Session PP9: Poster Session VI: Plasma Applications; Reversed Field Pinch; Basic Plasma Physics, Waves; Divertors, Edge Physics, and Fueling

Scaling of electron energies with microwave frequency in ECR ion source plasmas

The presence of high energy electrons cause ECR ion sources to be strong emitters of x-rays. The x-rays present hazards to personnel and can add substantial heat load to the cryostat of modern, superconducting ECR ion sources. Having an understanding of how x-ray energies and intensities scale with microwave frequency will be very important in the design of future, higher frequency ECR ion sources. In this study the effect that microwave heating frequency has on electron energy and x-ray intensity is examined. To explore how the electron energy and x-ray power scale with microwave frequency, x-ray measurements were taken with a semiconductor type x-ray detector on two different ECR ion sources at Lawrence Berkeley National Laboratory. These sources have frequencies of 6.4 GHz and 10-12 GHz/14 GHz. In addition, transient effects such as plasma formation and decay times were measured using diamagnetic loop signals.   

Progress on the ORNL high power, high particle flux helicon hydrogen plasma source

A high power, high particle flux helicon plasma source has been constructed at ORNL. This electrode-less, high ionization fraction RF plasma source is a short pulse ($\leq$ 3 s) prototype that will provide data needed to design a long pulse version for incorporation into in a high power flux ($\sim$ 20 MW/m$^{2}$) linear plasma-surface interaction (PSI) test stand. The source will produce high density hydrogen and deuterium plasmas with density $>$10$^{19}$ m$^{-3}$, and total ion production $\geq$ 10$^{21}$s$^{-1}$. It consists of a 1.3 m long, 15 cm diameter vacuum chamber, with a helical antenna transmitting rf power at f= 10-26 MHz through a cylindrical aluminum nitride (AlN) vacuum window 30 cm in length, and four circular coils creating an axial magnetic field with $|B| \leq$ 0.8T. Results of initial operation will be presented.   

Enhanced Magnetic Field Perturbations and Electric Currents Observed Downstream of the High Power Helicon

The high power helicon (HPH) is a compact plasma source that can generate downstream densities of 10$^{17}$-10$^{18}$ m$^{-3}$ and directed ion energies greater than 20 eV that continue to increase tens of centimeters downstream of the source. In order to understand the coupling mechanism between the helicon antenna and the plasma outside the immediate source region, measurements were made in the plasma plume downstream from the thruster of the propagating wave magnetic field and the perturbation of the axial bulk field. This magnetic field perturbation ($\Delta $B) peaks at more than 15 gauss in strength downstream of the plasma source and propagates tens of centimeters downstream, cancelling the base magnetic field as it propagates. Taking the curl of this measured magnetic perturbation and assuming azimuthal symmetry suggests a peak current density of 20 kA m$^{-2}$. Data will be presented that relates the cancellation of the base magnetic field to the propagation of the helicon wave and the region where the plasma current system closes.   

Ion Energy Distribution Measurements Downstream of the High Power Helicon Plasma Thruster with a Flux Conserving Nozzle Configuration

The high power helicon (HPH) deposits up to 40 kW of power into a plasma, generating a plasma beam with a measured source density of 1x10$^{20}$ m$^{-3}$ and energies in the range of 20-40 eV. Recently, the arrangement of magnetic nozzles downstream of the plasma source has been modified in order to produce a flux conserving configuration. Retarded field energy analyzer (RFEA) measurements of the ion energy distribution functions at two locations downstream of the plasma source, 67 cm and 144 cm away, have been carried out. Data on the number density, ion velocity, and energy density of the plasma beam at these locations will be presented. An improvement in performance over the previous nozzle configuration is observed. Additionally, results suggest that the energy density of the beam does not decrease with distance from the source between the two locations.   

Transport modeling of the ORNL high intensity linear RF plasma source

Recent progress in the electrode-less helicon hydrogenic plasma source\footnote{R.H. Goulding, et al., Proc. 18th Conf on RF Power in Plasmas, Gent, Belgium, June, 2009.} have motivated the development at ORNL of an RF-plasma source that magnetically links a helicon to a mirror cell in which the plasma is heated by EBW, ICH and whistler waves. The $<$4m long plasma column further includes a parallel transport region connected to a pumped target plate. Such a source is modeled at three levels using: a two-point model, a 1D-parallel Braginski's fluid model in which the plasma sources/sinks are computed using the kinetic Monte Carlo neutrals code DEGAS, and the 2D SOLPS code. The required source plasma parameters to achieve certain target plasma parameters, particularly at high plasma heat and particle fluxes, are found to be sensitive to the plasma and neutrals parameters in the helicon and RF mirror cells, the effective heating via various RF techniques, the plasma and neutrals boundary conditions at the target plates and around the RF-plasma heating zones, and the pumped reservoirs with partial backflow of thermal molecules. New results of this investigation will be reported.   

Optical Emission Spectroscopy for CO2 Dissociation using a Dielectric Barrier Discharge (VADER)

VADER (the Versatile Atmospheric Dielectric barrier discharge ExpeRiment) operates at atmospheric pressure and employs high voltage pulses across a quartz dielectric spanning an anode-cathode pair to create a high density, non-thermal, cool plasma in a variety of gasses. In CO$_{2}$ plasmas, energetic electrons from the tail of the non-thermal electron distribution excite CO$_{2}$ molecular states and provide a pathway for CO$_{2 }$dissociation that requires less energy per molecule than conventional thermal dissociation processes. CO$_{2}$ dissociation by-products can then be used as feedstock gasses for chemical synthesis. Here we have used optical emission spectroscopy in the reaction zone of VADER to monitor the density of reaction products and optimize the dissociation process. The optical emission measurements are correlated with real-time residual gas analyzer (RGA) measurements of the discharge exhaust gas.   

Initiation of pentaerythritol tetranitrate using a fast and high power plasma arc source

Initiation of high explosives (HE) is the process of transitioning the HE from a quiescent state to one containing a propagating release of chemical energy. Plasma arc initiation is driven by a discharge across the surface on or through the HE. Experiments have found that at least one conventional high explosive (pentaerythritol tetranitrate, PETN) can be arc-initiated with low threshold input energies. The underlying physics of these thresholds is not yet known. The ability to understand and predict plasma-based initiation is crucial for analyzing the safety of initiation systems. We are studying the high temperature plasma driven HE kinetics in these systems by using a plasma arc source that can deliver $\sim $200 mJ to the HE on the 10's ns time-scale. Here, we present both spatial and temporal characterization of the plasma temperature and density from this source via atomic emission spectroscopy. We also present preliminary kinetics results from time-resolved IR spectroscopic experiments of PETN films driven by these plasmas. Finally, we discuss simulations of these plasmas using a 1-D hydrodynamic model coupled with simple HE kinetics.   

Microwave Plasma Discharge Combustor

This work describes a numerical simulation of an experimental microwave plasma enhanced combustion system. The microwave torch is a co-axial structure where the inner electrode acts as an antenna to transport 2.45 GHz microwaves to the tip of the torch where energy from the plasma is coupled into the flame. The internal dimensions of the torch are adjusted so that an electromagnetic standing wave can be generated. A 2-D electromagnetic, particle-in-cell fluid hybrid simulation of the co-axial structure shows the electromagnetic power flow through the torch. The simulations show a standing wave coupled through the co-axial structure and along the plasma flame. The following profiles are provided: electric field, power flow, energy deposition, plasma density, and excited species. Furthermore, the plasma dielectric constant, which depends on the density and collision frequency, controls the spatial distribution of the standing wave.   

Particle-In-Cell Simulations of a Current-Free Double Layer

A current-free double layer that forms between a magnetized source plasma (upstream) and an unmagnetized plasma in a larger volume expansion chamber (downstream) is studied using a particle-in-cell code. The code is 1D in space and 3D in velocity phase-space. Plasma expansion is modeled by invoking a loss profile in the downstream region. This profile is obtained from the plasma volume expansion that results from the diverging magnetic field of the source chamber. Emphasis is placed on the electron velocity distribution functions (EVDFs). We find that the EVDFs perpendicular to the simulation axis are nearly Maxwellian. Upstream, the EVDF in the parallel direction is significantly depleted in the downstream facing ($+\hat{x}$) direction for energies greater than that required to escape the sheath at the upstream chamber wall. The EVDF is not significantly depleted in the ($-\hat{x}$) direction. Downstream, the EVDF has a two-temperature Maxwellian distribution with an additional depletion in the upstream facing ($-\hat{x}$) direction, corresponding to electrons escaping to the downstream wall. These findings are compared with the EVDFs assumed in previous analytic models and a modified model is developed based on the EVDFs from the simulations.   

Rayleigh Instability in a Hall Thruster: Effect of Ion Temperature and magnetic field

Hall thruster is suitable for a long term mission in space than other electric thrusters. In order to improve the performance of the Hall thruster, we need to understand the inner physical phenomena such as the instability of the discharge current and plasma oscillations. In most of the studies, the ion temperature has been neglected which affects significantly the efficiency and performance of the thruster via thermal motions of the ions. In this paper we have studied the effect of ion temperature and magnetic field strength on the growth rate and perturbed plasma potential. It is found that growth rate and perturbed potential is higher on the higher value of the ion temperature.   

Scaling Laws of Lissajous Helicon Plasma Accelerator toward Electric Propulsion in Space

Scaling law of Lissajous Helicon Plasma Accelerator(LHPA) is derived and tested via PIC simulations with code VORPAL. In the LHPA, rotating transverse electric field in external longitudinal uniform magnetic field drives azimuthal current via ExB drift then thrust is produced due to Lorentz force. An 1D analytical model is developed which includes field penetration and ExB current estimation based on trajectory analysis. Scaling law of thrust as a function of parameters of RF drive frequency, applied RF voltage, plasma density, size of the thruster will be shown.   

Study of Resistive Instability in a Hall Thruster

A Hall thruster is a cross-field device in which ions are accelerated in a quasineutral plasma, because of which a Hall thruster offers much higher thrust density than other types of ion thrusters. A strong electric field is normally established in the Hall thruster channel region of large magnetic fields. This electric field is responsible for the ion acceleration. In order to improve the performance of the Hall thruster, the inner physical phenomena such as the instability of the discharge current and plasma oscillations need to be understood well. In the literature, resistive instabilities in a Hall current plasma discharge have been investigated where it was observed that plasma perturbations in the acceleration channel are unstable in the presence of collisions. However, in most of the studies, the ion temperature has been neglected for the sake of simplicity. In our study, considering the thermal motions of the ions, we have found that an azimuthally propagating mode becomes unstable under certain conditions on the axial distribution of parameters inside the thruster channel. The effect of the temperature and magnetic field on the growth rate of the instability has been studied and it is observed that the growth rate increases for the higher ion temperature and stronger magnetic field.   

A halo-shaped energetic plasma flow from Hall thrusters

The use of permanent magnets instead of electromagnet coils for Hall thrusters can be advantageous for low power space applications. The plasma measurements revealed that the miniaturized cylindrical Hall thrusters with permanent magnets and electromagnet coils operate rather differently. In particular, the ion current density distribution from the permanent magnet thrusters has an unusual halo shape, with a majority of high energy ions flowing at large angles with respect to the thruster centerline. Plasma potential and LIF measurements showed that a stronger magnetic field outside the permanent magnet thruster as compared to the electromagnet thruster alters the electric field distribution in a way that a significant portion of the ion acceleration occurs in a defocusing electric field outside the thruster channel. A high speed imaging revealed a correlation between the electric field distribution and the occurrence of a low frequency E$\times $B rotating spoke in the thruster discharge.   

A comparison of inflection point and floating point emissive probe techniques for electric potential measurements in a Hall thruster plasma

Theory suggests that when increasing the electron emission of an emissive probe the floating potential will saturate $\sim T_e/e$ below the plasma potential. This can introduce significant errors in plasma potential measurements in Hall thrusters where $T_e > 10$ eV. The method of determining the plasma potential from the inflection point of emissive IV traces in the limit of zero emission may give a more accurate measurement of the plasma potential. The two methods are compared in a Hall thruster where $n_e \sim 10^{11}$ cm$^{-3}$, $T_e \sim 20$ eV, and ion flows are significant. The results can be generalized to other types of plasmas.   

Investigation of a Low-Frequency Rotating Spoke in Hall Thrusters

In recent studies~[1], we identified the presence of an E$\times $B rotating spoke in the cylindrical Hall thruster operating with a cusp-shaped magnetic field. The characteristics of the rotating spoke resemble previous observations for conventional~Hall thrusters of the annular geometry with predominantly radial magnetic field [2, 3]. It was suggested that the spoke is responsible for the enhancement of the electron current across the magnetic field. In this work, we use a high speed camera, electrostatic probes and segmented electrodes to measure the fraction of the cross-field current traversing the spoke towards the anode. \\[4pt] [1] J. Parker, et al., \textit{Appl. Phys. Lett} (Submitted 2010) \\[0pt] [2] G. S. Janes and R. S. Lowder, \textit{Phys. Fluids} (1966) \\[0pt] [3] A. I. Morozov, et al., \textit{Sov. Phys. Tech. Phys. }(1972)   

Kinetic simulations of a cylindrical Hall thruster

A newly developed 2d3v Particle-in-Cell code with Monte Carlo (MC) Collisions is used to simulate the operation of the PPPL 100 W cylindrical Hall thruster (CHT). The model includes all relevant collisional processes (Coulomb collisions, electron-neutral elastic, ionization and excitation collisions; ion-neutral momentum transfer and charge exchange collisions). The dynamics of the background gas is self-consistently resolved with direct MC simulation. Secondary electron emission at the thruster walls is accounted through a probabilistic MC model. The anomalous electron transport perpendicular to magnetic field is included in the simulation via Bohm-type diffusion. The computational domain includes the thruster channel and the near-field plume region. The self-sustained and the current overrun modes of the CHT operation are studied and compared with experiments.   

Numerical results of the Hall thruster discharge with an electrode in the near-plume region

Recent experimental results indicated that for Hall thrusters (HT), the use of additional intermediate electrodes between the cathode and the anode placed in the plasma plume or at the channel exit can affect large amplitude, low frequency oscillations of the discharge current and the plasma plume divergence. In this work, we study numerically the effects on the discharge current, and on the plume characteristics of the HT, of an electrode placed in the near-plume region. For that purpose, a 2D hybrid PIC-fluid code is used to simulate the plasma discharge. Preliminary results remark two important aspects. First, the electrode effects also modify the plasma magnitudes in the channel of the thruster, and second, the electron turbulence model is a key factor to understand the plasma discharge.   

Focusing of plasma flow in an E cross B discharge

ExB discharges can be used to accelerate ions in a quasi-neutral plasma. Large ion fluxes can be produced in this way because there is no space charge limitation, however difficulty in specifying the electric field distribution results in large flow divergence [1]. Recent work has identified new methods to control the flow divergence [2,3]. We present the results of new techniques that are designed to further reduce the divergence. \\[4pt] [1] A.I. Morozov and V.V. Savelyev, Reviews of Plasma Physics vol. 21, B. B. Kadomtsev and V. D. Shafranov, Eds. New York: Consultants Bureau, 2000. \newline [2] Y. Raitses, L.A. Dorf, A.A. Litvak, and N.J. Fisch, Journal of Applied Physics 88 (2000) 1263. \newline [3] A. Smirnov, Y. Raitses, and N.J. Fisch, IEEE Transactions on Plasma Science 36 (2008) 1998.   

Magnetic Nozzle Simulation Studies for Electric Propulsion

Electric Propulsion has recently re-gained interest as one of the key technologies to enable NASA's long-range space missions. Options are being considered also in the field of aneutronic fusion propulsion for high-power electric thrusters. To support these goals the study of the exhaust jet in a plasma thruster acquires a critical importance because the need of high-efficiency generation of thrust. A model of the plasma exhaust has been developed with the 3D magneto-fluid NIMROD code [1] to study the physics of the plasma detachment in correlation with experimentally relevant configurations. The simulations show the role of the plasma diamagnetism and of the magnetic reconnection process in the formation of a detached plasma. Furthermore, in direct fusion-propulsion concepts high-energy (MeV range) fusion products have to be efficiently converted into a slower and denser plasma jet (with specific impulse down to few 1000's seconds, for realistic missions in the Solar System). For this purpose, a two-stage conversion process is being modeled where high-energy ions are non-adiabatically injected and confined into a magnetic duct leading to the magnetic nozzle, transferring most of their energy into their gyro-motion and drifting at slower speed along with the plasma propellant. The propellant acquires then thermal energy that gets converted into the direction of thrust by the magnetic nozzle. [1] C. R. Sovinec et al., J. Comput. Phys. 195, 355 (2004).   

Two-Axis Velocimetry Study of Neutrals in a Hall Thruster

We present two-axis laser-induced fluorescence velocimetry measurements of the Xe I 6s'[1/2]$^{0}_{1} \quad \to $ 6p'[3/2]$_{2}$ transition at 834.9 nm (vacuum) inside a 6-kW laboratory Hall thruster. The spatial density of the data is higher near the walls than in the center of the channel to better capture the neutral-wall and neutral-sheath interactions. The thruster is operated under seven conditions spanning discharge voltages of 150-600 V and anode mass flow rates of 10-30 mg/s. The magnitude of velocity varies from 200 to 400 m/s along the channel centerline of the interrogated region but can go up to a few km/s near the downstream end of the channel walls. These high speed neutrals are speculated to be products of charge-exchange reactions and are only detectable over very small regions. The neutral velocity exhibits greater sensitivity to changes in anode mass flow rates than in discharge voltage. Comparison with simulation data shows that the presence of the plasma greatly affects the behavior of the neutral gas in this Hall thruster.   

Experimental study of electric wind produced by a tubular actuator

Three kinds of voltage waveforms (sinusoidal AC, positive/negative half-wave AC) are applied to a tubular DBD plasma actuator. The speeds and directions of wind induced by the three voltage waveforms are compared. It's found that the wind speed created by the sinusoidal high voltage is much higher than that of the positive or the negative half-wave voltage. On the other hand, the speeds and the directions created by the positive and negative half-wave voltage are nearly the same. This phenomenon suggests that the surface charging significantly affects the wind speed. In addition, it is found that the wind directions have no relation to the directions of electric field. The same results are obtained by both the tabular DBD and the DC corona discharge actuators. It suggests that, when the active electrode is positive, the plasma is positively charged and accelerated by the electric field to move away from the active electrode. When the active electrode is negative, the negative ions are predominant, which also moves from the active electrode to the ground electrode under the electric field. Therefore the induced electric wind has the same direction with the positive voltage.   

Laser initiated, RF sustained air plasmas

Measurements and analysis of air breakdown processes and plasma production by focusing 193 nm, 260 mJ, 10 MW high power laser radiation inside a 6 cm diameter helical RF coil. We observe quantum resonant multi-photon (REMPI)2 and collisional cascade laser ionization processes that produce high density (ne$\sim$7 $\times$ 1016/cm$^3$) cylindrical seed plasmas at 760 Torr. The focused laser and associated shock wave produces a seed plasma for sustaining by the RF (1-10 kW, 0.5-1.5 s) pulse. Measurements of the helical RF antenna load impedance obtained by measuring the reflection coefficient with and without the laser pulse and 105 mm wave interferometer density and temperature measurements are made. They demonstrate that the laser formed seed plasma allows RF sustainment at higher initial air pressures (15-30 Torr) than with RF only initiation. Spectroscopic measurements of the plasma and comparison with the SPECAIR code are made to determine rotational and vibrational temperatures. Comparison of the experimental measurements of helical antenna plasma loading with the ANTENAII code will be made and discussed.   

Characterization of RF-driven Plasma Filaments in a Plasma Globe

Filamentary structures have been observed in many types of plasma discharges, such as in DC sparks and AC dielectric barrier discharges (DBDs). Recent progress has been made in characterizing these structures, though their exact physical origin remains unclear. Commercial plasma globes (or plasma balls) are RF discharges in a primarily neon gas mixture near atmospheric pressures that clearly display filamentation. Recent work has provided the first characterization of plasma globe filaments [Campanell \textit{et al}, Physics of Plasmas 2010]. We have extended this initial work to investigate in greater detail the voltage dependence of filamentation, and also include observations on the role of filament flaring, branching, and system hysteresis. Our preliminary results using a custom apparatus will be presented.   

Preparation of amorphous carbon films using pulsed atmospheric-pressure glow discharges in a three-electrode configuration

Pulsed glow discharge plasmas at atmospheric pressure in a three-electrode configuration were developed in order to prepare large-area plasma processes for preparation of amorphous carbon films. The high-voltage, high-repetition bipolar pulse with a fast rise time was applied between parallel-plate electrodes with quartz glasses as dielectric barrier to generate the pulsed glow discharges using a mixed gas of He as a carrier gas and CH$_{4}$ as a precursor. The glow-like plasma with CH radicals was extracted to the stainless-steel substrate by applying the same potential as the parallel-plate electrodes. It was confirmed that an amorphous carbon film with a length of approximately 170 mm was successfully synthesized by the pulsed glow discharges at atmospheric pressure. However, a transition of the discharge behavior to a streamer-like discharge when increasing the input power. Therefore, it is planned that ceramic plates are used as dielectric barrier with large permittivity in order to increase plasma density. In addition, the bias voltage will be independently controlled to produce a stable glow discharge.   

Controllability of the arc plasma-based synthesis of single-walled carbon nanotubes

The focus of this work is to understand the mechanism of magnetic-field-enhanced plasma synthesis, further to establish the fundamental correlation between parameters of arc plasma and characteristics of single-walled carbon nanotubes (SWNT) and increase the controllability and flexibility of arc discharge method. The influence of magnetic field on SWNT parameters is demonstrated as following: (i) It can increase the length of SWNT by a factor of 2 and the population of long nanotubes with the length above 5 $\mu $m. (ii) It can result in substantial fractions of produced SWCNTs being of small diameter, less than 1.3 nm. (iii) It can change the chirality distribution of SWNT and the ratio of metallic to semiconducting SWNT. Additionally, the explanations of these findings are presented in the study of voltage-current characteristics of arc plasma, the analysis of size distribution of catalyst particles, the diffusion model of carbon adatom by Monte Carlo and numerical simulation of arc discharge ablation.   

Graphene production in magnetically enhanced anodic arc

Graphene is a one-atom-thick planar sheet of sp$^{2}$-bonded carbon atoms that are densely packed in a honeycomb crystal lattice. This new material, which combines aspects of semiconductors and metals, could be a leading candidate to replace silicon in applications ranging from high-speed computer chips to biochemical sensors. However, before graphene sheets can be applied to commercial applications, it is necessary to find lower cost methods of mass production. Recently, a new method of graphene synthesis in magnetically controlled anodic arc discharge was discovered.\footnote{M. Keidar, A. Shashurin, O. Volotskova, Y. Raitses, and I. I. Beilis Phys. Plasmas 17, 057101 (2010)} The effect of external magnetic field application to the discharge zone on production yield of graphene flakes is studied using specially-designed magnetic system able to create different magnetic field configurations with magnitudes of up to several kGauss. The considered here method may have broad commercial impact on production of bulk graphene for different energy, electronics, aerospace, mechanical, civil, and biomedical applications, and especially for newly emerging ultracapacitor industry.   

Effect of electronically excited states on the transport properties in thermal plasmas

Transport properties of LTE thermal plasma are studied by solving a set of Boltzmann equations in the framework of Chapman -- Enskog method. As the convergence of higher order contributions in plasma is slow as compared to that in gases, so a detailed investigation of the variation of these contributions with with temperature and pressure has been carried out. It has been observed that increase of the pressure shifts the ionization equilibrium towards the higher temperature, thereby exhibiting the strong pressure dependance. The electron transport properties such as electron thermal conductivity and electrical conductivity are affected particularly at high pressure by taking and neglecting electronically excited states of the plasma species.   

Plasma Flow Measurement using Mach Probe and Laser-Induced Fluorescence in Weakly Magnetized Plasmas

A Mach probe (MP) is generally used for the measurement of plasma flow velocity in the edge of magnetic fusion devices, space propulsion systems, processing plasmas, sheath and pre- sheath regions. Although several un-magnetized MP theories are available, their validity has not been fully confirmed, and should be checked by comparative (or simultaneous) measurement with another diagnostic tool such as laser-induced fluorescence (LIF). The plasma flow velocity was measured via an MP and LIF method in Ar plasma generated by LaB$_6$ cathode of Diversified Plasma Simulator - Modification (DiPS-M). The MP and LIF measurements are performed in the ranges of magnetic field intensities, $100 < B < 1000$ (Gauss) and the various neutral pressures, $1 [Preview Abstract]

Biomedical applications of the cold atmospheric plasma jets

The unique chemical and physical properties of cold atmospheric plasmas enable their numerous recent applications in biomedicine. This report is focused on the investigation of the interaction between the cold helium atmospheric plasma jet and living tissue. This study considers the ability of cold atmospheric plasmas to impact cell migration rates as a function of 1) the length of the plasma treatment time, 2) the number of hours after treatment that cell migration is assessed, and 3) localization of the treatment zone. Data show that the ability of plasma to reduce cell migration rates increases as a function of treatment time with a maximum of 30{\%} and that this affect persists for 33 hours after plasma treatment. Further, the persistence of the cells migration was studied as well. Fluorescence microscopy was used to evaluate inter-cellular changes. In order to characterize the optimal condition for cell treatment with plasma jet, its UV-vis-NIR spectrum was evaluated, as well as simulation of the cold plasma jet behavior was performed.   

High-speed sterilization technique using dielectric barrier discharge plasmas in atmospheric humid air

The inactivation of \textit{Bacillus atrophaeus }spores by a dielectric barrier discharge (DBD) plasma produced by an ac voltage application of 1 kHz in atmospheric humid air was investigated in order to develop low-temperature, low-cost and high-speed plasma sterilization technique. The biological indicators covered with a Tyvek sheet were set just outside the DBD plasma region, where the air temperature and humidity as a discharge gas were precisely controlled by an environmental test chamber. The results show that the inactivation of \textit{Bacillus atrophaeus }spores was found to be dependent strongly on the humidity, and was completed within 15 min at a relative humidity of 90 {\%} and a temperature of 30 \r{ }C. The treatment time for sterilization is shorter than those of conventional sterilization methods using ethylene oxide gas and dry heat treatment. It is considered that reactive species such as hydroxyl radicals that are effective for the inactivation of \textit{Bacillus atrophaeus }spores could be produced by the DBD plasma in the humid air. Repetitive micro-pulsed discharge plasmas in the humid air will be applied for the sterilization experiment to enhance the sterilization efficiency.   

Particle in Cell Simulations of Double Layers in Xe-Ar Helicon Plasmas

Recent experiments in plasma sheaths at grounded boundaries confirmed theoretical predictions that in a multi-ion species plasma presheath, ions do not accelerate up to their own Bohm speed but instead accelerate up to a common bulk sound speed at the sheath-presheath boundary. Since the double layer that forms in expanding helicon source plasmas is essentially a plasma sheath in the plasma volume instead of at the plasma boundary, it was expected that similar effects would be observed in the double layer presheath. In experiments on the Hot hELIcon Experiment (HELIX) at West Virginia University, we find that argon ions in a majority xenon plasma accelerate up to the common speed at the boundary of the double layer also. Simulations of these plasmas have been performed using a particle-in-cell code with Monte Carlo collisions, modified to handle the presence of two ion species in the plasma, We present majority ion flow data at different spatial locations as a function of minority ion doping fraction, and compare with the experimental results.   

2D/3D Monte Carlo Feature Profile Simulator FPS-3D

Numerical simulation of etching/deposition profiles is important for semiconductor industry, as it allows analysis and prediction of the outcome of materials processing on a micron and sub-micron scale. The difficulty, however, is in making such a simulator a reliable, general, and easy to use tool applicable to different situations, for example, with different ratios of ion to neutral fluxes, different chemistries, different energies of incoming particles, and different angular and energy dependencies for surface reactions, without recompiling the code each time when the parameters change. The FPS-3D simulator [1] does not need recompilation when the features, materials, gases, or plasma are changed -- modifications to input, chemistry, and flux files are enough. The code allows interaction of neutral low-energy species with the surface mono-layer, while considering finite penetration depth into the volume for fast particles and ions. The FPS-3D code can simulate etching and deposition processes, both for 2D and 3D geometries. FPS-3D is using an advanced graphics package from HFS for presenting real-time process and profile evolution. The presentation will discuss the FPS-3D code with examples for different process conditions. The author is thankful to Drs. S.-Y. Kang of TEL TDC and P. Miller of HFS for valuable discussions. \\[4pt] [1] P. Moroz, URP.00101, GEC, Saratoga, NY, 2009.   

Simulation studies on the role of ion-enhanced field emission in micro-discharges

Due to its promising future for a number of applications, micro-discharges have become more and more popular in the past decade. For micro-discharges with electrode spacing $<$ 10 $\mu $m, a very interesting feature is the possibility of electron field emission playing an important role. The short distance between the two electrodes enables a very high electric field under a reasonable electric potential. This high electric field makes field emission strong enough to serve as another electron source in the micro-discharge. This work investigates the impact of field emission by embedding it into a particle-in-cell (PIC) simulation. The well-developed Fowler-Nordheim field emission model does not take ion effects into accounts. Therefore to make it a step further, an ion-enhanced field emission model is built in this work. Parametric studies are used to assess the impact of ions on the field emission current density and other properties of the discharge itself. In addition, a revised emission model based on an ion-affected potential distribution is also implemented and is discussed in this work.   

Analytical and Numerical Studies of Non-Local Kinetics in Glow Discharge*

We have studied the formation of electron velocity distribution function (EVDF) in different parts of the energy range in a short glow discharge in light gases. Analytical models are simplified by the fact that the EVDF for most of the electrons produced in the cathode layer is highly anisotropic, and remains so as they propagate into the negative glow (NG). Another group of non-thermal electrons, originating closer to the cathode layer-NG boundary, has sufficiently small kinetic energies to undergo several collisions within NG and become close to isotropic (these energies can still be several times larger than the ionization potential). The former group can be described by a 1D EVDF, with scattering accounted by small corrections, and the latter group, which also includes the electrons produced in NG, can be treated under a diffusion approximation. Both groups are strongly non-Maxwellian. We present the analytical model and compare it to simulation results obtained with a particle-in-cell code EDIPIC. *This work was supported by the US DOE under Contract DE-SC0001939.   

Full-Scale 3D Simulation of a sputtering magnetron

PIC simulations have been used to study ion energy distributions in magnetron plasmas, and coupled with other simulations to relate plasma processes to properties of sputtered films. The plasma is weakly ionized and exchanges heat with the background gas by scattering and charge-exchange reactions. Resulting heating of neutral background gas up to $\sim $1200K, leading to $\sim $5X rarefaction and increased plasma impedance, was studied with coupled PIC and Direct Simulation Monte Carlo (DSMC) simulations. Effects of scaling the PIC simulations from 0.1X to 1X physical size, and modifying the plasma potential by a dc substrate bias, will be presented. Comparison to experimental I-V relations and importance for roughness and density of sputtered films will be discussed.   

Overview of MST results and plans

Advances in RFP performance and basic physics studies on MST are summarized. Pellet injection increases the density of improved confinement plasmas well above the empirical limit (up to $n/n_G$=1.5), and the beta value is likewise increased, without evidence for a hard limit. A 1 MW tangential neutral beam injector is operational, sparking several new topics for RFP research. The observed good confinement of fast ions bodes well for increasing beta and assessing energetic ion effects. The NBI's directed momentum affects the plasma flow, and altered tearing mode dynamics suggest localized current drive in the core. Energy confinement with oscillating field current drive is measured about the same as for steady induction, a promising result for this efficient form of current sustainment. Magnetic self-organization studies reveal that ion heating during magnetic reconnection events is increasingly anisotropic at lower density. Detailed measurements of the flow and magnetic structure of the reconnection region around the $q$=0 surface emphasize the importance of nonlinear mode coupling and two-fluid physics. Several new diagnostics are being implemented, including fast Thomson scattering, neutral particle analyzers, and fast ion CHERS. Continuing investigation of lower hybrid and electron Bernstein waves will be summarized. Work supported by USDoE and NSF.   

Neutral Beam Injection Systems on the MST

Recent upgrades to a 50kV diagnostic neutral beam (DNB) and the addition of a 1 MW 20 ms neutral beam injector (NBI) on MST posed many technical challenges in design, installation and commissioning. The DNB upgrades improved beam divergence and energy distribution and reduced temporal fluctuations in beam energy. The addition of the 1 MW NBI required many novel engineering solutions, such as a narrowly focused (10cm) beam and an ultra-high pumping-speed titanium-arc getter pump, to allow the injection of the beam into the MST plasma with minimal perturbations to the thick conductive shell and neutral pressure. An innovative snubber circuit mitigates the effects of parasitic stored energy in the long transmission lines separating the beam sources and power supplies, which are required due to space constraints. Both systems are largely self-contained with independent vacuum and control systems; the NBI is fully automated with a computer interface allowing remote control and data acquisition. Work supported by USDoE and NSF.   

Confinement and dynamics of neutral beam injected fast ions in the MST Reversed Field Pinch

The new 1MW neutral beam injector (97{\%} H, 3{\%} D) on MST provides a good test-bed for study of fast ions in the RFP. Analysis of the D-D fusion neutron flux decay at beam turn-off reveals that the confinement time of the fast ions is at least 10 ms, ten-fold larger than the thermal conferment times for particles and energy in standard stochastic plasmas. Also, the fast ion confinement increases with magnetic field strength. Dependence of fast ion confinement on plasma parameters, beam energy, and injection direction will be characterized and compared with TRANSP simulations. In addition, an advanced neutral particle analyzer and a prototype of fast ion charge exchange spectroscopy are under construction to measure neutralized fast ions and induced Doppler-shifted H$\alpha $ light, respectively, thereby resolving fast ion density and energy distribution. Initial measurements of fast-ion dynamics during magnetic reconnection events will be presented.   

Neutral Beam Heating On MST

A 1 MW, 20 ms tangential neutral beam injection system has recently been installed on MST. One of the key impacts that a high-power neutral beam has on a plasma is particle heating. Significant ion heating has been measured, with an increase in core Ti of more than 40 eV occurring after only a few milliseconds in a plasma with an initial ion temperature of approximately 200 eV, and a density of around 1E13 cm$^{-3}$. Simulations of classical collisional heating were performed, using TRANSP beam deposition simulations in a 1-D temperature profile evolution model. This model assumes a constant stored energy when the beam is not fired, and also assumes an anomalous ion heating source. Variations in parameters such as energy confinement time change the simulated heating results, but it is very difficult to replicate the observed heating by classical collisions alone. Measurement of the evolution of the electron temperature profile is currently underway, and will aid in the understanding of energy balance in MST, and the complete impact of the neutral beam on plasma temperature. Work supported by the USDOE.   

Effect of Neutral Beam Injection on QSH onset in the RFP

The recent installation of a 1MW neutral beam injector (NBI) on the Madison Symmetric Torus has enabled studies of several different phenomena of the RFP plasma. Non-reversed discharges are prone to conversion to a quasi-single helicity (QSH) state mode where the core-most resonant tearing mode (m=1,n=5) becomes significantly larger than all other modes within the plasma. The island experiences a significant drag torque and quickly decelerates and locks in the lab frame. It is observed that launching 1MW of 25kV neutral H atoms into these discharges can suppress the transition to QSH, resulting in significantly lower magnetic fluctuation amplitude. With this and the considerable momentum injected via NBI, the magnetic island can rotate much faster than in non-NBI discharges. Wall interactions are reduced and a substantially longer pulse length is observed. One possible explanation of QSH suppression is a small core-localized NBI current drive, which reduces the on-axis safety factor to slightly below 0.2 and removes the resonance condition. Equilibrium reconstructions can illustrate this subtle change in current density profile; no change to the total toroidal plasma current is measured. \noindent Work supported by USDOE.   

Study of Alfv\'{e}n Eigenmodes on the Madison Symmetric Torus

Alfv\'{e}n waves are likely of fundamental importance in the reversed-field pinch (RFP). Unstable tearing fluctuations can inject energy into Alfv\'{e}n modes, which could excite broadband magnetic turbulence and anomalous ion heating. An effort is in progress to understand toroidicity-induced Alfv\'{e}n eigenmodes (TAE's) through their structure, driving terms and damping mechanisms on the MST. Coupling of multiple continuum modes can introduce undamped Alfv\'{e}n eigenmodes with frequencies up to 1 MHz. These modes can also become unstable by inverse Landau damping due to fast ions, a condition pertinent to neutral beam injection heating on MST and fusion alpha particles in future RFP devices. Frequencies of weakly damped modes have been calculated by solving a 3D partial differential equation that describes shear Alfv\'{e}n dynamics numerically based on MST equilibrium conditions. To excite the calculated modes, a single strap poloidal antenna connected to a 1 kW broadband amplifier will be employed. A toroidal array of 32 fast magnetic loops resolves power spectra and mode numbers in the relevant range of frequencies. Preliminary data indicates some coherent magnetic activity below TAE frequency ranges. Supported by USDoE and NSF.   

Investigating the $\beta $ limit on MST with pellet injection and NBI

Pellet injection into improved confinement ohmically-heated MST plasmas has resulted in a density exceeding substantially the empirical Greenwald Limit and a total $\beta $, normalizing to the magnetic pressure at the plasma boundary, of 26{\%}. Although tearing mode amplitudes are larger at higher $\beta $, a clear $\beta $ limit has not been observed. The addition of a 1 MW heating neutral beam on MST will provide additional heating which will be utilized to further probe for a $\beta $ limit. The fast ion confinement is measured to be several times greater than the thermal particle confinement time, but the thermalization time tends to be long for low density improved confinement plasmas. The high density, low temperature pellet fueled plasmas should provide an ideal target for deposition of the beam power to the plasma. Work Supported by USDOE.   

Two-fluid and gyroviscosity effects on pressure-driven instabilities

Pressure-driven instabilities in reversed field pinch associated with unfavorable magnetic curvature, which becomes important at high beta, could limit confinement. Recently, a record high plasma beta of 26\% for the improved confinement MST has been achieved with no severe side effects. Here we examine in toroidal geometry the behavior of resistive interchange instability using the extended MHD code NIMROD. Equilibrium profiles from MSTFit are fitted and imported into the Grad-Shafranov solver NIMEQ. Two-fluid and MHD stability analysis of the high beta MST plasma is then performed using NIMROD at a Lundquist number of S=$10^6$. We perform computations for two sets of high beta MST equilibria. In the first equilibrium, the Mercier criterion parameter exceeds the ideal stability limit and in the second equilibrium the plasma is ideally stable. We find that when the ideal stability limit is violated, finite Larmor radius (FLR) effects (in the form of ion gyroviscosity and the Hall term in generalized Ohm's law) suppress the growth rate of localized resistive interchange modes, however they are not completely stabilized. Nonlinear two-fluid single helicity computations for pressure-driven modes are also presented. It will be shown that mean flows which are mainly concentrated in the outer half of the plasma volume due to the m=0 mode perturbation, are generated.   

Maximizing MST's Inductive Capability with Programmable Bt and Bp Power Supplies

There are now strong motivations on MST for increased plasma current, increased pulse length, and flexible waveform control. For example, improved confinement plasmas at near MST's present maximum current, 0.5 MA, exhibit RFP-record Te and Ti that are several-fold larger than temperatures achieved at 0.2 MA. The maximum dc current added by ac oscillating field current drive is not yet known since the added current ramps up on a time scale longer than the duration of present MST discharges. Inductive current profile control has generated substantially improved confinement, but the optimal control waveform and, hence, the maximum confinement are not yet known. These and other considerations motivate the installation of flexible, programmable power supplies. A power supply for Bt is already in operation. Based on IGBT switches, this supply provides waveform control with a bandwidth of several kHz at about 25 MW. The design of a similar but more powerful supply for Bp is underway. Together, these supplies will provide MST with the most advanced inductive control capability of any RFP in the world.   

Electron temperature and density fluctuations during improved confinement plasmas

We plan to present initial results of Thomson scattering measurements of $T_e$ and $n_e$ fluctuations associated with electrostatic transport during improved confinement plasmas (PPCD). Recent upgrades to the MST Thomson scattering diagnostic have allowed detailed measurements of electron temperature fluctuations in a variety of plasmas. For example, previous studies have mapped temperature fluctuations to magnetic perturbations, observing remnant island structures between sawteeth events in standard plasmas. Electron temperature fluctuations have also been shown to decrease significantly during PPCD plasmas and it is thought that electrostatic fluctuations may dominate transport. Further upgrades to the Thomson diagnostic may enable absolute density measurements, and the planned addition of a fast laser system will improve maximum time resolution from 25 kHz to 250 kHz. These capabilities will allow investigations into correlated $T_e$ and $n_e$ fluctuations associated with electrostatic transport. This work supported by the USDoE.   

Limits of stochastic thermal transport in PPCD discharges in MST

High current improved confinement plasmas in MST exhibit a large increase in the central electron temperature to $\sim $ 2 keV, and the temperature gradient region broadens. Interestingly the maximum temperature gradient tends to occur in a region near the toroidal field reversal surface (r/a $\sim $ 0.8) where the density of resonant surfaces for tearing modes is highest. Recent upgrades to the Thomson scattering diagnostic on MST have greatly improved the capability for single-shot profile measurements with high time resolution. Improved confinement discharges obtained using Pulsed Poloidal Current Drive (PPCD), a form of transient inductive current profile control, can therefore be analyzed to better understand the mechanisms which lead to the best confinement performance. This permits an investigation of the role of residual stochasticity of the magnetic field in energy transport, taking advantage of the shot-to-shot variability in the level of tearing mode suppression that happens naturally with PPCD. Measurements of the evolution of the Te(r,t) profile and local power balance are compared with the evolution of the tearing mode spectrum to investigate this physics. Supported by USDoE.   

Helical Magnetic Self-Organization in the RFX-mod and MST devices

Self-organization of the reversed field pinch with large helical structure (QSH regimes) is predominant as plasma current is increased. In RFX-mod, the persistence and strength of the QSH state increases markedly above 1 MA. An internal transport barrier appears, and plasma thermalization within the helical magnetic surfaces reflects improved confinement. The QSH regime is also obtained in MST plasmas, which operates with plasma current up to nearly 0.6 MA. We report here a statistical analysis of the tearing mode behavior in MST (e.g., amplitudes and QSH persistency) that reveals a trend with plasma current similar to that observed in RFX-Mod. This trend supports an expectation for universal behavior that depends on parameters such as the Lundquist number that vary with the plasma current. Analysis of the common database from the two devices should help reveal key physics for QSH onset and dynamics. Planned Thomson scattering measurements and transport analysis on MST will be important to compare with the confinement behavior established for RFX-Mod. Work supported by USDoE.   

Measurements of plasma potential with a heavy ion beam probe, and momentum transport in improved confinement RFP discharges

Spatially localized measurements of the electric potential with a heavy ion beam probe (HIBP) in the interior of improved confinement reversed-field pinch plasmas indicate positive potentials $\sim $ 1-1.5 kV which decrease as a function of time. This suggests a mitigated loss of electrons due to reduction of magnetic stochasticity. The potential magnitude loosely tracks the dominant magnetic mode ($m$=1, $n$=6) velocity (flow); dependencies on profiles of $T_{e}$, $T_{i}$, and $n_{e}$ are also speculated. Ambipolarity constraints on stochastic particle transport predict an outwardly directed electric field which couples to momentum balance (but fluid stresses have also been measured to be large). These issues are essential to understanding particle and momentum transport, yet there is not a good explanation for the origin of rotation in ohmically heated RFP plasmas. We will present initial investigations of these issues using a combination of theory, modeling, and HIBP measurements.   

Prediction and refinement of magnetic equilibrium and Heavy Ion Beam trajectories in MST

Operation of a heavy ion beam probe on the Madison Symmetric Torus presents a unique fundamental challenge in accurate prediction of beam trajectory and sample volume. The magnetic field, generated primarily by plasma currents, is central to this issue; it is determined by Grad-Shafranov equilibrium reconstructions produced by MSTFit. The precision of the reconstruction is dependent on both components of the model and diagnostic data used as constraints. Efforts to model magnetic field (error and fluctuation) asymmetries and fine-tune equilibrium plasma profiles have been successful in more precisely computing the heavy ion trajectory. As the beam path is largely determined by the magnetic field, it also contains unique information about the field. Accurate modeling of fluctuation profiles and beam trajectories allows extraction of information which can be used to determine local fluctuation magnitudes. The process of determining the trajectory which not only benefits from, but also improves, the accuracy of equilibrium reconstructions is discussed. This work is supported by the US DOE.   

Advances in the measurement of electrostatic fluctuations and inference of particle flux with a Heavy Ion Beam Probe in improved confinement MST plasmas

Electrostatic fluctuations may be the dominant transport mechanism in improved confinement MST plasmas when magnetic fluctuations are suppressed. Simultaneous measurements of density and potential fluctuations have been made with a heavy ion beam probe in the plasma interior; spectra are broadband with most power below 100 kHz. A noise reduction method has been developed to reduce UV induced detector noise to a level on the order of other sources. A technique to estimate wavenumbers ($k)$ which assumes a fluid fluctuation frequency Doppler shifted by a plasma flow velocity is being explored; mean $k$ estimates give $k$*\textit{$\rho $}$_{s}$ = 0.14$\sim $0.21 which are within a range of ITG modes recently predicted by simulations. The noise reduction and $k$ estimate advances are enabling the inference of electrostatic particle flux and its study relative to estimates from previous work on global particle balance in core of MST. Measurements and comparisons with modeling will be presented. (Work supported by US DOE.)   

Development of o-carborane boronization for MST

A boronization technique using o-carborane sublimation, which is much safer and simpler than those using diborane or trimethylboron, is being developed for MST. Due to the pulsed nature of the wall-conditioning discharges used in MST, efforts have been made to increase the repetition rate of the discharges in order to increase the boronization duty cycle. The gaseous o-carborane, created by heating a powder, can be injected in a pulsed manner. In this case, throughput is enhanced by the flow of helium, the working gas of the discharge, but is limited by the short opening time of the puffing valve. Density control and plasma parameters such as D-alpha and impurity radiation will be compared before and after the boronization. Two aluminum coupons flush with the inner wall of the MST vessel, one located near the o-carborane injection port and the other opposite to it, will be used to analyze the film and to examine the uniformity of the coating.   

Measurement and Modeling of Fast Electrons with Lower Hybrid in MST

Lower hybrid wave injection experiments are underway on the MST RFP, driving current at r/a $\sim$ 0.75 in order to stabilize tearing fluctuations. Predicting the effects of LH deposition requires a detailed understanding of the fast electron population. CQL3D, a Fokker-Planck code, has been used to study this population. X-ray emission has been accurately predicted in reduced-tearing, improved confinement discharges. It has however proved challenging to model standard RFP discharges, where CQL3D is shown to over-predict x-ray emission. Attempts to reconcile the code and experiment are presented. A tungsten target probe has been used to stimulate bremsstrahlung emission during standard discharges with LH. Measurements indicate radially localized fast electrons associated with LH starting at an insertion depth of 3 cm ($r/a \sim 0.94$). X-ray flux is observed to increase into $r/a \sim 0.9$, the limit of probe insertion for high current plasmas. CQL3D can also be used to determine the efficiency of LH absorption in various plasma conditions. Studies indicate that Rechester-Rosenbluth diffusion prevents the efficient accumulation of LH current in highly stochastic plasmas. In addition to these results, estimates of the power required to maintain the current in stabilized plasmas will be presented. Work supported by US DOE.   

Fast electron generation on MST via radio-frequency waves

Lower hybrid current drive has been proposed as a means of improving confinement in the reversed field pinch by reducing tearing fluctuations. The particular constraints of the Madison Symmetric Torus have led to the use of a novel interdigital-line traveling wave antenna structure rather than the traditional waveguide grill antenna. Since hard x-ray (HXR) flux from bremsstrahlung is a standard indicator for current drive, HXR surveys have been performed up to 160 kW of input power. While there are no definite indications of current drive in the HXR regime at the target absorption region, toroidally localized hard x-rays with energies up to 50 keV have been observed. Monte Carlo modeling supports the hypothesis that gradients in the rf electric fields at the antenna face pull out a high energy perpendicular tail in the distribution. Some fraction of these electrons are lost to the antenna structure. Those not immediately lost are subject to trapping which allows for multiple passes through the near-field and can explain asymmetries in the HXR flux.   

Electron Bernstein Wave Studies in MST

The electron Bernstein wave (EBW) has potential to stabilize resistive tearing modes with off-axis current drive for further improvement of RFP confinement. Hardware upgrades to the MST-EBW experiment include a 5.5GHz radar klystron tube capable of 1MW power output driven by a novel resonant switchmode power supply and directed toward the RFP plasma edge through a cylindrical molybdenum wave guide antenna. By utilizing XB conversion, the X-mode evanescently decays in the narrow region between the R and UH layers and couples to the Bernstein mode at the UH layer. The Bernstein wave is strongly damped at the electron cyclotron resonance where it coupled to the electron gyromotion, thereby altering the electron distribution. By external control of magnetic field, either Fisch-Boozer or Ohkawa current drive mechanisms can be activated to drive off axis current in the plasma. Current profile may then be optimized experimentally to reduce particle transport. Initial experiments are presented to verify high power coupling and understand heating via observed x-ray emission and compared to Fokker-Plank modeling.   

Particle and Momentum Transport in a Stochastic Magnetic Field

3-D Resonant Magnetic Perturbations (RMP) have been successfully utilized to suppress or mitigate ELMS in tokamak plasmas. However, the mechanism for particle transport (density pumpout) and flow change in a stochastic magnetic field remains unclear. In the MST reversed field pinch (RFP), where a stochastic magnetic field is produced by multiple overlapping tearing modes, we observe a strong particle pump-out and parallel flow change similar to RMP experiments. Detailed measurements in the interior of the high-temperature RFP indicate that density fluctuations in a stochastic magnetic field play an important role both in particle and momentum transport. The particle flux primarily results from strong nonlinear mode coupling. A common physics basis for fluctuation-induced transport in the tokamak and RFP toroidal magnetic confinement configurations is explored. Work is supported by US DOE and NSF.   

Upgrade of Far-Infrared Interferometer-Polarimeter Diagnostic on MST

Recently, the 3-wave far-infrared interferometric diagnostic system on MST has been improved by (1) upgrading the infrared pump source, (2) developing more precise alignment techniques, and (3) refining system calibration. System upgrades serve to reduce phase noise and lower systematic errors thereby improving overall resolution. This multichannel system can be configured to make combined phase measurements of interferometry/Faraday rotation or interferometry/differential interferometry leading to direct determination of 3 equilibrium (density, magnetic field and current density) and 5 fluctuating (density and density gradient, poloidal and radial magnetic field, current density) quantities. Combined measurements have system bandwidth of $\sim $250 kHz while individual measurements can be up to $\sim $2 MHz. Results for standard sawtoothing and high-performance plasmas will be presented, including 3-D effects during quasi-single helicity conditions and core fluctuation characteristics for MST plasmas with 1 MW neutral beam heating.   

New soft x-ray spectrometer on MST

Measurements of x-ray spectra in the MST are used to investigate the transport of energetic electrons and to estimate the effective charge $Z_{\mathrm{eff}}$. A new set of x-ray detectors is being implemented on the MST to detect x-rays in the energy range of $2$-$10\,\mathrm{keV}$. The new detectors are six Amptek XR-100CR modules with preamplifier and cooling. The detectors are connected to Cremat Gaussian shaping amplifiers with shaping times of either $500$ or $100\,\mathrm{ns}$. The shaping amplifier output is directly digitized at $60\,\mathrm{MHz}$, and the x-ray pulses are processed using a new code capable of correctly fitting multiple overlapping pulses. This configuration should allow a maximum count rate of $2$-$5\,\mathrm{MHz}$. The detectors can be placed in any of 17 ports covering $r/a$ values from 0.87 inboard to 0.84 outboard allowing measurements of inboard-outboard symmetry. The new detectors compliment the current system composed of 13 $\mathrm{CdZnTe}$ detectors detecting hard x-rays in the $10$-$150\,\mathrm{keV}$ energy range. The composite energy spectra of these x-ray diagnostics will be used with CQL3D Fokker-Planck modeling to constrain key parameters such as the electron radial diffusion coefficient and $Z_{\mathrm{eff}}$.   

Fast Electron Transport in Improved-Confinement RFP Plasmas

Hard x rays (HXRs), with energies reaching 150 keV, are detected in MST discharges with reduced tearing mode amplitudes, indicative of improved confinement of fast electrons. In standard discharges, tearing modes create stochastic magnetic fields, fast electrons diffuse out of the core at a rate proportional to their velocity, and emitted x rays do not exceed energies of about 10 keV. By comparison, when tearing modes are sufficiently reduced, magnetic flux surfaces are restored and fast electrons become well-confined, diffusing at a rate independent of velocity. HXRs are measured from the core of PPCD discharges, in which the current profile is inductively modified to reduce tearing mode amplitudes. For these plasmas, the Fokker-Planck code CQL3D can be used to infer $Z_{eff}$ and the particle diffusion coefficient $D_{r}$ from measured spectra; typical values are $Z_{eff}$ = 5 and $D_{r}$ = 5 m$^{2}$/s. HXRs are also detected when a large magnetic island forms in the plasma core, usually the result of quasi-single helicity, where one tearing mode grows large while the rest are suppressed. Stochasticity is reduced within the island and fast electrons are well-confined in this region. Work supported by the USDOE.   

An Upgraded Soft X-Ray Tomography Diagnostic to Measure Electron Temperature on MST

The upgraded soft x-ray (SXR) diagnostic will measure electron temperature on MST using two complementary methods. Both methods are based on the two-color technique, which calculates temperature based on the ratio of SXR bremsstrahlung emission from the plasma in two different energy ranges. Improvements over the previous diagnostic include individual detection diodes to reduce noise, a new geometry that improves tomographic reconstructions, and a new filter design that provides direct two-color measurements for each line-of-sight. The new diagnostic retains the ability to measure temperature from tomographically reconstructed emissivity. Additionally, shared lines-of-sight also enable the two-color technique to be applied directly to the measured brightness. Extensive modeling demonstrates advantages and limitations in both techniques. For example, the direct brightness technique provides a robust radial profile of temperature, while the tomographic technique provides a 2D temperature map but is sensitive to mathematical artifacts. This work supported by U.S. D.O.E.   

Deuterium alpha line measurements and neutral density modeling for the Madison Symmetric Torus

Accurate measurements of deuterium alpha line emissions play an important role in determining many plasma parameters, including neutral particle density, electron source rate, particle confinement time, diffusion rate, and thermal conductivity. Knowledge of the core neutral particle density is required to calculate fast ion confinement time and interpret charge exchange recombination spectroscopy data. In the Madison Symmetric Torus (MST), a filtered photodiode array collects D$_{\alpha}$ photons along several viewing chords at a fixed toroidal location. A new analysis of this chord-integrated data is being developed using the NENE Monte Carlo particle tracing code to model neutral particle diffusion inside the plasma. This method is compared to previous modeling via Abel inversion with a two-dimensional asymmetric correction. Data resolution and fit quality are also being improved by substantially increasing the number of detector viewing chords. These improvements will allow for more accurate calculation of particle transport quantities in MST.   

Advances in the pulse-burst laser system for high-repetition-rate Thomson scattering on MST

A pulse-burst laser has been installed for Thomson scattering measurements on MST. The laser design is a master-oscillator power-amplifier which is capable of \emph{Q}-switching at frequencies between 5-250~kHz. Single pulses through the first (four) Nd:YAG amplifier stages give energies up to 1.5~J, and the gain for each stage has been measured. Repetitive pulsing at 10~kHz has also been performed for 2~ms bursts giving average pulse energies of 0.53~J with $\Delta E/E$ of 4.6\%, where $\Delta E$ is the standard deviation between pulses. The final Nd:glass amplifier stages require flashlamps operated at 1800~V and 1800~A. At these currents, inductive turnoff spikes can become large even for small circuit inductances. The flashlamp power supplies have been modified to reduce inductance and increase snubber capacitance, and now reliably produce pulse trains (10 pulses at 1 kHz) at maximum flashlamp drive current. In addition, the beam path is being extended to the MST vacuum vessel. This work is supported by the U. S. Department of Energy and the National Science Foundation.   

Characterization of electrostatic turbulence in the MST reversed field pinch

Low-frequency fluctuations (10-30 kHz) associated with tearing modes dominate the fluctuation power spectrum in RFP plasmas. High-frequency turbulence is also present and may play a significant role in particle and energy transport and in anomalous ion heating. Recent magnetic fluctuation measurements suggest the broadband turbulence results from a nonlinear cascade, but the nature of the constituent fluctuations is poorly understood. The present work shows results from high-frequency electrostatic fluctuation measurements made with an insertable multi-tip probe in the edge plasma region. Wave number spectra have both power-law and exponential characteristics, indicating regions of constant energy transfer rate from large to small scales, as well as energy dissipation at small scales. High-frequency magnetic fluctuations have been measured simultaneously, allowing the coherence and relative phases between electrostatic and magnetic fluctuations to be determined. Partitioning of the kinetic and magnetic energy will be discussed, as will candidates for the source of high-frequency turbulence.   

Flow Dynamics and Transport in the Edge of MST

Insertable probes are used to investigate plasma flows and transport associated with tearing mode structures that occur in MST during quasi-periodic bursts of tearing mode instabilities (sawteeth). Novel ensemble techniques are used to reconstruct a signal's spatial variation in the rotating reference frame of the plasma from single point probe measurements. This allows for a detailed examination of the flow dynamics and spatial structure during a sawtooth crash. A Mach probe, a spectroscopic probe, and a triple tip Langmuir probe are used to measure components of the plasma flow as well as plasma density and electron and ion temperature. Edge resonant tearing modes phase lock together during a crash and form a complex island structure. Flows associated with this structure are measured and compared to predictions from a nonlinear cylindrical DEBS code and a toroidal NIMROD calculation. Fluctuation-induced particle transport, measured as $\langle\tilde{n}_{\mathrm{e}} \tilde{v}_{\mathrm{r}}\rangle$, increases dramatically during a crash. The flux is found to be non-axisymmetric and correlated with the tearing modes. The implications these measurements have on our understanding of reconnection, momentum transport, particle transport, and ion heating will be presented. Work is supported by the US DOE and the NSF.   

Measurements of Nonlinear Hall-Driven Reconnection in the Reversed Field Pinch

Previous measurements have established that reversed field pinch (RFP) sawtooth relaxation is characterized by spontaneous reconnection occurring simultaneously at multiple sites. Here, we report measurements of the magnetic fields and terms in Ohm's law associated with reconnection in the edge region of MST plasmas. The magnetic field structure is measured by probes and compared with theoretical predictions computed in both toroidal and cylindrical geometry. The composite magnetic structure from modes with toroidal mode numbers n=1-4 resonant at the toroidal field reversal surface reveals a complex but still coherent edge structure. Key terms of Ohm's law for the dominant mode ($n=1$) are accessible from magnetic field measurements and reveal the ordering ($\frac{1}{ne}J\times B >> E>\eta J$), clearly indicating that single fluid physics is not sufficient to explain this reconnection. In particular, nonlinear three-wave coupling through the Hall term acts as a driving mechanism for this linearly stable mode. The observed coherent structures and strong nonlinear interaction terms highlight the substantial role of collective mode phase matching during sawtooth events.   

Oscillating-Field Current-Drive Experiments on MST

Oscillating-field current drive (OFCD) is a proposed method of efficient, steady-state current drive in which applied AC poloidal and toroidal loop voltages interact with magnetic relaxation to produce a DC plasma current. OFCD at a moderate power level is added to Ohmically sustained reversed-field pinch plasmas in the MST device, and its effects on equilibrium profile evolution, global magnetic fluctuations, and energy balance are examined using a variety of measurements. For the optimal phase between the two applied AC voltages, the cycle-average plasma current increases by up to 10\% with Ohmic efficiency, while both the energy confinement time $\tau_{\rm{E}}$ and normalized thermal pressure $\beta$ slightly improve, consistent with a reduction in magnetic fluctuation amplitudes. Nonlinear, 3D, resistive-MHD simulations reproduce the main experimental features, especially the phase dependence of the added current. Internal fluctuation measurements are underway to examine changes in the relaxation dynamics. A new programmable power supply is to be used in optimizing OFCD performance with longer pulses, more power, and improved waveform control, including nonsinusoidal OFCD.   

Magnetic Relaxation with Oscillating Field Current Drive on MST

In oscillating field current drive (OFCD), poloidal and toroidal frequency-matched ac magnetic fields are inductively applied to the plasma to drive dc plasma current through magnetic relaxation. Measurements of the dynamo mechanisms associated with magnetic relaxation are conducted during OFCD both to better understand the relaxation dynamics and to aid in optimizing OFCD performance. The full [$<\tilde {E}\cdot \tilde {B}>$] and Hall [$\frac{<\tilde {j}\times \tilde {B}>}{ne}$] dynamo and the fluctuation-induced magnetic helicity flux [$<\tilde {\phi }\tilde {B}_r >$] associated with magnetic relaxation are measured in the edge using insertable probes. They are enhanced during OFCD by $\sim $100{\%} relative to standard RFP operation and, as expected, exhibit the opposite radial direction of induced transport of helicity. Probes used include a secondary-emission capacitive probe developed to measure electric fields and compared to Langmuir probe measurements. Measurements of the MHD [$<\tilde {v}\times \tilde {B}>$] and Hall [$\frac{<\tilde {j}\times \tilde {B}>}{ne}$] dynamo in the core using charge exchange recombination spectroscopy and far-infrared interferometry-polarimetry are in progress as well. This work is supported by the US DOE.   

Revisiting the Mode-Beating Model of AC Helicity Injection

Oscillating field current drive (OFCD), or AC helicity injection, is an important candidate for current sustainment in reversed-field pinch devices. Bellan examined AC helicity injection in a slab geometry and described it as a beating between two plasma modes that produces a mean current parallel to the equilibrium magnetic field [P. M. Bellan. Phys. Rev. Lett. 54, 1381 (1985)]. This mean current is confined to within a classical resistive skin depth of the plasma surface, and plasma relaxation is responsible for transporting this current to the core. We revisit this analytical work and examine how this wave-beating effect is represented in zero-beta MHD simulations, including consideration of the choice of boundary conditions. In addition to the expected parallel current, numerical simulations show a pinch effect from a cycle-averaged current that is perpendicular to the mean magnetic field, which is not described in Bellan's original work. Our results are discussed with respect to Boozer's general anti-dynamo theorem [A. H. Boozer. Phys. Fluids B Vol. 5, 2271 (1993)].   

Studies of the Two-Fluid RFP Dynamo and Mode Coupling

The nonlinear evolution of finite beta, two-fluid tearing modes in a cylindrical, force-free pinch is investigated with the NIMROD code. A multihelicity case with $R/a\simeq3$ simulates the two-fluid reversed field pinch (RFP) fluctuation driven dynamo electric fields. The model includes warm ions and ion gyroviscosity with an ion sound gyroradius, $\rho_{s}=0.05$, and $\beta=0.1$ which are realistic for the Madison Symmetric Torus (MST). During a relaxation event, we find that the Hall dynamo plays an important role reinforcing the MHD dynamo, driving poloidal current, and reversing the toroidal magnetic field at the wall. We compare this model to single-fluid MHD computations with zero and finite $\beta$. The coupling of the core mode fluctuations nonlinearly drive the $m=0$ edge mode, resonant ($\mathbf{k}\cdot\mathbf{B}_{0}=0$) near the wall, through significant fluctuation induced MHD and Hall contributions to the $m=0$ parallel electric field. Single helicity profiles are compared to the analogous core modes in the multihelicity simulation in order to understand how the nonlinear effects modify the profiles and produce flows near the wall. Comparison of laboratory measurements with simulation results investigate the role of the Hall effect in the nonlinear mode coupling.   

Transport and linear stability studies for PPCD optimization in RFPs

We have combined 1D transport simulations of pulsed poloidal current drive (PPCD) together with linear stability studies for a wide spectrum of m=1 and m=0 modes. The model includes a ``dynamo'' term in Ohm's law which is gradually decreased to zero during the early part of the PPCD cycle, simulating the decrease in tearing mode activity as PPCD progresses. We present several initial studies with ad-hoc waveforms in time for the wall electric field, and measure the fraction of time over which m=1, m=0 stability is achieved. We have also developed a more systematic scheme by which we decrease the toroidal electric field at the wall and determine the poloidal electric field there by requiring the parallel electric field to be zero there. This programming is designed to match at a specific time a self-similar rampdown (SSRD) state. After this time the wall electric field components are then programmed to decay exponentially. A PPCD scenario is considered optimal if it has a slow decay rate and develops into a SSRD state well inside the stable SSRD regime.   

Tearing Instabilities in the Reversed-Field Pinch in the Presence of Hall Currents and Pressure Gradients

We present simulations of the linear and nonlinear behavior of tearing modes in reversed-field pinch equilibria. In particular, we perform parameter studies, varying the strength of the Hall term and the plasma beta, and investigate the role of the pressure gradient and diamagnetic stabilization. Simulations are performed using our Magnetic Reconnection Code (MRC), a fully 3D extended MHD simulation code which includes Hall current and electron pressure gradient in a generalized Ohm's law. The MRC is an MPI-parallelized finite-volume based simulation code that integrates the extended MHD equations. It supports arbitrary curvilinear coordinate mappings, allowing it to be adapted to cylindrical and toroidal geometries. In order to overcome restrictive time-step limits, it uses implicit time integration. We have carried out comparisons of linear resistive MHD stability results with NIMROD, showing excellent agreement. In the presence of diamagnetic effects, the nonlinear tearing modes exhibit vortical structures and flow-through reconnection. Implications for observations in MST on sawtooth crashes and the dynamo effect will be discussed.   

Application of the Generalized Weighted Residual Method to stability problems within ideal and resistive MHD

Initial-value stability and transport problems formulated in resistive MHD usually require extensive computations using a very large number of time steps. Although spectral methods are used for the spatial domains, finite steps are traditionally used for the temporal domain with resulting constraints in terms of CFL-like stability conditions for explicit and accuracy-related issues for implicit methods. The Generalized Weighted Residual Method (GWRM) alleviates these problems by representing the time domain in the form of a Chebyshev series. The solution is obtained as an approximate semi-analytical expression through solving a global system of algebraic equations for the expansion coefficients, valid for all time, spatial and physical parameter domains. We demonstrate solutions in terms of eigenvalues and eigenfunctions for the z-pinch, using the linearized ideal MHD equations. Including resistivity, results for resistive g-modes of the reversed-field pinch are also presented.   

Overview of Recent Results and Future Plans of RFX-MOD

RFX-mod is a flexible RFP (R=2m, a=0.46m) device designed for operation up to 2MA, to (a) explore RFP approach to fusion (b) provide state-of-the-art contribution to stability feedback control (c) focus on 3D magnetic shaping. Operation up to 1.9 MA was achieved. Single helical axis equilibria with strong electron internal transport barriers [Nat. Phys. 5, 2009] lead to $T_{e,core} \approx$ 1.5 keV. eITBs develop in regions with low/null magnetic shear and appear at $n/n_{Greenwald}$~0.2; this operational limit is due to $n_{edge}$ accumulation. A combination of H-pellet fuelling and lithization is explored to reach better density control and more peaked $n_e$ profiles. Gyrokinetic calculations show that microtearings may drive transport across eITB. Full 3D equilibrium is reconstructed with VMEC, adapted to the RFP and allowing the application of other codes originally developed for stellarators. MHD feedback control gives new results, also in low-current RFX-tokamak plasmas, where a (2,1) current driven is actively stabilized, reinforcing the collaboration with tokamak community on MHD active control.   

Impurity transport in enhanced confinement regimes in RFX-mod Reversed Field Pinch

The results of impurity transport studies in RFX-mod enhanced confinement quasi-single helicity (QSH) and single helical axis (SHAx) regimes are presented and discussed. The impurity diffusion coefficient and pinch velocity are obtained through comparing experimental emission pattern (line emission and SXR time evolutions, SXR profiles) with the results of a 1-D impurity transport code. Previous analysis [S. Menmuir et al. to be published in Plasma Phys. Contr. Fus.] of impurity transport in RFX-mod standard discharges showed that the impurity pinch velocity, always directed outwards, features a barrier with high values around r/a = 0.8, where the diffusion coefficient decreases by one order of magnitude. In the QSH regime, the transition region in D and v is more internal and the barrier in velocity is wider and stronger. New results have been obtained in experiments with Ni laser blow-off (LBO) injection in high current discharges (Ip$>$1.5 MA) with long lasting QSH, to better characterize the Ni behavior inside the helical magnetic topology.   

Neoclassical transport in the helical Reversed-field pinch

Test particle evaluation of the diffusion coefficient in a fusion plasma in the reversed-field pinch (RFP) configuration shows distinct similarities with Stellarators when the plasma spontaneously evolves towards a helical shape with reduced magnetic chaos. In particular, we recover the classical Tokamak and Stellarator transition from the banana to the plateau and Pfirsch-Schl\"uter regimes. The almost total absence of helically trapped (``superbanana'') particles with the values of $q$ typical of the RFP ($|q| < 0.16$) and at the levels of helical deformation seen in experiment ($B_h/B = 10$\%) causes transport to be proportional to collision frequency (at low collisions). This fact excludes the possibility that the minimum conceivable transport could be inversely proportional to collision frequency, which is typical of un-optimized Stellarators. This result strengthens the perspectives of the helical RFP as a fusion configuration.   

Trapped electrons and microinstabilities in the reversed-field pinch

We study the role of trapped electrons in the anomalous transport induced by microinstabilities in the Reversed Field Pinch. This investigation is justified by the fraction of trapped particles present in the core, which is almost the same as in the tokamak.\footnote{M. Gobbin {\em et al.}, J. Plasma Fusion Res. Series {\bf 8}, 1147 (2009)} This fraction turns out not to have a strong influence on ITG mode frequency/growth rate. Trapped electron modes are revealed especially in correspondence to relatively large density gradients and slight electron temperature slopes. Analytical approaches are carried out with fluid/kinetic ions. Numerical results are obtained with the codes GS2\footnote{M. Kotschenreuther {\em et al.}, Comput. Phys. Commun. {\bf 88}, 128 (1995); W. Dorland {\em et al.}, Phys. Rev. Lett., {\bf 85}, 5579 (2000)} and TRB.\footnote{X. Garbet {\em et al.}, Phys. Rev. Lett., {\bf 91}, 035001 (2003)} Some recent advances on ITG and microtearing mode studies are also summarized.   

Application of VMEC to RFX Equilibria and Data Analysis

Recent experimental data in RFX (Reversed Field Experiment) indicate the formation of long-lived helical structures. The modification of VMEC (Variational Moments Equilibrium Code) to compute these three-dimensional equilibria is described. Previous applications of VMEC relied on the presence of a strong (non-reversing) background toroidal magnetic field. The reversal point in RFX, where the toroidal field changes sign inside the plasma, requires a new internal magnetic field representation in VMEC. A rich suite of codes for computing transport (DKES) and stability (COBRA and TERPSICHORE) has been developed around VMEC and can now be applied to RFX as well. VMEC equilibria for RFX are compared with ones computed with the SHEq code, which is based on a perturbative approach.   

Soft-X ray imaging diagnostics of helical structures in the low-aspect-ratio RFP RELAX

Soft-X ray (SXR) imaging diagnostic techniques using tangential pinhole camera have been developed and applied to the RELAX ($R$/$a$=0.5m/0.25m, $A$=2) for the study of three-dimensional (3-D) magnetic structures in a low-aspect-ratio RFP configuration. An experimental 2-D image has been compared with calculated tangential images using model profiles for emissivity, to identify the most plausible SXR emissivity profile for the experimental 2-D image. It has been found that in shallow-reversal RFP plasmas with relatively high fill pressure, a helically deformed core has been formed in RELAX. The SXR emissivity in the helical core is estimated to be 2-3 times higher than in the background. According to time variation measured by AXUV array, the SXR emissivity is the highest in the core region with quasi-periodic oscillation of the peak position, which is an indication of toroidal rotation of the helical hot core.   

Turbulent Transport of Fast Ions in the Large Plasma Device (LAPD)

Due to gyroradius averaging and drift-orbit averaging, the transport of fast ions by microturbulence is often smaller than for thermal ions. In this experiment, Strong drift wave turbulence is observed in LAPD on gradients produced by a plate obstacle. Energetic lithium ions orbit through the turbulent region. Scans with a collimated analyzer and with probes give detailed profiles of the fast ion spatial distribution and of the fluctuating fields. The fast-ion transport decreases rapidly with increasing fast-ion gyroradius. Unlike the diffusive transport caused by Coulomb collisions, in this case the turbulent transport is non-diffusive. Analysis and simulation suggest that the fast ions interact ballistically with stationary two-dimensional electrostatic turbulence. The energy dependence of the transport is well explained by gyro-averaging theory. In new experiments, different sources and obstacles alter the drift-wave turbulence to modify the nature of the transport.   

Hybrid Modeling of Alfv\'{e}n Wave Propagation in a Helicon Plasma Source

A 2D hybrid fluid-PIC code is used to study the propagation of Alfv\'{e}n waves from the low density edge to the high density central axis of a helicon plasma source. We will present the results from simulations that include reflecting boundary conditions and realistic plasma parameters. Waves are launched both internally and externally. For the internal excitation case, waves are launched from a ring antenna immersed in the plasma. The simulation results are compared to recent experiments in which internally launched waves were ducted along the high-density plasma core and mode converted from compressional waves into shear waves. In the case of externally launched waves, an {\em m}=1 coil placed at the plasma edge is used to excite shear waves in the edge. We will present simulation results demonstrating wave reflection and coupling to the high density plasma core.   

Neutral Density Profile Measurement in a Helium Helicon Plasma and its Effect on Alfv\'{e}n Waves Dispersion

Recently we developed a novel technique to measure the neutral density in a helium plasma using laser induced fluorescence and the plasma opacity [\textit{Houshmandyar et al.}, Rev. Sci. Instrum. Oct 2010]. Knowledge of neutral density profile provides insight into the propagation of shear Alfv\'{e}n waves in partially ionized plasmas; in which finite ion-neutral collisions radically alter the wave characteristics, $i.e.$, the dispersion relation. Here we report an extension to our earlier work, including measurements of the neutral density at different radial positions for two gas flow configurations, continuous inlet flow and static (no flow). Also, we show that the measured values of neutral densities at different radii are in good agreement with of those deduced from the kinetic Alfv\'{e}n dispersion relation when measured at different radii.   

Ion acceleration in Ar-Xe and Ar-He plasmas

Ion velocity distribution functions (ivdf) are investigated by laser induced fluorescence in Ar-Xe and Ar-He expanding helicon plasmas as a function of gas composition. In the case of Ar-Xe plasma it was found that in the helicon source both Ar+ and Xe+ vdfs are unimodal. Their parallel speeds are subsonic and unaffected by changes in gas composition. At the end of the source the argon ivdf shows a bimodal structure indicative of an electric double layer upstream of the measurement location. The fast argon ion component parallel velocity increases with Xe fraction from 6.7 to 8 km/s as the Xe fraction increases from 0 to 4{\%}. In the expansion region, the bimodal character of Ar ivdf is maintained with a supersonic fast component reaching parallel speeds of 10.5 km/s. For all studied plasma conditions and different spatial locations, the Xe+ vdf exhibits a unimodal structure with a maximum parallel flow velocity of 2.2 km/s at the end of the source. For Ar-He plasma, the Ar ivdf is bimodal with the fast ion component parallel velocity increasing from 5.2 to 7.8 km/s as the He fraction increases from 0 to 30{\%}. For the same He fraction range, the slow argon ion population distribution changes from a single Gaussian to a wide distribution extending all the way from the speed of the fast population to 0 m/s.   

Simultaneous two-dimensional laser induced fluorescence measurements in a helicon plasma

Recent upgrades to the laser induced fluorescence system at West Virginia University have made possible simultaneous two-dimensional ion velocity distribution function measurements in the Hot hELIcon eXperiment. A single dye laser is split into two beams, each modulated at a different frequency using a mechanical chopper. The two beams are then injected into the plasma chamber with one parallel and one perpendicular to the background magnetic field. Fluorescence signal from each beam is then collected simultaneously at single location using a single set of collection optics. We will present measurements of the thermal anisotropy as a function of the plasma radius obtained with a two-dimensional scanning mount.   

Studies of Electrostatic Instabilities During Double Layer Formation Using Time-Resolved LIF

Previous time resolved laser induced fluorescence (LIF) measurements of the parallel ion velocity distribution function (ivdf) in expanding, helicon plasma demonstrated a possible correlation between beam and bulk ion populations and electrostatic fluctuations. For a strong mirror ratio, where the formation and collapse of a beam population is observed, the correlation between fluctuations in the ion populations and the electrostatic fluctuations are strong. For a weak mirror ratio, the beam is less correlated with the electrostatic fluctuations than the bulk ion population and the beam persists throughout the discharge. We present results from detailed study of the effects of varying the mirror ratio and rf frequency in the source.   

A comparison of laser induced florescence and continuous wave ring down spectroscopy IVDF measurements in an argon helicon plasma

In this work, we compare two spectroscopic methods for measuring the ion velocity distribution functions (IVDF) in an argon helicon plasma: laser induced florescence (LIF) and continuous wave cavity ring down spectroscopy (CW-CRDS). An established and powerful technique, LIF suffers from a requirement that the initial state of the LIF sequence have a substantial density. In most cases, this requirement limits LIF to ions and atoms with large metastable state densities for the given plasma conditions. CW-CRDS is considerably more sensitive than LIF and can potentially be applied to much lower density populations of ion and atom states. CRDS is a line integrated technique without the spatial resolution of LIF. CRDS is a proven, ultra-sensitive, cavity enhanced absorption spectroscopy technique and when combined with a CW diode laser that has a sufficiently narrow linewidth, the Doppler broadened absorption line, i.e., the IVDF, can be measured. We will present CW-CRDS and LIF measurements of the IVDF in argon using the 668.614 nm (in vacuum) line of Ar II.   

Continuous wave cavity ring down spectroscopy measurements of ion velocity distribution functions in argon helicon plasma

The West Virginia University helicon source group routinely employs laser induced fluorescence (LIF) to measure velocity distribution functions (VDFs) of argon ions, argon neutrals, helium neutrals and xenon ions. We are developing a continuous wave cavity ring down spectroscopy (CW-CRDS) diagnostic with a narrow linewidth, tunable diode laser as an alternative technique to measure VDFs in species where LIF is inapplicable. Being an ultra-sensitive, cavity enhanced absorption spectroscopy technique, CRDS can also provide a direct measurement of the absolute metastable state density. Here we present Ar II ion VDFs obtained through measurements of the Doppler broadened absorption spectrum of Ar II at 668.614 nm (in vacuum), a standard initial state for conventional Ar II LIF.   

Alfv\'{e}n Wave Heating of Argon Ions in the Hot hELicon eXperiment (HELIX) at West Virginia University

Alfv\'{e}n wave damping is the dominant physical process invoked in leading theoretical models of ion heating in the solar corona.~ The construction of a new external antenna by the West Virginia University helicon source group to launch large-amplitude (B$_{1}$~ $\sim $ 10{\%} of B$_{0})$ shear Alfv\'{e}n waves in argon plasma provides a new experimental tool to investigate ion heating due to the damping of these waves.~ The ion temperatures are measured with Laser Induced Fluorescence (LIF) while magnetic sense coils are used to measure the phase velocity and amplitudes of the propagating waves. Here we present completed design schematics for the antenna as well as analysis of the launched waves and resonant ion heating results.   

Static and Dynamic Structure Factors with Account of the Ion Structure for High-temperature Alkali and Alkaline Earth Plasmas

The electron-electron, electron-ion, ion-ion and charge-charge static structure factors (SSF) are calculated for alkali and Be$^{2+}$ plasmas at various temperatures and concentrations using the method described by G. Gregori et al., Phys. Rev. E \textbf{74}, 026402 (2006); High Energy Density Phys. \textbf{3}, 99 (2007). The dynamic structure factors (DSF) for alkali plasmas are calculated using the method of moments developed by V. M. Adamjan et al., High. Temp. \textbf{21}, 307 (1983). In both methods the screened Hellmann-Gurskii-Krasko potential, obtained on the basis of Bogolyubov's method, has been used taking into account not only the quantum-mechanical effects but also reflects important features of the ion structure (S. Sadykova et al., Contrib. Plasma Phys. \textbf{49}, 76 (2009)). Our results on the SSFs for Be$^{2+}$ plasma deviate from the data obtained by Gregori et al., while our DSFs are in a reasonable agreement with those of S. V. Adamjan et al., Phys. Rev. E \textbf{48}, 2067 (1993). We conclude that the short range forces, which we take into account by means of the HGK model potential, which deviates from the Coulomb and Deutsch ones, employed by S. V. Adamjan et al. and Gregori et al. correspondingly, influence the SSFs and DSFs significantly.   

Betatron radiation from an off-axis electron bunch in a PWFA

In a plasma wakefield accelerator (PWFA) with a drive bunch density higher than the plasma, a pure ion column is formed behind the drive bunch (blow-out regime). Due to the ion restoring force, which is linearly increasing with radius, beam electrons perform betatron oscillations. We consider the case of a witness bunch entering the plasma with a radial offset or a transverse momentum component. In this case, the whole witness bunch oscillates about the beam axis defined by the drive bunch. We use the particle in cell code QUICKPIC [1] to simulate the plasma wakefields and we study the radiation characteristics as a function of the electron bunches and plasma parameters. We place the witness bunch at the position where the synchrotorn- radiated power is compensated for by the energy gain from the wakefields. Detailed results will be presented.\\[4pt] [1] C.H. Huang, et al., J. Comp.Phys., 217(2), 658, (2006).   

Numerical study of plasma wakefields driven by trains of electron bunches

We study numerically the excitation of plasma wakefields by a train of electron bunches. The purpose is to find a regime in which the wakefield excited by individual electron bunches add and have a large amplitude and a large transformer ratio. This scheme will produce a high energy accelerated bunch with a low energy drive train in a single plasma wakefield accelerator stage. The transverse size of the bunches must be maintained long enough for the driving bunches to efficiently transfer their energy to a trailing witness bunch. It is studied experimentally at the Brookhaven National Laboratory Accelerator Test Facility (ATF). We also investigate the effect of a transverse electron plasma profile on the period of the excited wakefield, an effect that may play a role in the experiments using a capillary discharge as a plasma source. Detailed simulation results will be presented.   

Adaptable Particle-in-Cell Algorithms for Graphical Processing Units

Emerging computer architectures consist of an increasing number of shared memory computing cores in a chip, often with vector (SIMD) co-processors. Future exascale high performance systems will consist of a hierarchy of such nodes, which will require different algorithms at different levels. Since no one knows exactly how the future will evolve, we have begun development of an adaptable Particle-in-Cell (PIC) code, whose parameters can match different hardware configurations. The data structures reflect three levels of parallelism, contiguous vectors and non-contiguous blocks of vectors, which can share memory, and groups of blocks which do not. Particles are kept ordered at each time step, and the size of a sorting cell is an adjustable parameter. We have implemented a simple 2D electrostatic skeleton code whose inner loop (containing 6 subroutines) runs entirely on the NVIDIA Tesla C1060. We obtained speedups of about 16-25 compared to a 2.66 GHz Intel i7 (Nehalem), depending on the plasma temperature, with an asymptotic limit of 40 for a frozen plasma. We expect speedups of about 70 for an 2D electromagnetic code and about 100 for a 3D electromagnetic code, which have higher computational intensities (more flops/memory access).   

Whistler Wave Resonances in Laboratory Plasma

Standing whistler wave patterns have been investigated in the Naval Research Laboratory's Space Physics Simulation Chamber. In the original experimental configuration, partial reflection of the antenna-launched whistler waves from the chamber end boundaries occurs, setting up a combination of standing and traveling waves. By controlling the axial magnetic field strength profile, cyclotron absorption of the whistler waves can be induced before reflection occurs, leaving only the forward propagating waves. By comparing standing wave amplitudes to that when the wave is prevented from reflecting, cavity Q's in excess of 30 have been observed. Under uniform axial magnetic field conditions, the addition of planar conducting grids across the vacuum chamber cross-section at the ends of the plasma column provides improved reflecting surfaces and corresponding increases in the value of Q.   

Experiments correlating antenna impedance and wave propagation in the NRL Space Physics Simulation Chamber

We have demonstrated in previous work that an electrically short exciter in a plasma will have the largest energy deposition around the plasma-sheath resonant frequency which is about half the electron plasma frequency $\omega_{pe}$. When the exciter is electrically long, there is the possibility of competing channels of absorption from resonances associated with antenna radiation resistance. This brings up questions when operating in a parameter space where both energy deposition mechanisms are present: a) If both plasma-sheath and resonant wavelength coupling are present, can we control which is dominant, and b) is it advantageous to do so if the goal is to drive large amplitude waves? Presented are results of experiments controlling the antenna plasma sheath and electrical length to overlap the two modes of coupling. Concurrently we can monitor the propagated wave amplitude as we attempt to preference one mode versus the other. Results are expected to shed light on questions of antenna design and its relation to power deposition.   

Network analyzer-based plasma diagnostics using an rf impedance probe

We have recently completed an extended experimental series demonstrating the usefulness of a network analyzer in plasma diagnostics in the thin sheath limit.\footnote{In Review, Phys. Plasmas (July 2010)} An rf signal, much smaller than probe dc bias voltages, is applied to a spherical probe by the analyzer which returns both real and imaginary parts of the complex plasma impedance as a function of frequency.\footnote{Phys. Plasmas 13,032108 (2006)} Plasma impedance is determined by comparing the incident signal to that reflected from the plasma. We show that this information can be used to determine plasma potential and the electron distribution function. We will present data showing results for three spherical probes in addition to preliminary observations using cylindrical probes. The analysis method we present has general application to diverse areas of plasma investigations in the laboratory or in space. It can be used with \textit{in situ} instrumentation and can be extended to provide an estimate of sheath structure about the probe.   

The effect of dirty walls on the plasma potential in a multi-dipole chamber

In a multi-dipole chamber with dirty walls, the plasma potential is observed to be negative with respect to the grounded wall in the tens of volts. The plasma is generated by hot filaments releasing monoenergetic ionizing electrons that can be chosen to be from 35 to 60eV. Measurements with a collecting Langmuir probe suggest that the electron energy distribution is bi-Maxwellian e.g. Te = 2 and 4 eV. It is observed that the bulk plasma potential becomes more negative with increasing relative concentrations of hotter electrons. The potential profile next to the wall was measured using an emissive probe in the limit of zero emission. A virtual cathode forms approximately 8mm from the wall under a wide range of discharge parameters to confine secondary electrons caused by the ionizing electrons hitting the dirt.   

LVPD Plasma for Studies on ETG Turbulence

Role of electron temperature gradient (ETG) in plasma turbulence in tokamaks has been recognized in various experimental and theoretical investigations on plasma transport. However, investigations of ETG turbulence in tokamak devices have been based upon indirect inferences drawn from experimental database. Nor has it been experimentally investigated in basic plasma devices as it is very difficult to produce electron temperature gradient. In this paper, we show that this can be secured by making use of a magnetic electron energy filter. The plasma in Large Volume Plasma Device is characterized by plasma density, n$_{e0 }\sim $ 3$\times $10$^{11}$ cm$^{-3}$ and T$_{e0} \quad \sim $ 3 eV. In this device, we have provided a magnetic filter consisting of 155- turns rectangular coil of 4 cm width and varying length of $\sim $ 192 to15 cm from centre to the edge. The field at the centre of the coil is $\le $ 150 G, leaving a residual magnetic field of $\le $ 1 G in the region of experimental investigations. Our results show that $\nabla $T$_{e}$ in the flat electron density region can be secured by providing radial inhomogenity in filter magnetic field. Scale length of gradient can be varied from 200 cm to 40 cm in a controlled manner. This has allowed us to undertake detailed investigations on ETG turbulence.   

Magnetic field generation via the Kelvin-Helmholtz instability

Collisionless plasma instabilities have been proposed as candidates to explain the origin of magnetic fields required by models for non-thermal radiation emission in GRBs. Since these extreme scenarios are usually associated with strain and rapid variability of the ejecta, it is likely that strong velocity shears are present, triggering the collisionless Kelvin-Helmholtz instability (KHI). Seed magnetic fields, generated at the early development of the KHI, can be amplified by the dynamo effect in KHI-induced turbulence. In this work we generalize the relativistic collisionless KHI calculations to include arbitrary density jumps/flows. We observe that the onset of the KHI is robust to density jumps, making this instability ubiquitous in astrophysical scenarios. We present the first fully kinetic 3D simulation of a KHI scenario, which reveals the transverse dynamics of the KHI. Both structure formation and underlying physical processes are discussed. We also present a detailed comparison between the KHI in scenarios with electron-positron and with electron-proton clouds.   

Improved Matching and Plasma Formation Using Frequency Tuning During RF Pulsing in a Helicon Source

A flowing argon helicon plasma is formed in a 10 cm diameter, 1.5 m long Pyrex chamber with an axial magnetic field in nozzle or flat configuration, variable up to 1 kG in the source region. RF power is fed from a 1 kW tube-based supply into a capacitive matchbox that is tuned for low reflected power ($<$ 5\%) during steady-state helicon operation. A 18 cm long, 12 cm diameter half-turn double-helix antenna is used to excite helicon waves in the source. During pulsed operation, a high ($ \sim 10^{14}$ cm$^{-3}$) transient electron density is observed that leads to a poor RF match during the transient. Calculated variation of the RF frequency (from 12 MHz to 15 MHz) during the pulse allows for low reflected powers during the gas breakdown and the approach to and formation of the steady state plasma. Microwave interferometry (105 GHz), collisional radiative spectroscopic codes and diamagnetic loops are used to measure electron density and temperature during pulsed (5 ms) RF operation. Potential other circuit schemes are investigated that offer a larger matching bandwidth or an increased range of ``matchable'' plasma/antenna impedances.   

Drift Trajectories In Tokamaks With Radial Electric Fields

An interesting approach to achieving steady state tokamak operation is using radial electric fields to provide some, or all of the rotational transform. We have classified the trajectories of individual, collision free particles moving in fields relevant to this approach in terms of the exact invariants $\mathcal{H}$, $p_\phi$, and the adiabatic invariant $\mu$. In addition, by employing classical averaging techniques we have derived a differential equation in two variables, the minor radius $r$ and poloidal angle $\theta$, that determines the time dependence of such trajectories.   

A transformation identifying the Caldeira-Leggett model with the linear Vlasov-Poisson system

The Caldeira-Leggett model is a Hamiltonian system that describes a simple harmonic oscillator coupled to a continuous spectrum of simple harmonic oscillators. It was invented to study quantum tunnelling in dissipative systems. We show that the damping mechanism in the Caldeira-Leggett model is analogous to Landau damping and derive an integral transformation from solutions of the Caldeira-Leggett model to solutions of the linearized Vlasov-Poisson equation. This establishes the equivalence of the two models and provides an example of a transformation that diagonalizes a Hamiltonian systems with a continuous spectrum\footnote{P.~J.~Morrison. Hamiltonian description of Vlasov dynamics: Action-angle variables for the continuous spectrum. Trans.\ Theory and Stat.\ Phys., 29:397-414, 2000.}. We let the discrete oscillator have negative energy and derive an analog of the Penrose criterion for the stability of solutions. We extend our proof of Krein's theorem for the linearized Vlasov-Poisson equation\footnote{G.~I.~Hagstrom and P.~J.~Morrison. On Krein-like theorems for noncanonical Hamiltonian systems with continuous spectra: application to Vlasov-Poisson. arXiv:1002.1039. To appear in Trans.\ Theory and Stat.\ Phys.} to the Caldeira-Leggett model.   

Parametric study of gravito-MHD waves

Quasi-periodic oscillations (QPO) are commonly observed in the gamma-ray flares by magnetars, the neutron stars with internal magnetic field as high as $10^{16}$ G. The QPOs can have very low frequencies, e.g. tens of Hertz. To understand the source of QPO and the damping mechanism in such frequency range, it is of interest to examine the characteristics of gravito-MHD waves in such a plasma where sound wave is close to the light speed and hence far greater than the Alfven speed. To this end, we consider a planar geometry and study the dispersion relation of the system for arbitrary angles between the equilibrium magnetic field and gravity and by changing the relative strength of the sound and Alfven speeds. Gravity does not change the number of characteristic MHD waves. Furthermore, in the limit when the sound speed of the system is much larger than the Alfven speed, the system supports two low-frequency waves (with phase velocity proportional to the Alfven speed) and one high-frequency wave (with phase velocity proportional to the sound speed). The behavior of such waves in a spherical geometry, especially the emergence of discrete modes as opposed to a continuum, will be explored. Work supported by LANL LDRD.   

Scattering of magnetic mirror trapped electrons by an Alfven wave

Energetic electrons produced naturally or artificially can be trapped in earth's magnetic field for months, threatening the growing population of space satellites. An experimental study of artificially de-trapping these particles is performed on Large Plasma Device (LaPD) at UCLA (the quiescent afterglow plasma in which the experiment was done had :$n\approx 5\times 10^{11}cm^{-3},B=860G,T_e =0.5eV,L=18m,diameter=60cm)$, matching critical parameter ratios in the lab plasma to those in space. In this experiment, electrons with large $v_\bot $ are produced by microwave heating (2.45GHz, 5kW) at upper-hybrid frequency, and trapped by a magnetic mirror field (mirror ratio $\approx $1.5). A shear Alfven wave (f=220kHz,$B_{wave} =2G)$ is launched with a rotating magnetic field source. It is observed that the wave eliminates the trapped electrons. This effect is observed via different diagnostics. Plasma density and temperature perturbations from the Alfven wave are observed along with the scattering. The scattering mechanism is under investigation.   

Pitch Angle Scattering of Electrons by Alfven Waves Generated with Rotating Magnetic Field Source

The pitch angle scattering by large amplitude whistler and Alfven waves is an attractive mechanism for the precipitation of electrons in the radiation belt. Recent experiments in LAPD/UCLA have shown rapid loss of energetic electrons in the presence of waves and this is studied in detail using simulations with experimental parameters. The Alfven waves generated by a Rotating Magnetic Field (RMF) antenna in LAPD have sharp gradients in the transverse direction. These rapidly varying magnetic fields leads to the breaking of the adiabatic invariant of electrons and precipitate them to the loss cone via non-resonant scattering. The generation of Alfven waves by the RMF antenna is simulated with a two fluid code and the resulting fields are used in a particle tracing code to study the pitch angle scattering of electrons. It is found that the pitch angle diffusion coefficient for the non-resonant scattering scales as the square of the ratio of the electron Larmor radius to the transverse wavelength. Further, the fluctuations generated by the two loop RMF source fills the plasma volume more effectively and is more effective in pitch angle scattering the electrons than using one loop antenna Work supported by ONR MURI grant.   

Alfven Wave Generation by a Rotating Magnetic Field Source: Theory, Modeling and Experimental Results

Recent experiments conducted in the Large Plasma Device (LAPD) located at UCLA demonstrated efficient excitation of whistler and shear Alfven waves by a Rotating Magnetic Field (RMF) source. We present analytical theory, computational modeling and experimental results of the shear Alfven wave excitation by RMF source created by a phased orthogonal two-loop antenna in a plasma. An analytical theory and simulations using a three-dimensional cold two-fluid model of Alfven wave excitation were developed and compared with experiments. These comparisons show good agreement on linear shear Alfven wave properties, namely, spatio-temporal wave structure, dispersion relation, and the dependence of wave magnitude on the wave frequency. From the simulations it was found that the energy of the Alfven wave generated by the rotating magnetic field source is distributed among the kinetic energies of ions and electrons and the electromagnetic energy of the wave. The wave magnetic field power calculated from the experimental data and using a fluid model agrees within $\sim$1 percent. The RMF source is thus very efficient in generating shear Alfven waves. Work supported by ONR MURI grant.   

Design and Use of an Els\"{a}sser Probe for Projection of Alfv\'{e}n Wave Fields According to Wave Direction

Measurement of plasma transport using probes usually requires simultaneous measurement of multiple quantities from which transport can be inferred. Particle and energy transport have received the most attention. We have designed and built a new probe to simultaneously measure fluctuating $E$ and $B$ fields in order to evaluate wave Poynting flux for application to Alfv\'{e}n wave experiments in the Large Plasma Device (LAPD) at UCLA. This new probe allows projection of measured wave fields onto Els\"{a}sser variables $Z^{\pm }$\equiv (\textit{E $\times $ B}$_{0})$/$\vert B_{0}\vert ^{2}\pm $ $B/$(4\textit{$\rho $}$_{0}$\textit{$\pi $})$^{0.5}$ where the time averaged background field $B_{0}$ and plasma mass density \textit{$\rho $}$_{0}$ are measured separately. Experiments were conducted in a singly ionized He-H plasma in the LAPD and these measurements are presented. The results were compared with existing measurement techniques for this type of plasma in the LAPD [1]. Findings will be discussed at the conference.\\[4pt] [1] C. A. Kletzing, D. J. Thuecks, F. Skiff, S. R. Bounds, and S. Vincena, Phys. Rev. Lett. \textbf{104}, 095001 (2010).   

Ion-ion Hybrid Alfven Wave Resonator

In a magnetized plasma consisting of two ion species, the perpendicular dielectric coefficient vanishes at the ion-ion hybrid frequency, where shear Alfven waves have zero parallel group velocity and experience a cut-off. Since the ion-ion hybrid frequency is proportional to the magnetic field, it is possible for propagating shear waves to be reflected in regions of increasing magnetic field. Thus, in principle, it is possible for a magnetic well configuration in a two ion plasma to behave as an Alfven wave resonator, as may be encountered in a tokamak or a planetary magnetosphere. This study explores the possibility of establishing such a resonator in the linear plasma column generated in the Large Plasma Device (LAPD) at UCLA using H-He and He-Ne mixtures. The resonator response is investigated by launching monochromatic waves or sharp tone bursts from a rectangular magnetic loop antenna. In a magnetic well, the radial profiles of the wave magnetic field broaden and a line structure emerges in the power spectrum. The profiles of the lines in the spectra resemble global radial modes rather than waves launched from a localized source. The observations are compared to predictions of a square wave model of the well and the Budden transmission model.   

Laser produced plasma clouds as a source and obstacle for multi-ion Alfven wave propagation

We present results on the interaction of Alfven waves and energetic, laser-produced plasmas (LPP's). The experiment consists of a large, ambient, magnetized plasma, within which the carbon LPP is created. The LPP is generated using a turn-key Nd:YAG (1064nm, 1J, 10ns) laser. The background plasma is generated by the Large Plasma Device (LAPD) at UCLA. The background species is helium, $n=10^{12}$ cm$^{-3}$, D=60cm, L=1660cm cylinder. The LPP acts as a source of energetic carbon ions: 10$^{15}$ particles, v=$10^{7}$ cm/s = 0.1$v_{A}$ whose expansion is directed primarily along the background magnetic field; the resulting cloud is of interest for two reasons: (1) the carbon cyclotron motions act as an electric dipole antenna, which radiates shear Alfven waves in the He plasma, which outrun the cloud expansion. (2) the relatively slow expansion of the LPP compared to the Alfven speed produces a spatially localized region of a two-ion species plasma which can act as a barrier for externally launched shear Alfven waves due to the ion-ion hybrid resonance layer in the cloud. Wave field measurements will be presented of both outgoing and incoming shear Alfven waves to the carbon cloud. Comparisons of the cyclotron emission are made to predictions by the OSIRIS PIC code.   

Contained Alfven eigenmodes in mirrors with sheared rotation

In rotating mirrors, a fixed azimuthal perturbation in the lab frame appears as a wave in the rotating frame. If there is sheared rotation, the plasma-frame frequency will also vary radially due to the Doppler shift. This can lead to radially localized Alfven eigenmodes with high azimuthal mode numbers. Such contained Alfven modes are found both for peaked and non-peaked rotation profiles. These modes might be useful for alpha channeling or ion heating, as the high azimuthal wave number allows waves with zero frequency in the lab frame to exceed the ion cyclotron frequency in the rotating frame for reasonable values of the rotation frequency.   

Revisiting linear plasma waves for finite value of the plasma parameter

We investigate through theory and PIC simulations the Landau-damping of plasma waves with finite plasma parameter. We concentrate on the linear regime, $\gamma \gg \omega_B$, where the waves are typically small and below the thermal noise. We simulate these condition using 1,2,3D electrostatic PIC codes (BEPS), noting that modern computers now allow us to simulate cases where ($n\lambda_D^3 = [1e2;1e6]$). We study these waves by using a subtraction technique in which two simulations are carried out. In the first, a small wave is initialized or driven, in the second no wave is excited. The results are subtracted to provide a clean signal that can be studied. As $n\lambda_D^3$ is decreased, the number of resonant electrons can be small for linear waves. We show how the damping changes as a result of having few resonant particles. We also find that for small $n\lambda_D^3$ fluctuations can cause the electrons to undergo collisions that eventually destroy the initial wave. A quantity of interest is the the life time of a particular mode which depends on the plasma parameter and the wave number. The life time is estimated and then compared with the numerical results. A surprising result is that even for large values of $n\lambda_D^3$ some non-Vlasov discreteness effects appear to be important.   

Streaming Ultracold Neutral Plasmas

Ultracold Neutral Plasmas (UNPs) are formed by near-threshold photoionization of laser-cooled atoms. They are orders of magnitude colder than other neutral plasmas and have extremely clean and controllable initial conditions. With appropriate masking of the ionization laser, we modify the initial density distribution to create two hemispheres of plasma that stream into each other during expansion. We will discuss the interaction of these streaming plasmas, their collisionality, penetration, and stopping power. This new technique of shaping the initial density enables the study of various collective modes, plasma collisions and the effects of strong coupling.   

Ion Acoustic Waves in Ultracold Neutral Plasmas

Ultracold neutral plasmas (UNP), created by photoionization of laser-cooled atoms near ionization threshold, are orders of magnitude colder than any other neutral plasma. They have extremely clean and controllable initial conditions that enable the study of strongly coupled physics, plasma expansion and collective wave phenomena. Here, we excite ion acoustic waves in UNPs through direct imprinting of ion density modulations during plasma formation, where the density modulations are implemented by applying a mask to the ionizing laser. Laser-induced fluorescence imaging of the plasma reveals density perturbations that oscillate in space and time. In spite of the UNP's finite size, expansion, and inhomogeneous density; the dispersion relation of these oscillations matches that of ion acoustic waves.   

Investigations into beam-plasma interactions

The interactions of plasmas with non-thermal electron populations can result in instabilities that are of importance in a number of applications and phenomena. Such instabilities include the anomalous Doppler instability that may occur in magnetic confinement fusion, the two-stream instability which is of importance to fast ignition inertial confinement fusion and cyclotron maser emission which can be found in auroral kilometric radiation and chorus. Numerical simulations of these instabilities, particularly the anomalous Doppler and two-stream instabilities, have been undertaken to provide a design and benchmark for the controlled, low temperature, low density experimental studies to follow.   

Ionization damping of linear waves in cold plasmas

For an arbitrary linear wave in cold plasma, an analytical model is proposed to describe wave damping caused by above-threshold ionization and recombination. When either of the two processes dominates, the wave energy is also found as a function of the frequency.   

Quasilinear Model for Energetic Particles Interacting with TAE Modes

TAE instabilities are thought to be a major source of Energetic Particle transport which could set limits on operational scenarios, especially for burning plasmas, and causes damage to the first wall. The quasilinear model proposed by Berk et al.\footnote{H. L. Berk et al, \emph{Nucl. Fusion}, 35:1661, 1995.} relies on diffusion mechanisms for particle dynamics to captures the evolution of the energetic particle distribution function and the associated mode amplitude. Using the bump-on-tail as a paradigm, we analyze the dynamics near the resonances for accurate diffusion coefficient representation. We verify the model to get the predicted single mode saturation levels and benchmark the case of multimode overlap against particle codes. Using the TAE mode structures computed by the ideal MHD code NOVA, we generalize this method to relax energetic particles' profiles in the full 3D phase space.   

Gyrokinetic particle simulation of the beta-induced Alfven eigen mode

The beta-induced Alfven eigen mode (BAE) is studied using the global gyrokinetic particle code GTC. In our simulation, BAE is successfully excited by antenna and energetic particle density gradient. Through the antenna frequency scan, we can measure the BAE frequency and damping rate by numerical fitting the saturation amplitude. BAE excitation by energetic particles shows that the BAE propagates in the ion diamagnetic direction and the frequency has a little downshift, which is due to modification of the energetic particles. The frequency and growth rate in gyrokinetic simulation is a little different from drift kinetic simulation, which is expected due to the finite larmor radius effect. We also find that the BAE frequency is related to the wavelength and the plasma beta while the growth rate is sensitive to the energetic particle properties. Benchmarks between GTC and HMGC are also done through initial perturbation, antenna excitation and energetic particle excitation. The simulation results agree with each other very well.   

The interchange and geodesic acoustic type modes in two-fluid theory

Interchange and geodesic acoustic type modes are considered within the unified approach of two-fluid theory. It is assumed that electrons are in the adiabatic, $\omega <>k_\parallel v_{Ti} $. Appropriate moment equations are used for both components taking into account the the magnetic field curvature, which is modeled by the addition of the constant gravity forces, as well as electromagnetic effects. Both electron and ion drift effects are taken into account including the finite Larmor radius (FLR) effects for ions due to gyroviscosity. Within this approach , ion temperature gradient, interchange, and geodesic acoustic modes are recovered. It is shown that geodesic acoustic modes are intrinsically related to interchange modes. Both types of modes occur as a result of the essential balance between the radial diamagnetic and inertial (polarization) currents. It is shown that coupling of drift and geodesic curvature effects leads to the destabilization of geodesic acoustic modes.   

Characteristics of Helicon-Plasma Produced Using a Segmented Multi-Loop Antenna III

Tokai Helicon Device (THD), which has been used to study lower-hybrid cavitons [1], is a helicon-type plasma device which utilizes a specially designed flat segmented multi-loop antenna to excite a helicon wave. The antenna, which is installed just outside a quartz window at the end of the vacuum vessel (20 cm i.d. and 100 cm length), consists of four concentric loops [2]. Each loop is divided into several segments. By changing the electrical connections among the antenna segments, it is possible to excite $m$ = +-1 and +-2 modes as well as $m$ = 0 mode, where $m$ is the azimuthal mode number. The discharge characteristics of the $m$ = 0 excitation cases were previously reported in detail [2]. The discharge characteristics of the $m$ = +-1 and +-2 cases will be presented. Excited wave characteristics of produced plasma will also be discussed. We have been using THD to develop a new plasma acceleration scheme for an advanced concept electric propulsion system. This application will be briefly discussed.\\[4pt] [1] T. Tanikawa \textit{et al}., Bull. Ameri. Phys. Soc. \textbf{54} (15), 125 (2009).\\[0pt] [2] T. Tanikawa \textit{et al}., Bull. Ameri. Phys. Soc. \textbf{51} (7), 164 (2006); T. Tanikawa and S. Shinohara, \textit{ibid}. \textbf{53} (14), 174 (2008).   

Electrostatic turbulence in the low-density plasma column

Electron plasma density fluctuations are observed in plasma when a radial pressure gradient excites drift waves. The linear machine GyM (R=0.125 m, L= 2.11 m, B$<$0.1T), operating at IFP-CNR since 2008, has started experiments aimed at characterizing drift waves excited in its non-uniform magnetized plasma. Two different plasma sources (magnetron 2.45 GHz or hot filament) have been used to sustain plasma with adjustable sections (1.5 cm$<$r$<$10 cm). The diagnostic system is composed by different sets of movable electrostatic probes and by optical emission spectroscopy, dedicated to the electron temperature measurement. Fluctuations in plasma density have been observed and characterized as a function of the injected RF power. The dynamic (frequency and amplitude) of such fluctuations has been related to the spontaneous radial electric field consequence of different electron density profiles. The results from the new probe array, recently implemented in GyM to provide a deeper study of the spatial distribution of turbulence, are shown.   

Experimental Study of Current Filamentation Instability

Current Filamentation Instability, CFI, is of central importance for the propagation of relativistic electron beams in plasmas and could play an important role in the generation of magnetic fields and of radiation in the after-glow of gamma ray bursts and for energy transport in the fast-igniter inertial confinement fusion concept. Simulations of the ATF beam and plasma parameters using the particle-in-cell code QuickPIC indicate the presence of the CFI. Our goal is to experimentally study and characterize the CFI at the ATF at BNL as a function of beam and plasma parameters. The experiment has two stages; the first stage incorporates longitudinal diagnostics (longitudinal density) and the second is transverse diagnostics (magnetic field and transverse density). We have started the first phase and present simulation results, experimental setup and considerations, results to date and transverse diagnostic design.   

Ion heating by the alpha-channeling mode in mirror machines

In some mirror reactor designs, it might be advantageous to redirect a part of the extracted energy to the fuel ions in the central cell and the device plug rather than heat plasma electrons. Previously identified modes suitable for alpha-channeling in simple mirror machines were shown to damp most of their energy on electrons. Two techniques based on the minority ion species injection and mode coupling capable of redirecting a part of the wave energy to the ion heating are demonstrated and their efficiency is estimated. Also, a method of calculating the alpha-channeling mode structure is proposed to evaluate the feasibility of energy redirection in practical mirror devices.   

Continuum Damping of Free-boundary TAE with AEGIS

An extension has been added to the ideal MHD code AEGIS (Adaptive EiGenfunction Independent Solutions) to estimate continuum damping of an Alfvenic mode. In our scheme we analyze the determinant arising from attempting to match solutions at the surface of the plasma vacuum interface. A zero of the determinant corresponds to an eigenvalue of the system. When continuum damping exists in a stable system, the eigenmode cannot be calculated by an integration along the real axis (in principle integration in deformed regions of the complex plane is required). The approach we take here is to scan the value of the determinant as a function of complex frequency where the imaginary part of the frequency is positive. The analytic continuation of the determinant gives an estimate of the root in the lower half plane, from which the damping rate is extracted. A complicating factor in our procedure is that the positions of a pole and zero of a determinant is frequently comparable to the damping rate. Hence, the search procedure must account for both the zero and pole structure of the determinant. It is interesting to note that the root of the pole corresponds to the eigenvalue of the problem where an ideal conducting wall is placed on the plasma vacuum interface. We are attempting to apply our new subroutine to realistic equilibria, such as C-Mod.   

Two Stream Instability in an Inhomogeneous Magnetized Plasma

A plasma can support several kinds of electrostatic and electromagnetic waves depending upon what the external perturbations are and how the plasma is generated or under what conditions the waves are excited. In the presence of magnetic field, more new kinds of the waves can evolve in the plasma. Moreover, if the free energy is available in the system, then these waves can grow at the cost of this free energy and hence instabilities can take place. In the present investigation, we make an analytical study of a two stream instability in an inhomogeneous magnetized plasma having drifting ions and electrons. We consider a general situation by taking an oblique propagation of the wave from the magnetic field and a constant ionization. Using normal mode analysis, we derive dispersion relation which is solved numerically for the growth rate of the instability. The behavior of growth rate with magnetic field and the propagation angle along with ionization constant has been studied for two different wavelengths of the oscillations. We observe two type of the instabilities out of which one grows at a faster rate and hence is named as fast growing instability. The growth rate of the other slowly growing instability is also examined. We find that the growth rates of both the instabilities attain higher magnitudes at smaller wavelength of the oscillations.   

Visualization of Rotating Spoke Instabilities in a Hall Thruster

High-speed imaging of several Hall thrusters spanning an order of magnitude in discharge current has revealed an omnipresent visible rotating spoke mode, propagating at hundreds to the low thousands of meters per second in the direction of the ExB drift. An extensive software suite has been developed to facilitate image processing and visualization of these spokes in annular and cylindrical Hall thruster discharges. Modes with regular structure ranging from 2 to 5 simultaneous spokes have been observed, sometimes as single stable patterns lasting for a few seconds (the length of a typical video) and other times in a turbulent transition between several modes trading off every few milliseconds. Such azimuthal waves have long been suggested as a mechanism for cross-field electron transport, yet the visible spoke frequencies are not seen in conventional discharge current measurements. To address this disagreement we outline the design and construction of an azimuthally segmented anode capable of directly measuring rotation of the Hall thruster discharge.~   

Autoresonant BGK modes

Resonant wave interactions play a major role in plasmas and other nonlinear media. The interacting waves may exhibit autoresonance, i.e. a continuous nonlinear phase-locking despite variation of system parameters [1]. We present a kinetic wave autoresonance paradigm, where a driven BGK mode is excited by slow variation of the driving wave frequency. We assume a flat- top initial electron distribution, drive the system by a slowly varying ponderomotive wave. Initially, the drive's phase velocity is outside the velocity distribution, so the electrons do not experience Cherenkov resonance. Later, as the driving frequency decreases, a stable phase-space hole is formed and, under certain conditions, remains phase-locked with the drive [2,3]. The electrostatic field associated with this structure comprises the autoresonant BGK mode. We study the problem analytically and numerically within the Vlasov-Poisson system and use the Whitham's averaged variational principle [4] to describe the adiabatic evolution of the BGK mode. [1] L. Friedland, Scholarpedia 4, 5473 (2009). [2] L. Friedland, P. Khain, and A.G. Shagalov, Phys. Rev. Lett. 96, 225001 (2006). [3] P. Khain and L. Friedland, Phys. Plasmas 14, 082110 (2007). [4] G.B. Whitham, Linear and Nonlinear Waves, Willey, New York, 1974.   

A nonlinear theory of the parallel firehose and gyrothermal instabilities

Weakly collisional magnetized plasmas tend to develop pressure anisotropies which trigger fast ($\sim$ ion cyclotron period) plasma instabilities at scales between the ion Larmor radius $\rho_i$ and the mean free path $\lambda_{mfp}$. These can dramatically affect the global ($\gg\lambda_{mfp}$) dynamics and their nonlinear evolution should drive pressure anisotropies towards marginal stability values, controlled by the plasma beta $\beta_i$. This nonlinear evolution is worked out in an {\em ab initio} kinetic calculation for the parallel ($k_\perp=0$) firehose instability in a high-beta plasma. We use a particular physical asymptotic ordering to derive a closed nonlinear equation for the firehose turbulence, which we solve. We find secular ($\propto t$) growth of magnetic fluctuations and a $k_\parallel^{-3}$ spectrum, starting at scales $\ga \rho_i$. When a parallel ion heat flux is present, the parallel firehose instability mutates into the new {\em gyrothermal instability}. Its nonlinear evolution also involves secular magnetic energy growth, but its spectrum is eventually dominated by modes with a maximal scale $\sim\rho_il_T/\lambda_{mfp}$, ($l_T$ is the parallel temperature gradient scale).   

LIF diagnostic for flow shear instability studies at the NRL SPSC

An ongoing series of experiments at the Naval Research Laboratory Space Physics Simulation Chamber (NRL SPSC) has investigated ion cyclotron instabilities due to sheared plasma flow. Both electrostatic and, most recently, electromagnetic instabilities have been observed (see E. Tejero {\it et al.}, this conference). A three axis laser induced fluorescence (LIF) diagnostic is being constructed for direct measurement of the flow layer profile, which previously has been inferred from electric field measurements. The spatial resolution (2mm) will be sufficient to observe the ion gyroscale flow shear. This diagnostic uses a singly ionized Argon level scheme: a 500mW diode laser pumps the transition at 668nm and fluorescence is observed at 442nm.   

Spontaneous Electromagnetic Emission from a Strongly Localized Plasma Flow

Laboratory observations of electromagnetic ion cyclotron waves generated by a localized transverse dc electric field are reported. Experiments indicate that these waves result from a strong \textbf{E}$\times $\textbf{B} flow inhomogeneity in a mildly collisional plasma with sub-critical magnetic field-aligned current. The wave amplitude scales with the magnitude of the applied radial dc electric field. The electromagnetic signatures become stronger with increasing plasma $\beta $, and the radial extent of the power is larger than that of the electrostatic counterpart.   

Development and Implementation of Magnetic Probe Diagnostics on a Linear Magnetized Plasma Column

Electromagnetic ion cyclotron (EMIC) waves in the near-Earth space environment are of interest because of their ability to interact with and scatter energetic electrons. Experiments in the Auburn Linear Experiment for Instability Studies (ALEXIS), a 170 cm long, 10 cm diameter magnetized plasma column, are part of a combined study with the Naval Research Laboratory to study the generation and propagation of EMIC waves. This presentation reports on the development of a single axis magnetic loop probe for measuring the changing magnetic field. Results will be presented on the development of the magnetic loop probe and calibration data. Results of initial measurements of magnetic fluctuations in the ALEXIS plasma may also be presented.   

Potential and flow modification in a linear magnetized plasma column

Transverse and parallel sheared flows are important topics in both space and fusion plasmas, and have been the subjects of extensive study. The scale size of space plasmas, and high temperatures of fusion plasmas, provide unique diagnostic challenges. Small scale laboratory experiments are often more flexible and provide easier diagnostic access than other plasma environments. The Auburn Linear EXperiment for Instability Studies (ALEXIS) is a 170 cm long, 10 cm diameter, linear magnetized, rf generated plasma column, which, in addition to existing Langmuir and emissive probes, has recently been outfitted with a Laser Induced Fluorescence system. Recent experiments have focused on modifying the plasma potential and characterizing the plasma response. Initial results indicate that modification of the radial electric field results in modification of both the azimuthal and radial ion flows. Measurements will be presented on the correlation between different low frequency wave features and the electric field, density, and flow structures in the plasma.   

\textit{S/XB} measurements for Mo I and W I lines

In the spectroscopic method to determine sputtered impurity influxes, the ionization events per photon (\textit{S/XB}) value is essential to convert the line emission intensity into a particle flux [1]. However, experimental data of \textit{S/XB} values for Mo I are scarce and for W I sometimes inconsistent. In the linear divertor simulator PISCES-B, we have determined \textit{S/XB} values of Mo I and W I lines by measuring the line emission of sputtered atoms by He or Ar plasma bombardment. While our measured values for the Mo I transition of $z$ $^{7}$P$^{o} \rightarrow \quad a \quad ^{7}$S (379.8, 386.4, 390.3 nm) are systematically $\sim $2-3x lower than theoretical values, agreement for other transitions is more satisfactory. For the W I line at 400.8 nm, we reported that our values of $\sim $100-200 were $\sim $5-10x larger than previously reported experimental data at electron temperature $>$ 10 eV [2]. Our recent measurements provide \textit{S/XB} $\sim $ 55 at higher electron density, where the geometrical loss flux from the plasma column is further reduced. Supported by the US DOE contract No. DE-FG02-07ER54912.\\[4pt] [1] A. Pospieszczyk et al., J. Phys. B \textbf{43}, 144017 (2010). [2] D. Nishijima et al., Phys. Plasmas \textbf{16}, 122503 (2009).   

Kinetic simulations of divertor heat load profile and dependence on plasma current and heating power

One performance target for the DOE Office of Fusion Energy Sciences this year is to improve understanding of heat transport in the tokamak scrape-off layer plasma and divertor conditions in ITER. Divertor heat flux and edge plasma profiles have been measured in multiple tokamak devices for H-mode discharges with similar dimensionless plasma parameters and scans over plasma current and heating power. For this FY 2010 Joint Research Target milestone, the Center for Plasma Edge Simulation has performed kinetic simulations of the edge region for a subset of these discharges using the discrete guiding-center neoclassical transport code XGC0. This code includes realistic X-point geometry, a neutrals source model and collisions, and a diffusion model for turbulent transport. Plasma particles are initialized with measured density and temperature profiles. Particle and heat load data is collected on the inner and outer divertor plates. In this presentation, we compare heat load estimates for DIII-D, NSTX and Alcator C-Mod similarity discharges and show the trends with varying plasma current and heating power.   

H-mode Snowflake Divertor Plasmas on TCV

An ELMy H-mode ``snowflake'' divertor is established and studied for the first time in the TCV tokamak. The H-mode access and the ELM dynamics is compared to a conventional single-null diverted configuration. The snowflake configuration exhibits $15\%$ higher confinement and 2 to 3 times lower ELM frequency. Ideal MHD stability analysis suggests enhanced stability of the snowflake H-mode pedestal to mid- to high-toroidal-mode-number modes. The capability of the snowflake to redistribute the edge power on the additional strike points has been confirmed experimentally.   

Plausible collective phenomena associated with dust in fusion plasma

We consider collective impact of nano-scale dust on edge plasma phenomena. We investigate possible impact of nano-dust on MARFE and find that it is plausible that dust, providing plasma particle sink and causing plasma flow, which overcomes the thermal force on impurity localizing in low temperature region. We also show that rather modest amount of dust, compatible with tokamak conditions, can significantly reduce the growth-gate of flute instability (which is the simplest proxy of ballooning mode) making it more susceptible to other stabilization mechanisms (e.g. shear of plasma flow velocity).   

Impact of dust originated impurities on fusion edge plasmas

The impact of dust originated impurities on fusion edge plasmas is analyzed using coupled DUSTT/UEDGE code, which allows self-consistent modeling of dust and plasma transport in the edge. The code validation is performed using 3D reconstructed dust trajectories measured on NSTX. Various scenarios of injection of dust particles with different sizes, speeds, and composition materials are simulated and compared with available experimental data from NSTX and other tokamaks. The modeling demonstrates that dust injected with rates of order or larger than 10mg/s has profound effects on the edge plasma profiles, transport, and stability. In particular, it is shown that the plasma contamination with the dust originated impurities can lead to substantial increase of the radiation power losses and decrease of the radial pressure gradients and of the radial plasma fluxes in the edge. Significant reduction of the power load to the divertor plates is also observed for the simulated dust injection rates. The differences between the injection of dust and of equivalent amounts of gaseous impurities are discussed.   

Modeling of the 2007 JET $^{13}$C migration experiments

Using the last run day of the 2007 JET experimental campaign, $^{13}$CH$_{4}$ was introduced repeatedly from the vessel top into a single plasma type (H-mode, I$_{p}$= 1.6 MA, B$_{t}$= 1.6 T). Similar experiments were performed in 2001 (vessel top into L-Mode) and 2004 (outer divertor into H-Mode). Divertor and wall tiles were removed and been analysed using secondary ion mass spectrometry (SIMS) and Rutherford backscattering (RBS) to determine the $^{13}$C migration. $^{13}$C was observed to migrate both to the inner (largest deposit), outer divertor (less) , and the floor tiles (least). This paper reports the EDGE2D/NIMBUS based modelling of the carbon migration. The emphasis is on the comparison of the 2007 results with the 2001 results where both injections were from the machine top but ELMs were present in 2007 but not present in 2001. The ELMs seemed to cause more $^{13}$C re-erosion near the inner strike point. Also of interest is the difference in the Private Flux Region deposits where the changes in divertor geometry between 2004 and 2007 caused differences in the deposits. In 2007, the tilting of the load bearing tile caused regions of the PFR to be shadowed from the inner strike point which were not shadowed in 2004, indicating $^{13}$C neutrals originated from the OSP.   

Dynamics of Charged Dust Particle near Conducting Wall in TOKAMAK

A substantial amount of dust has been observed to be present near the first walls of fusion devices. The impact of dust on plasma parameters in current and future fusion devices is not clear and may cause a significant safety threat. It is therefore important to understand the dynamics of dust particles after formation. A surface charge is induced on the wall of a conducting material in the presence of a charged particle. The charged particle is then attracted to the wall by this induced charge causing the charge in the wall to redistribute and thus increasing the force of attraction further. In this work we study the dynamics of this attraction and the dissipation of electromagnetic energy via joule heating within the conducting wall.   

Super{\_}X divertor geometries and plasma simulations

Potential implementations on NSTX, a component test facility (CTF) and a fission-fusion hybrid are considered. Magnetic equilibria are devised using CORSICA for desirable core plasma configurations, to allow considerable flexibility, and without unacceptable coil currents or positions. Using the state of the art edge simulation code SOLPS, plasma simulations of these configurations are considered, to examine their advantages.   

Influencing Edge Turbulence via X-point Geometry

Turbulent fluctuations near the last closed flux surface of a diverted tokamak are strongly influenced by the combination of parallel electron motion and magnetic geometry. Close to the X-point, localized magnetic shear distorts flutelike perturbations, exponentially enhancing $k_{\perp}$. The parallel current allows the resulting high-$k_{\perp}$ X-point physics to effectively react back on the turbulent drive in the outboard midplane region, suggesting that turbulence and transport in the edge region could be influenced through variation of the X-point geometry. Effects of such variation, for example the separate and joint rotation of the separatrices at the X-point, are investigated via systematic field-line projection of warm-ion gyrofluid equations onto a parametrized family of model X-point magnetic geometries. Physical effects are clarified via the explicit dependence of the advection and dissipation operators on the geometric parameters. Possible experimental applications are discussed.   

Kinetic Simulation of pedestal transport with self-consistent RMP penetration

It is of the highest importance for ITER program to understand the pedestal transport and the RMP penetration physics self-consistently. Coupled kinetic-fluid simulation is performed in a realistic DIII-D diverted geometry, with electrons and ions orbiting under self-consistent 3D RMPs, radial electric field, Coulomb collisions, neutral kinetic transport with wall recycling, impurity radiation, and a heat flux and a torque from the core. For the kinetic edge transport simulation, we use the full-function kinetic ion-electron- neutral guiding-center PIC code XGC0. For fluid simulation, we have installed a proper fluid equation in M3D-OMP. The result shows that the RMP amplitude weakens significantly from the vacuum value within the magnetic separatrix, and that electron density is significantly reduced in a manner qualitatively consistent with experiments. The study reveals that the kinetic effect is essential in understanding the stochastic plasma transport in a toroidal magnetic confinement device.   

Stabilization of bursty transport by externally imposed magnetic perturbations

Resonant magnetic field perturbations (RMP) introduced by external coils are found to mitigate type I ELMS in tokamak plasmas. Here we present results using the DRBM code, which solves the Braginskii fluid equations in a flux-tube geometry, to study the effect of externally imposed magnetic perturbations to mimic the RMPs. Transport bursts in simulations of the edge plasma in tokamaks are suppressed by the application of magnetic field perturbations. The amplitude of the applied magnetic field perturbations is characterized by a stochasticity parameter $S$. As $S$ increases, but is still less than unity so that most of the magnetic flux surfaces are still preserved in the simulations, the magnitude of the ELM bursts decreases. The size of bursts is reduced to a very small value while $S$ remains below unity. Widespread magnetic stochasticity is not a requirement for the stabilization of bursty transport observed in the simulations by the magnetic field perturbations. The magnetic field perturbations are found to suppress the growth of the resistive ballooning instability that underlies the bursty transport.   

New approach in multi-fluid modeling of edge plasma transport with high intermittency due to blobs and ELMs

Strong fluctuations of edge plasma parameters via blobs result in a strong spatiotemporal variation of all nonlinear functions of plasma parameters (e.g. ionization rate, heat/momentum fluxes, plasma-wall interactions), hence, causing serious problems with both interpretation of experimental data and modeling using 2-D fluid transport codes based on averaged plasma parameters. In new modeling approach we include spatiotemporal features of blobs and ELMs using the 2-D transport code UEDGE in a time-dependent mode. Our model is based on multi-fluid simulation of an ensemble of plasma ``macro-blobs'' appropriately seeded in the edge plasma according to experimental statistics of blobs. The model properly projecting inherently 3-D filamentary structures associated with blobs on the 2-D poloidal geometry. We report the initial results from UEDGE modeling of ``macro-blob'' dynamics and non-linear interaction with background plasma. The impact of model assumptions and initial conditions on individual blob dynamics and background plasma will be discussed. Next steps of this work are (1) development of 3-D tools performing the spatiotemporal averaging to calculate the signals of tokamak diagnostics from the UEDGE plasma with blobs; (2) simulation of ensemble of ``macro-blobs''; and (3) benchmarking the approach against experimental data from current tokamaks.   

Diffusive-convective transition for scrape-off layer transport

Transport of plasma from the edge pedestal gradient region into the scrape-off layer (SOL) impacts the SOL width, critical for future tokamak devices. This width is believed to be set by a competition between classical parallel transport and turbulent cross-field transport. In previous work, [J.R. Myra et al, submitted to J. Nucl. Mater. (2010)] focusing on modeling of the heat flux width in NSTX experiments, the possibility of a transition from quasi-diffusive to convective transport in the SOL was noticed. This transition is explored here theoretically and through SOL turbulence simulations using the SOLT code [D. A. Russell, et al, Phys. Plasmas 16, 122304 (2009)]. At the transition, the transport becomes intermittent, and the SOL width is broadened due to blob emission. Critical dimensionless parameters for the transition are investigated. The inputs to the model include the net heat flux into the SOL and the parallel particle and energy confinement times, which are related to the connection length and plasma collisionality.   

When is it Valid to Assume that Heat Flux is Parallel to $B$?

It is frequently assumed that heat flow in the plasma scrape-off-layer is everywhere parallel to \textbf{\textit{B}}, due to the strong anisotropy in electron thermal conductivity. This assumption is convenient but paradoxical. Here are examined three situations where this assumption has sometimes been applied: 1)~extrapolating from midplane \textit{Te}($R)$ measurements to divertor heat flux profile, 2)~determining the location of the separatrix from measured midplane \textit{Te}($R)$, combined with total heat flux leaving the plasma, and 3)~predicting the heat flux to plasma-facing components in the scrape-off-layer of diverted plasmas. Numerical solution of the anisotropic, nonlinear heat equation suggests that the first application is poor, the second well justified, and the third very far from accurate. Additional plasma physics effects may mitigate these results, but the simple assumption of dominant parallel heat flow due to anisotropy in electron thermal conductivity is not supported in many important cases.   

Analysis of time-dependent particle transport in the tokamak boundary plasma

Plasma particle transport in the edge and scrape-off layer of tokamaks is not well understood but is important for core fueling, helium removal, and impurity intrusion. A simple 1D model is presented to clarify the time-dependent impact of different possible mechanisms including penetration and ionization of recycled or injected neutrals, and plasma diffusion and convection. More detail of edge profile dynamics between Edge-Localized-Modes (ELMs) corresponding to re-building of the pedestal region is studied with the 2D UEDGE transport code. The influence of the ion pinch associated with perpendicular ion viscosity is evaluated. It is assumed that during quasi-steady-state discharges with regularly-spaced bursts of Edge Localized Modes (ELMs), the net pumping of all walls and pumps averaged over an ELM cycle is just sufficient to remove the small neutral beam particle source. The simulation results are then compared with similar time-dependent data for DIII-D edge density profiles between ELMs.   

General properties of the magnetic field in a snowflake divertor

The power-law series for the poloidal magnetic flux function, up to the third order terms, are presented for the case where two nulls of the poloidal magnetic field are separated by a small distance, as in a snowflake divertor. Distinct from the earlier results, no assumptions about the field symmetry are made. Conditions for the realization of an exact snowflake are expressed in terms of the coefficients of the power series. It is shown that, by a proper choice of the coordinate frame in the poloidal plane, one can obtain efficient similarity solutions for the separatrices and flux surfaces in the divertor region: the whole variety of flux surface shapes can be characterized by a single dimensionless parameter. Transition from a snowflake-minus to snowflake-plus configuration in the case of no particular symmetry is described. The effect of the finite toroidal current density in the divertor region is assessed. A possibility of creating a near-snowflake configuration in the ITER-scale facilities is discussed. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.   

Progress with and status of COGENT

COGENT is a continuum gyrokinetic code being developed by the Edge Simulation Laboratory for edge plasmas. The code is distinguished by application of 4th order conservative discretization, and mapped multiblock grid technology to handle the geometric complexity of the tokamak edge. We report on a verification campaign involving simulation of geodesic acoustic modes. We have performed simulations over a range of safety factors and at multiple values of $T_e/T_i$, and extended the evaluation of the analytic formalism of Gao {\it et al.}\ to encompass the simulation parameter domain. We compare COGENT results with the theory and with results from other codes, and in particular find excellent agreement between COGENT and theory. We will also report on several lines of code upgrade activity, including application of the mapped multiblock capability to divertor geometry, incorporation of a simple collision operator and a spatial diffusion operator to model the effects of anomalous transport, implementation of implicit time advance for diffusive operators, and an automated code testing system.   

Nonlinear Simulations of Peeling-Ballooning modes in ITER H-mode scenario

A minimum set of equations based on the Peeling-Ballooning (P-B) model with non-ideal physics effects (toroidal flow shear, diamagnetic drift, ExB drift, resistivity, and anomalous electron viscosity) is found to produce some essential features of pedestal collapse when using the BOUT++ simulation code. It is found from nonlinear simulations for a realistic high Lundquist number that the pedestal collapses are limited to the edge region and the ELMs size is about 8-10\% of of the pedestal stored energy, which is consistent with many observations of large ELMs. Nonlinear simulations demonstrate that the nonlinear P-B modes trigger magnetic reconnection, which then leads to the partial collapse of the pedestal. For one of the latest designs of the ITER 15MA inductive H-mode scenario (under the burning condition), linear growth rate, the ELM size, and power deposition pattern on ITER plasma facing components will be quantified.   

Ion orbit loss effects on radial electric field in tokamak edge for standard and snowflake divertor configurations

Neoclassical orbits and related non-ambipolar losses in tokamak edge are sensitive to the radial electric field Er; therefore an electric field in the edge may be affected by the ion orbit prompt loss at the null point [1]. This effect may be sensitive to the details of magnetic field near the null-point. In the snowflake divertor configuration [2], the prompt loss is quite different from that in the standard X-point configuration since the zone affected by the prompt loss in snowflake geometry is significantly larger than that for the standard X-point [3]. Here we extend the analytic study in [3] by including the effects of electric field and collisions using numerical calculation of drift orbits. Radial electric field induced by the prompt loss is compared for otherwise similar standard x-point configuration and a snowflake. \\[4pt] [1] C.S. Chang, S. Kue, H. Weitzner, Phys. Plasmas, 9, 3854 (2002).\\[0pt] [2] D.D. Ryutov, Phys. Plasmas, 14, 064502, June 2007.\\[0pt] [3] D.D. Ryutov and M.V. Umansky, Phys. Plasmas, 17, 014501 (2009)   

Gyrokinetic Models for Edge Plasmas*

The use of gyrokinetic equations for the simulation of magnetic fusion edge and scrapeoff-layer plasmas requires that the equations be valid for large relative perturbation amplitudes and, possibly, large flows. The Hamiltonian gyrokinetic theory has therefore been extended to two new orderings [1,2] that are more general than the standard ones in that they allow for potential perturbations or \textbf{\textit{E$\times $B}} flows of order the thermal levels. These theories both generalize and show that additional terms should have been present some related prior work. Here, full (low-\textit{$\beta $}) electromagnetic toroidal equation sets are presented, and he energy conservation relations are derived using Noether's theorem in a Lagrangian variational approach. Useful subsidiary and reduced orderings are also considered that result in considerable simplification, and methods for the numerical implementation of the new terms in the equations will also be discussed. *This work was performed for US DOE by LLNL under Contract DE-AC52-07NA27344 and is part of the ESL. \\[4pt] [1] A.M. Dimits et al., Phys. Fluids B\textbf{4}, 274 (1992). \\[0pt] [2] A.M. Dimits, Phys. Plasmas \textbf{17}, 055901 (2010).   

High-resolution numerical schemes for cross-magnetic-field drift in UEDGE and comparison with impurity flow measurements

A first order upwind scheme has been employed by the UEDGE edge transport code for the cross-magnetic-field drift terms even though it induces numerical diffusion. In a steep gradient H-mode, the physical diffusion coefficient is believed to be small in the transport barrier region where numerical diffusion may be as large as, or even exceed, the physical diffusion. Therefore the first order upwind scheme can degrade simulation results especially currents in the scrape-off-layer. An approach by Rozhansky et al. reduces radial numerical diffusion at the expense of a larger poloidal numerical diffusion. Here an alternate approach of higher-order schemes are investigated for UEDGE. The convergence and accuracy of the new numerical schemes are compared to the first order upwind scheme. These numerical improvements are further tested in comparing simulations with edge carbon flow measurements using a new Fourier Transform Spectrometer on DIII-D.   

Using Non-Axisymmetric Scrape-Off Layer Current to Control Edge-Localized Modes and Equilibrium Profiles

Control of both the time-averaged divertor heat loads and the impulsive heat fluxes that accompany edge-localized modes is critical to the success of future tokamaks, including ITER. Driving toroidally non-axisymmetric current through the scrape-off layer (SOL) plasma can achieve both goals simultaneously by broadening the SOL by driving radial convection [1] and by driving edge-resonant magnetic perturbations in order to control the edge pressure gradient [2]. We analyze reactor-relevant SOL current-drive mechanisms that reduce power requirements and do not require internal insulators using analytic theory and reduced 1d and 2d numerical models using the UEDGE code. Calculations show that toroidally localized puffing and pumping of neutral gas and/or impurities can generate substantial SOL current and that choosing the appropriate mode number, width and phasing at the target plate can amplify the desired effects. \\[4pt] [1] R.H. Cohen and D.D. Ryutov, Nucl. Fusion 37, 621 (1997). \\[0pt] [2] I. Joseph, R.H. Cohen and D.D. Ryutov, Phys. Plasmas 16, 052510 (2009).   

Radiation Resistant Hydrogen Microsensors for Fusion Applications

Quantifying the flux and energy of charge exchange neutrals to the walls of fusion experiments is important to understanding wall erosion and energy balance. Quantification of these fluxes is made much more difficult because they have very strong poloidal and toroidal variations. To facilitate such measurements, we have been developing compact, palladium metal oxide semiconductor (Pd-MOS) detectors. These devices are dosemetric detectors, which can evaluate differences between plasma discharges. To become widely used, however, such detectors must be made resistant to UV and x-ray induced damage, as well as high energy particle bombardment. We report here on the fabrication of Schottky diode Pd-MOS devices in which we have minimized the oxide thickness (to reduce the production of charges from UV and x-rays) and increased the Pd overlayer (to reduce charge production from high energy particles). The fabrication has been facilitated through use of an array of metallic posts to improve the Pd film adhesion. The efficacy of the film adhesion and comparison with standard detectors will be examined. Testing and calibration of the detectors is reported as a function of hydrogen flux and energy.