Thesis (PH. D)--University of Pennsylvania. / Cover title.

No description available.

The physics of partially neutralized plasmas is largely unexplored, partly because of the difficulty of confining such plasmas. Plasmas are confined in a stellarator without the need for a plasma current, and regardless of the degree of neutralization. The Columbia Non-neutral Torus (CNT) is a stellarator dedicated to the study of non-neutral, and partially neutralized plasmas. This thesis describes the first systematic studies of plasmas of arbitrary neutrality. The degree of neutralization of the plasma can be parameterized through the quantity η ≡ |n_e - Z n_i|/|n_e + Z n_i|. In CNT, η can be varied continuously from pure electron (η = 1) to quasi-neutral (η ≈ 0) by adjusting the neutral pressure in the chamber, which controls the volumetric ionization rate. Pure electron plasmas are in macroscopically stable equilibria, and have strong self electric potentials dictated by the emitter filament bias voltage on the magnetic axis. As η decreases, the plasma potential decouples from the emitter, and spontaneous fluctuations begin to appear. Partially neutralized plasmas (10^-3 < η < 10^-1) generally exhibit multi-mode oscillations in CNT. However, when magnetized ions are present, the electron-rich plasma oscillates at a single dominant mode (20 - 100 kHz). As the plasma approaches quasi-neutrality (η < 10^-5), it also reverts to single mode behavior (1 - 20 kHz). A parametric characterization of the single mode fluctuations detected in plasmas of arbitrary neutrality is presented in this thesis along with measurements of the spatial structure of the oscillations. The single mode fluctuations observed for η ≈ 0.01 to 0.8 are identified as an ion resonant instability propagating close to the E × B velocity of the plasma. The experiments also show that these oscillations present a poloidal mode number m = 1, and a toroidal number n = 0, which is identical to the spatial structure of the diocotron instability in pure-toroidal traps, and implies that the ion-driven instability breaks parallel force balance and the conservation of poloidal flux in CNT. The low frequency oscillations detected in the quasi-neutral regime are a global instability convected by the E × B flow of the plasma. In this case, the mode aligns almost perfectly with the field lines, and presents a resonant m = 3 poloidal structure.

The Columbia Non-neutral Torus (CNT) is the first experiment designed to create and study small Debye length non-neutral plasmas confined by magnetic surfaces. This thesis describes experimental confinement studies of non-neutral plasmas on magnetic surfaces in CNT. Open orbits exist in CNT resulting in electron loss rates that are much faster than initially predicted. For this reason a conforming boundary was designed and installed to address what is believed to be the primary cause of open orbits: the existence of a sizable mismatch between the electrostatic potential surfaces and the magnetic surfaces. After installation a record confinement time of 337 ms was measured, more than an order of magnitude improvement over the previous 20 ms record. This improvement was a combination of the predicted improvement in orbit quality, a reduced Debye length that resulted in decreased transport due to the perturbing insulated rods, and improved operating parameters not indicative of any new physics. The perturbation caused by the insulated rods that hold emitters on axis in CNT is a source of electron transport and would provide a loss mechanism for positrons in future positron-electron plasma experiments. For these reasons an emitter capable of creating plasmas then being removed faster than the confinement time was built and installed. Measurements of plasma decay after emitter retraction indicate that ion accumulation reduces the length of time that plasmas are confined. Plasmas have been measured after retraction with decay times as long as 92 ms after the emitter has left the last closed flux surface. Experimental observations show that obstructing one side of an emitting filament with a nearby insulator substantially improves confinement. As a result, experiments have been performed to determine whether a two stream instability affects confinement in CNT. Results indicate that the improvement is not caused by reducing a two stream instability. Instead, the improvement is a result of altering the sheath of the emitting filament which allows the plasma to reach an equilibrium state with improved confinement. These measurements agree with confinement times for plasmas created by unobstructed emission that are in the same improved confinement state.

This thesis describes the first results of passive and active MHD spectroscopy experiments on the High Beta Tokamak-Extended Pulse (HBT-EP) device using a new array of magnetic diagnostics and coils. The capabilities of the HBT-EP experiment are significantly extended with the installation of a new adjustable conducting wall, high-power modular control coil arrays, and an extensive set of 216 magnetic sensors that allow simultaneous high-resolution detection of multimode MHD phenomena. The design, construction, and calibration of this system are described. The capability of this new magnetic diagnostic set is demonstrated by biorthogonal decomposition analysis of passive measurements of rotating resistive wall modes (RWMs). A strong multimode effect is detected for the first time in HBT-EP plasmas consisting of the simultaneous existence of m/n=3/1 and 6/2 RWMs which cause the plasma to evolve in a non-rigid multimode manner. Additional mode numbers as high as n=3 are also observed. Active MHD spectroscopy experiments using a "phase-flip" resonant magnetic perturbation (RMP) are able to excite a clear three-dimensional response. By adjusting the helicity of the magnetic field applied by the control coils, the driven plasma response is shown to be predominantly resonant field amplification. When the amplitude of the applied field is not too large, the driven resonant response appears linear, independent of the presence of background MHD phenomena and consistent with the predictions of single-helicity modeling of kink mode dynamics. The spatial structures of both the naturally rotating kink mode and the externally driven response are observed to be identical, while the temporal evolutions are approximately independent. The phase-flip driven plasma response is measured as a function of edge safety factor, plasma rotation, and the amplitude of the applied magnetic perturbation. As the RMP amplitude increases, the plasma response is shown to be linear, saturated, and ultimately, disruptive.

This thesis presents the first systematic study of multimode external kink structure and dynamics in a tokamak using a high-resolution magnetic sensor set. Multimode effects are directly measured, rather than inferred from anomalies in single-mode behavior. In order to accomplish this, an extensive set of 216 poloidal and radial magnetic field sensors has been installed in the High Beta Tokamak -- Extended Pulse (HBT-EP) device for high-resolution measurements of three-dimensional mode activity. An analysis technique known as biorthogonal decomposition (BD) is described, and simulations are presented to justify its use for studying kink mode dynamics in HBT-EP data. Coherent activity of multiple simultaneous modes is observed using the BD without needing to define a mode structure basis beforehand. Poloidal mode numbers up to m=8 are observed via sensor arrays with full 360 degree coverage. Higher poloidal mode numbers are suggested by the data, but cannot be well-resolved with the available diagnostics. Toroidal mode numbers up to n=4 are observed. Non-rigid, multimode activity is observed for coexisting external kinks having m/n=3/1 and 6/2 structures. Despite sharing the same helicity and same resonant surface, rotation of 6/2 modes is independent of 3/1 mode rotation -- the n=2 mode does not simply rotate with double the frequency of the n=1 mode. During periods of 3/1-dominated activity, the 6/2 mode is observed to modulate the 3/1 amplitude, and in brief instances can overpower the 3/1. Statistical analysis over many shots reveals the multimode nature of the 3/1 kink to be more significant when the resonant q=3 surface begins internal, then is ejected from the plasma. This inference is based on the relative amplitudes of secondary modes during 3/1-dominated activity, as well as spectral content of the modes. Conformal conducting wall segments were also retracted away from the plasma surface using low-order poloidal and toroidal asymmetries to excite measurable differences in low m- and n-number modes. Kink mode amplitudes increase as the wall segments are withdrawn, and non-symmetric wall configurations modulate the amplitude and frequency of the rotating modes depending upon their toroidal orientation with respect to the non-symmetric wall. Modulations of mode amplitude and rotation are larger for the toroidal wall asymmetry than for the poloidal wall asymmetry.

The application of electrostatic bias to both low density plasma with coherent fluctuations and high density plasma with turbulent fluctuations confined by a magnetic dipole are investigated. Previously, electrostatic biasing of low density plasma was symmetric, drove rapid plasma rotation, and excited the centrifugal interchange instability. This research investigates the application of non-symmetric bias and the influence of broken symmetry on strongly turbulent plasmas. Non- symmetric bias is applied through either point biasing or an equatorial array spanning the device. In both cases, the spatial symmetry of applied bias dramatically effects the plasma fluctuations. With bias applied, the plasma achieves a new equilibrium characterized by amplified low order modes and diminished amplitude of higher order modes. Although the turbulent spectrum changes, the RMS fluctuation level is unchanged by the bias. Bias also causes the turbulent electrostatic fluctuations to coalesce into a quasi-coherent mode and the appearance of increased coherence. The effect of bias configuration is also seen to change the measured levels of nonlinear coupling. Non-symmetric biasing increases nonlinear coupling while symmetric biasing leaves the coupling unchanged. These results represent the first experimental demonstration of symmetry breaking driving the inverse energy cascade in a quasi-two dimensional plasma system. The application of dynamic and rotating electrostatic bias as well as plans for applying turbulent feedback are discussed.

Understanding the maintenance and stability of plasma pressure confined by a strong magnetic field is a fundamental challenge in both laboratory and space plasma physics. Using magnetic and X-ray measurements on the Levitated Dipole Experiment (LDX), the equilibrium plasma pressure has been reconstructed, and variations of the plasma pressure for different plasma conditions have been examined. The relationship of these profiles to the magnetohydrodynamic (MHD) stability limit, and to the enhanced stability limit that results from a fraction of energetic trapped electrons, has been analyzed. In each case, the measured pressure profiles and the estimated fractional densities of energetic electrons were qualitatively consistent with expectations of plasma stability. LDX confines high temperature and high pressure plasma in the field of a superconducting dipole magnet. The strong dipole magnet can be either mechanically supported or magnetically levitated. When the dipole was mechanically supported, the plasma density profile was generally uniform while the plasma pressure was highly peaked. The uniform density was attributed to the thermal plasma being rapidly lost along the field to the mechanical supports. In contrast, the strongly peaked plasma pressure resulted from a fraction of energetic, mirror trapped electrons created by microwave heating at the electron cyclotron resonance (ECRH). These hot electrons are known to be gyrokinetically stabilized by the background plasma and can adopt pressure profiles steeper than the MHD limit. X-ray measurements indicated that this hot electron population could be described by an energy distribution in the range 50-100 keV. Combining information from the magnetic reconstruction of the pressure profile, multi-chord interferometer measurements of the electron density profile, and X-ray measurements of the hot electron energy distribution, the fraction of energetic electrons at the pressure peak was estimated to be about 35% of the total electron population. When the dipole was magnetically levitated the plasma density increased substantially because particle losses to the mechanical supports were eliminated so particles could only be lost via slower cross-field transport processes. The pressure profile was observed to be broader during levitated operation than it was during supported operation, and the pressure appeared to be contained in both a thermal population and an energetic electron population. X-ray spectra indicated that the X-rays came from a similar hot electron population during levitated and supported operation; however, the hot electron fraction was an order of magnitude smaller during levitated operation (<3% of the total electron population). Pressure gradients for both supported and levitated plasmas were compared to the MHD limit. Levitated plasmas had pressure profiles that were (i) steeper than, (ii) shallower than, or (iii) near the MHD limit dependent on plasma conditions. However, those profiles that exceeded the MHD limit were observed to have larger fractions of energetic electrons. When the dipole magnet was supported, high pressure plasmas always had profiles that exceeded the MHD interchange stability limit, but the high pressure in these plasmas appeared to arise entirely from a population of energetic trapped electrons.

This thesis presents the systematic study of the multimode external kink mode structure and dynamics in the High-Beta Tokamak Extended-Pulse experiment (HBT-EP) when the plasma rotation is externally controlled using a source of toroidal momentum input. The capabilities of the HBT-EP tokamak to study rotation physics was greatly extended during a 2009-2010 major upgrade, when a new adjustable conducting wall, a high-power modular control coil array system, and an extensive set of 216 poloidal and radial magnetic sensors were installed on the machine. HBT-EP was additionally equipped with a biased edge electrode which made it possible to adjust the plasma ion and plasma magnetohydrodynamics (MHD) mode rotation frequencies by imparting an electromagnetic torque on the plasma. The design of this biased edge electrode, and its capability to torque the plasma is described. The rotation frequency of the helical kink modes was directly inferred from analysis of the magnetics dataset. To directly measure the plasma ion acceleration as the plasma was torqued by the biased electrode, a novel high-throughput and fast-response spectroscopic rotation diagnostic was installed on HBT-EP. This spectroscopic rotation diagnostic was designed to measure the velocity of He ions, therefore when conducting experiments using the spectroscopic rotation diagnostic a gas mixture of 90%D and 10%He was used. With its current power supplies the bias probe is capable of accelerating the primary m/n=3/1 helical kink mode (which has a natural rotation frequency between +7-+9kHz) to somewhere between -50kHz-+25kHz depending on the probe bias. At a probe voltage of +175V the He impurity ions were seen to accelerate by 3km/sec. Biorthogonal decomposition (BD) analysis was applied to the large magnetics dataset and used to determine the multimode m/n spectrum of the helical kink modes present in HBT-EP. The dominant helicities present as revealed by the BD are the m/n=3/1 and m/n=6/2 modes, which represent about 85% and 8% of the total MHD activity respectively. This percentages remain consistent across the entire range of 3/1 mode rotation frequencies obtainable from the bias probe, (-50kHz-25kHz). The Hilbert transform technique was also applied to magnetic sensor data to determine the instantaneous amplitude and frequency of the total MHD activity. The total MHD amplitude was seen to decrease with increasing plasma rotation, a 35% reduction as the 3/1 mode was accelerated from +6-+24kHz. Active MHD spectroscopy experiments using a resonant magnetic perturbation (RMP) are able to excite a clear three-dimensional plasma response. Plasma rotation is theoretically expected to increase plasma stability to external resonant error elds, and in HBT-EP the plasma amplitude response to a m/n=3/1 RMP increases by a factor of 2.7 when the plasma rotation is decreased from +25kHz to +-2kHz. As the RMP amplitude increases, slower plasmas are seen to disrupt at a lower perturbation amplitude than unperturbed or rapidly rotating modes. The 6/2 helical kink mode also shows an amplitude and phase response to the 3/1 RMP, and like the 3/1 mode the amplitude response is largest when the plasma is slowly rotating. The ratio between the plasma 6/2 amplication and the 3/1 amplication to a 3/1 RMP is nearly constant, regardless of the plasma rotation or the RMP amplitude.

In this thesis, I present measurements from a high-speed video camera diagnostic on the High Beta Tokamak - Extended Pulse (HBT-EP). This work represents the first use of video data to analyze and understand the behavior of long wavelength kink perturbations in a wall-stabilized tokamak. A Phantom v7.3 camera was installed to capture the plasma's global behavior using visible light emissions and it operates at frame rates from 63 to 125 kfps. A USB2000 spectrometer was used to identify the dominant wavelength of light emitted in HBT-EP. At 656 nm, it is consistent with the D-alpha light expected from interactions between neutral deuterium and plasma electrons. The fast camera in combination with an Acktar vacuum black background produced images which were inverted using Abel techniques to determine the average radial emissivity profiles. These profiles were found to be hollow with a radial scale length of approximately 4 cm at the plasma edge. As a result, the behavior measured and analyzed using visible light videography is limited to the edge region. Using difference subtraction, biorthogonal decomposition and Fourier analysis, the structures of the observed edge fluctuations are computed. By comparing forward modelling results to measurements, the plasma is found to have an m/n = 3/1 helical shape that rotates in the electron drift direction with a lab-frame frequency between 5 and 10 kHz. The fast camera was also used to measure the plasma's response to applied helical magnetic perturbations which resonate with the equilibrium magnetic field at the plasma's edge. The static plasma response to non-rotating resonant magnetic perturbations (RMPs) is measured by comparing changes in the recorded image following a fast reversal, or phase flip, of the applied RMP. The programmed toroidal angle of the RMP is directly inferred from the resulting images of the plasma response. The plasma response and the intensityof the RMP are compared under different conditions. I found the resulting amplitude correlations to be consistent with previous measurements of the static response using an array of magnetic sensors. My work has shown that high-speed videography can be an extremely useful diagnostic for measuring edge perturbations in a tokamak. Future measurements, such as those using multiple cameras with different views, are expected to improve our understanding of plasma behavior and to detect edge fluctuations with higher temporal and spatial resolution. Supplementary Videos: Video 1 - This is an example of the video data from Shot 77324, an unforced plasma shot taken with the shells inserted. Video 2 - The strongest naturally-rotating mode has been extracted from a subset of the raw data shown in Video 1 using a biorthogonal decomposition. Long striations can be seen which are common in shots that have the shells inserted. Video 3 - In this video of the raw data from Shot 77537, the shells are retracted. The smooth non-reflective Acktar black background can be seen between the shells. Video 4 - The dominant BD mode from Shot 77537 shows pinwheel-like behavior. With the shells retracted, the plasma encounters fewer physical structures for neutral recycling and this affects the light emissions. Video 5 - This video shows the dominant BD modes from Shot 78029 during which a phase-flip RMP was used to influence the plasma. The mode seems to slow in its rotation as it resonates with the externally-applied field.