Radio frequency inductively coupled plasma sources are widely used in low temperature industrial processing. Recent computer simulations and experiments indicate significant improvement in processing results with the use of pulsed plasmas. This thesis reports simultaneous measurements of nearly all plasma and electromagnetic properties of an ICP in three-dimensions under pulsed operation.The plasma density profile is measured by rf compensated Langmuir probes, hairpin probes, and a 60 GHz microwave interferometer. Line integrated electron density measured from these three very distinct diagnostic techniques (interferometer, Langmuir probe, Hairpin resonator probe) shows agreement. Three-axis Bdot probes are used to measure the magnetic field. Inductive plasma current is calculated from measured magnetic field using Ampere’s law. With the observed cylindrical symmetry, the electromagnetic electric field is derived from Bdot data using Faraday's law. The above field and current are used to calculate the inductive power and Poynting vector. A side by side comparison in pulsed plasma operation is presented for two industry-relevant neutral pressures (5 mTorr, 25 mTorr). Distinct plasma evolution is observed with the two cases. The observations from the experiment are explained with the help of a simple lumped element circuit model.

Highly energetic electrons produced naturally or artificially can be trapped in the Earth's radiation belts for months, posing a severe danger to valuable space satellites. Concepts that can lead to radiation belts mitigation have drawn a great deal of interest. This dissertation investigates the loss of fast electrons from a magnetic mirror trap when irradiated by a shear Alfvén wave. The experiment is performed in the quiescent after-glow plasma in the Large Plasma Device (LaPD). The background magnetic field is programmed to include a magnetic mirror section (mirror ratio ≈ 2, length = 3.5 m). A trapped fast electron population is generated in the mirror section by X-mode high power microwave pulses. Different power levels, pulse durations and heating schemes lead to different electron energies. Two distinct electron energy ranges are studied separately in this work, which are around 100 eV and 100 keV respectively. Shear Alfvén waves of arbitrary polarization are launched externally by a Rotating Magnetic Field (RMF) source (δB/B0 ≈ 0.1%, λ_|| ≈ 9 m). It is demonstrated that the shear Alfvén wave can effectively de-trap energetic electrons from the magnetic mirror field in both electron energy ranges.

Due to grad-B and curvature drift, the highly energetic electrons (above 100 keV) form a hot electron ring in the magnetic mirror. Experimental evidence indicates that the ring is deformed in the right-handed (RH) circularly polarized shear Alfvén wave field. Electron losses are observed in both the radial and the axial direction of the mirror field, with modulation at the Alfvén wave frequency. The periodical loss continues even after the termination of the wave.

A test particle simulation is performed, which confirms that the single particle motion of the energetic electrons in the Alfvén wave field is not adequate to explain experimental observations, and the hot electron collective behavior must play a role in the de-trapping effect.

It is proposed that the hot electron ring is deformed due to E_Alfvén×B₀ drift in the Alfvén wave field. The ring deformation grows when the electron azimuthal drift matches the rotation of the RH shear Alfvén wave pattern, which generates a collective mode of the ring that degrades its confinement and leads to electron loss from the magnetic mirror.

Magnetized plasma expansions from explosive phenomena often have characteristically large ratios of kinetic ram pressure to the ambient magnetic field pressure, β≫1. In the presence of a tenuous, ambient plasma, collisions are incapable of transferring much energy due to the high relative velocities. These expansions, however, generate large magnetic field variations, ΔB/B∼1, in the form of a diamagnetic cavity as well as potentially large electric fields.

A high-β expansion was created using a laser-produced plasma that expanded, v_exp=1.28?〖10〗^7 cm/s, β∼〖10〗^6, into a uniform, magnetized background plasma. The processes that are capable of transferring energy and momentum from the high-β expansion to the ambient plasma without collisions were explored. The combination of a magnetic probe and a novel emissive probe yielded measurements of the total electromagnetic field in three-dimensions. These constituted the first measurements of the total electric field in such an environment. The electrostatic field structure comprised the predicted inward field of a diamagnetic cavity as well as previously unobserved features including an outward field associated with a magnetic field compression and an intense electrostatic pulse preceding the LPP. All these components were stronger than the largest observed induced electric field from Faraday’s law. Direct measurements of argon ion velocities moving through these fields were made with a planar, laser-induced fluorescence diagnostic which showed ions being pulled inward against the expansion direction. Orbit solvers show that the characteristic velocity observed, v_r=-3?〖10〗^5 cm/s, is consistent with the measured fields. The inward electrostatic field exhibited a linear variation with the magnetic field while the outward field and pulse exhibited at most a weak dependence. No significant differences in the fields were observed between helium and argon background plasmas. A qualitative model to describe the evolution of the high-β expansion in the context of weak coupling was developed. The model and the experimental field structure yielded important scaling relations for similar expansions and a qualitative extrapolation to the strong-coupling case.

Low temperature radio frequency (RF) inductively coupled plasma (ICP) is widely used in semiconductor industry for surface etching and processing. The plasma needs to be highly controllable and uniform. Electronegative gases are often used for their high reactivity, which adds to the complexity. Various measurements were performed in an effort to characterize a pulsed ICP in an industrial etch tool driven by a 2 MHz planar coil, with improved temporal and spatial resolution from previous literature. The operating gas was Ar/O2. The wafer on the bottom electrode can be biased at 1 MHz independently. The decoupling of the driving RF and the bias RF enables separate control over the plasma parameters in the sheath and the bulk region. In the sheath region in contact with the wafer, laser induced fluoresence experiment was conducted. Ar ion angle and energy distributions, drift velocity and sheath dynamics were investigated for different cases of wafer bias. The results showed an instantaneous sheath motion in response to the wafer bias. The ion energy distribution showed a continuous tail extending close to the bias voltage when the bias was turned on during the plasma glow, and a bimodal shape with a high energy peak when the bias was turned on in the afterglow. In the bulk region, RF compensated Langmuir probe, hairpin probe, Bdot probe and photodetachment were used to obtain essential plasma parameters. The electron and negative ions spatial distribution showed significantly different structures in Ar and Ar/O2. Different propagation patterns of the fields and inductive power deposition were also observed.

Ion cyclotron resonance heating (ICRH) systems are critical components of current and future tokamak experiments aimed at producing nuclear fusion energy. During ICRH a host of deleterious effects occur, including increased heat flux to plasma facing components and modification of launched wave power. A suspected root cause of these effects is the radio frequency (RF) rectification of the plasma potential. Interest in the antenna scrape-off layer (SOL) region has drawn increasing interest, as it is recognized that mitigating these effects is necessary to achieving fusion power. This dissertation investigates the RF rectification of the plasma potential and the resulting cross-field flows that form due to an active RF antenna. The experiment is performed in the Large Plasma Device (LAPD) utilizing a fast wave antenna and RF amplifier system developed for these studies. The RF system is capable of 150 kW output power for a 1 ms pulse that is repeated at the 1 Hz repetition rate of the LAPD plasma discharge.

Upon application of the RF pulse to the antenna, the DC plasma potential, measured with an emissive probe, dramatically increases in certain spatial locations by a factor greater than 10 Te. The largest plasma potentials are observed at locations magnetically connected to the top and bottom of the antenna, and they exist only in the private SOL created between the antenna and a limiter placed 3.6 m away along the LAPD axis. The DC rectified potentials scale linearly with the antenna current over a factor of 12x in the applied current. These DC potentials increase plasma materials interactions (PMI), resulting in the sputtering of antenna materials whose presence is detected in the bulk plasma by the coatings that develop on probe diagnostics. The DC rectified potentials persist in the plasma long after the RF current in the antenna has rung down on the same time scales as the change in the density.

At the top and bottom of the antenna are circular flows, often called convective cells. These 𝑬�𝑩𝟎 flows arise due to the spatial variation of the RF rectified potentials across the background magnetic field. The maximum strength of the electric field causing these flows was

found to scale quadratically with antenna current, giving rise to drift velocities that are a substantial fraction of the local sound speed, v_drift/c_s ≈ 0.8. These flows have a dramatic effect on the density, which increases in the SOL and develops poloidal asymmetries in the plasma region magnetically connected to the front face of the antenna. The convective cells cause a density depletion at the antenna midplane that increases with antenna current until a threshold current. The 2-D density modification is physically consistent with the calculated 𝑬�𝑩𝟎 flows.

These results show a plethora of problems that must be solved for successful ICRH operation, even at low antenna powers. These deleterious effects may be mitigated by antenna designs that reduce rectified potentials and utilize PMI-resilient materials. The saturation in the density depletion at the antenna midplane suggests methods for targeted density injection may be successful in improving antenna wave coupling.

Ion dynamics are investigated in a single and dual radio frequency sheath as a function of radius above a 30 cm diameter biased silicon wafer for the first time in an industrial inductively coupled (440 kHz, 500 W) plasma etch tool. Ion velocity distribution (IVD) function measurements in the argon plasma are taken using laser induced fluorescence (LIF). Planar sheets of laser light enter the chamber both parallel and perpendicular to the surface of the wafer in order to measure both parallel and perpendicular IVDs at thousands of spatial positions. A fast (30 ns exposure) CCD camera measures the resulting fluorescence with a spatial resolution of 0.4 mm. The dual-frequency bias on the wafer is comprised of a 2 MHz low frequency (LF) bias and a 19 MHz high frequency (HF) bias. The laser is phase locked to the LF bias and IVD measurements are taken at several different LF phases. Ion energy distribution (IED) function measurements and calculated moments are compared for several cases. For the LF case (no HF), the IEDs were found to be highly phase dependent and were varied radially up to 10%. Calculated mean velocity vectors showed large impact angles near the surface of the wafer with the largest angles observed near the wafer edge. The LF experimental results are compared with simulations designed specifically for this particular plasma tool and showed good qualitative agreement. For the dual frequency case, IEDs were measured at two disparate phases of the phase-locked LF bias. IEDs were found to be multi-peaked and were well-approximated by a sum of Maxwellian distributions. The calculated fluxes in the dual frequency case were found to be substantially more radially uniform than the single frequency bias case. For industrial applications, this radially uniform ion flux is evidently a trade off with the undesirable multi-peaked structure in the IEDs.

The objective of this dissertation is to characterize the physics of a boundary layer between a magnetized plasma and a neutral gas along the direction of a confining magnetic field. A series of experiments are performed at the Enormous Toroidal Plasma Device (ETPD) at UCLA to study this field aligned Neutral Boundary Layer (NBL) at the end of the plasma. A Lanthanum Hexaboride (LaB6) cathode and semi-transparent anode creates a magnetized, current-free helium plasma which terminates on a neutral helium gas without touching any walls. Probes are inserted into the plasma to measure the basic plasma parameters and study the transport in the NBL. The experiment is performed in the weakly ionized limit where the plasma density (ne) is much less than the neutral density (nn) such that ne/nn < 5%.

The NBL is characterized by a field-aligned electric field which begins at the point where the plasma pressure equilibrates with the neutral gas pressure. Beyond the pressure equilibration point the electrons and ions lose their momentum by collisions with the neutral gas and come to rest. An electric field is established self consistently to maintain a current-free termination through equilibration of the different species' stopping rates in the neutral gas. The electric field resembles a collisional quasineutral sheath with a length 10 times the electron-ion collision length, 100 times the neutral collision length, and 10,000 times the Debye length.

Collisions with the neutral gas dominate the losses in the system. The measured plasma density loss rates are above the classical cross-field current-free ambipolar rate, but below the anomalous Bohm diffusion rate. The electron temperature is below the ionization threshold of the gas, 2.2 eV in helium. The ions are in thermal equilibrium with the neutral gas.

A generalized theory of plasma termination in a Neutral Boundary Layer is applied to this case using a two-fluid, current-free, weakly ionized transport model. The electron and ion momentum equations along the field are combined in a generalized Ohm's law which predicts the axial electric field required to maintain a current-free termination. The pressure balance criteria for termination and the predicted electric field are confirmed over a scaling of plasma parameters. The experiment and the model are relevant for studying NBLs in other systems, such as the atmospheric termination of the aurora or detached gaseous divertors.

A steady state modified ambipolar system is measured in the ETPD NBL. The drift speeds associated with these currents are a small fraction of the plasma flow speeds and the problem is treated as a perturbation to the termination model. The current-free condition on the model is relaxed to explain the presence of the divergence free current.

The energization of ions in a dynamic system containing two colliding magnetic flux ropes is investigated in an experimental study presented in this dissertation. Two kink-unstable flux ropes on the Large Plasma Device (LAPD) are made to collide in order to trigger magnetic reconnection, a process in which magnetic energy stored in the fields is dissipated into thermal and kinetic energy that is picked up by surrounding ions. The local energy distribution of the ions is measured by a four-grid ion retarding field energy analyzer that was constructed specifically for this study. The average energy ⟨2E⟩, a quantity equivalent to the Maxwellian temperature for non-Maxwellian distribution functions, was plotted as a function of time for two different flux rope conditions. In both cases, ⟨2E⟩ spikes indicated the presence of an ion beam with a sub-Alfv�nic drift velocity of 9 to 15�eV. The beam does not appear to be heated and the ion temperature of the ropes is estimated to be between 4 to 6�eV. This is found to be consistent with a spectrometer’s line-of-sight, volume-averaged measurement involving Doppler broadening of the 320.3�nm He�II spectral line. Using polar plots as a visualization tool for the two-dimensional ion distribution function, the beam appears to travel primarily in the +z direction along the magnetic field and out of the reconnection plane. The presence of the beam is also correlated with magnetic reconnection events that were identified by plotting line contour planes of the magnetic vector potential A_(z). This means that the ions are likely to be accelerated by -dA_(z)/dt, the inductive electric fields created during magnetic reconnection. Furthermore, the energy density produced by the induced electric fields (0.5�J/m�) is comparable to the energy per unit volume required to heat the ions from 6 to 10�eV (0.3�J/m�) after accounting for a less than 50% efficiency of energy transferred to the ions. Overall, the evidence presented in this study strongly suggests that the ions in the beam are produced by magnetic reconnection. To the author’s knowledge, this is the first experimental observation of a field-aligned ion beam generated in a reconnection experiment.

Experiments are performed in the Large Plasma Device

We report on observations of shear Alfven waves radiated from a source of small transverse size, and the subsequent radial confinement of wave magnetic field energy within a cylindrical plasma. The radius of confinement lies between the kinetic regime of the bulk plasma and the inertial regime at the plasma edge; this radius is found to be a function of wave frequency. Numerical calculations using kinetic theory predict a zero in the perpendicular group velocity at a radius which varies in accord with the observations. An analytic expression for the perpendicular group velocity (valid for small perpendicular wave numbers) is given in the vicinity of the zero crossing.