turbulence and the self-generated macroscopic flows in magnetically confined plasmas. We formulate reduced models that self-consistently describe the evolution of turbulence and mean plasma profiles (including flows) and recover trends obtained from the CSDX device and HL-2A tokamak. The dissertation is divided to three parts. The first part presents a reduced model that describes the interplay between drift wave turbulence and zonal and axial flows in the adiabatic plasma of CSDX, where the electron response is Boltzmann. The model explains how free energy released from the density gradient accelerates both axial and azimuthal flows in CSDX. A description of the interactions between the disparate scales of the plasma via the parallel and perpendicular Reynolds stresses $\langle \tilde v_x \tilde v_z \rangle $ and $\langle \tilde v_x \tilde v_y \rangle $ is presented. Expressions for these stresses are decomposed into a diffusive component that relaxes the flow profile, and a residual stress responsible for accelerating the corresponding flow. Moreover, parallel and perpendicular flow dynamics are described using an extended mixing length approach. This accounts for the degree of symmetry breaking in the parallel direction and parametrizes the efficiency of $\nabla n$ in driving the axial flow. In the second part of the dissertation, the relationship between drift waves and zonal flows is examined in depth via a more specific model. Analytical results obtained from this model confirm the published experimental data showing a suppression of turbulence with the increase in magnitude of the magnetic field \textbf{B}. A new criterion for access to enhanced confinement is introduced. This criterion captured by the dimensionless quantity $R_{DT}$, compares the production rate of turbulent enstrophy due to relaxation of the mean profiles, to the corresponding destruction rate via coupling to the mean flow. When $R_{DT} >1$, the profiles steepen and enhanced confinement is accessible. In the third paper, a novel idea for understanding the physics of the density limit problem in low $\beta$ tokamaks is presented. The collapse of the zonal shear flow when the electron response transitions from Boltzmann to hydrodynamic scaling, along with cooling of the edge and the onset of MHD activity is predicted by the observation that the zonal flow drive will drop as the electron parallel diffusion time increases with density. This leads to a simple, verified understanding of the density limit phenomenon in $L$-modes.