Session NP1 - Poster Session VIII.
POSTER session, Wednesday afternoon, October 25
Exhibit Hall AB, Qu\'{e}bec City Convention Centre

[NP1.001] Partially Ionized and Low Temperature Plasmas II (DPP)

This abstract not available.

[NP1.002] Athermal Annealing of Silicon Implanted with As, P, and B

We present results of experiments in which silicon samples ion-implanted with phosphorus, arsenic, or boron are annealed, i.e. electrically activated, without direct heating. Instead, the samples are annealed by illuminating a single small, ~1-square-millimeter, spot on their surface with one pulse from a laser beam whose intensity is high enough to create a hot plasma within the focal spot. Annealing occurs outside the laser spot to a distance of 1-2 cm, where no direct heat is applied. We will show that the electrical activation of these samples is equivalent to thermally annealed controls, and that dopant diffusion into the samples is very small. Such low diffusion is necessary for the manufacture of very high density semiconductor circuits in which the spacing between circuit elements is very small. A plausible explanation for the annealing is that shock or high-amplitude sound waves generated by backpressure from plasma within the laser focal spot provide the energy for annealing. We will provide measurements that support this hypothesis.

[NP1.003] Plasma Assisted Fabrication of Carbon Based Materials

The Plasma Physics Laboratory at the University of Saskatchewan has recently initiated research on plasma based material processing. A glow discharge device has been constructed to deposite synthesized diamond films on silicon substrates heated by a hot filament. The substrates were pretreated by various means before they were installed in the reaction chamber. The scanning electron microscope (SEM) photographs and x-ray diffraction analysis have revealed diamond grain characteristics of the film. More recently, an arc discharge device using graphite electrodes is being designed for fabrication of fullerene and carbon nanotubes. The properties of diamond films deposited under various conditions and preliminary arc discharge results will be presented.

[NP1.004] Epitaxial Growth of AlInN by RF Magnetron Sputtering

With the bandgap varying from 6.2 eV in AlN to 1.9 eV in InN, the ternary AlInN has a great promise for applications in light-emitting diodes and laser diodes operating in the ultraviolet to orange spectrum, low-cost solar cells with high efficiency, and various types of sensors. However, little attention has been paid to this material because of the difficulty in the growth from a thermodynamical viewpoint. In this paper, we report the growth properties of AlInN films obtained by low temperature sputter method. The AlInN films were grown by radio frequency magnetron sputtering in an ambient of argon and nitrogen, using aluminum and indium targets. The structural properties were assessed by reflection high-energy electron diffraction and high-resolution triple-axis x-ray diffraction measurements and the optical properties were investigated with an optical absorption method. It was revealed that epitaxial AlInN films with wurtzite structure can be obtained on (0001) Sapphire and (111) GaAs substrates at a growth temperature as low as 100 ^\circC. The films can be prepared over the large composition range by controlling the plasma reaction and no solid-phase miscibility gap in the films was observed. The bandgap of the AlInN films was found to be increased with increasing aluminum content.

[NP1.005] Intense negative ion beams from a plasma source for focused ion beam applications

With the goal to augment the present state of the art of focused ion beam (FIB) systems for different applications in microelectronics arena a new approach using intense, high-brightness negative ion beams from a plasma source is being pursued. A compact Penning-type plasma source has been developed, and detailed studies of the characteristics of ion beams, specially H^- beams, have been made. Results with hydrogen ions show a beam brightness of > 10^5 A cm^-2 sr^-1 and an energy spread of ~ 2 eV for an angular beam intensity of >40 mA/sr. These values of the characteristic beam parameters are extremely attractive to develop an efficient FIB system for different applications including ion beam microscopy and ion beam milling. Results with a simple electrostatic focusing system suggest that a sub-10 nm spot size with current density of > 1 A/cm^2 is achievable with beams from this source. Current FIB systems primarily use ion beams from point sources, namely, liquid metal ion source. Another approach, currently being pursued, is based on gas-field ionization sources. Our approach with a plasma source has merits to provide stable, high throughput beams of widely different species from gas precursors using a relatively simple and robust technology. A roadmap detailing the plasma source, its focusing optics and needs for relevant applications will be discussed.

[NP1.006] Atmospheric Pressure RF Capacitive Plasma Source

An atmospheric pressure plasma source has been developed and it exhibits many desirable properties for a wide range of plasma applications without the complication and expense of an attached vacuum system. An atmospheric pressure plasma jet (APPJ) operates with rf power and produces a \alpha-mode capacitive discharge that is stable, steady-state, and non-thermal. The plasma parameters of this source have been measured: electron densities of ~ 10^11 cm^-3 and electron temperatures of ~ 2 eV by using neutral bremsstrahlung emission. The gas temperature of less than 150 ^oC is determined in the discharge by using rotational intensity distribution. Localized electron heating near the sheath boundary has been observed and is related to the discharge stability and subsequent \alpha to \gamma mode (or arcing) transition. Recent progress has been made to replace the helium in the discharge while maintaining discharge stability and reactivity, thus broadening the appeal of this source. Thus far, the APPJ has been used to etch polyimide, tungsten, tantalum, and silicon dioxide and to deposit silicon dioxide films at rates comparable to those in low pressure plasma sources.

[NP1.007] Killing Microorganisms with the One Atmosphere Uniform Glow Discharge Plasma

There is an urgent need for the development of new technologies for sterilization and decontamination in the fields of healthcare and industrial and food processing that are safe, cost-effective, broad-spectrum, and not deleterious to samples. One technology that meets these criteria is the One Atmosphere Uniform Glow Discharge Plasma (OAUGDP). The OAUGDP operates in air and produces uniform plasma without filamentary discharges at room temperature, making this technology advantageous for sterilization of heat sensitive materials. The OAUGDP operates in a frequency band determined by the ion trapping mechanisms provided that, for air, the electric field is above 8.5kV/cm. The OAUGDP efficiently generates plasma reactive oxygen species (ROS) including atomic oxygen and oxygen free radicals without the requirement of a vacuum system. We have demonstrated the efficacy of the OAUGDP in killing microorganisms including bacteria, yeast, viruses, and spores in seconds to minutes on a variety of surfaces such as glass, films and fabrics, stainless steel, paper, and agar.

[NP1.008] Plasma Processing with a One Atmosphere Uniform Glow Discharge Plasma (OAUGDP)

The vast majority of all industrial plasma processing is conducted with glow discharges at pressures below 10 torr. This has limited applications to high value workpieces as a result of the large capital cost of vacuum systems and the production constraints of batch processing. It has long been recognized that glow discharges would play a much larger industrial role if they could be operated at one atmosphere. The One Atmosphere Uniform Glow Discharge Plasma (OAUGDP) has been developed at the University of Tennessee Plasma Sciences Laboratory. The OAUGDP is non-thermal RF plasma with the time-resolved characteristics of a classical low pressure DC normal glow discharge. An interdisciplinary team was formed to conduct exploratory investigations of the physics and applications of the OAUGDP. This team includes collaborators from the UTK Textiles and Nonwovens Development Center (TANDEC) and the Departments of Electrical and Computer Engineering, Microbiology, Food Science and Technology, and Mechanical and Aerospace Engineering and Engineering Science. Exploratory tests were conducted on a variety of potential plasma processing and other applications. These include the use of OAUGDP to sterilize medical and dental equipment and air filters; diesel soot removal; plasma aerodynamic effects; electrohydrodynamic (EDH) flow control of the neutral working gas; increasing the surface energy of materials; increasing the wettability and wickability of fabrics; and plasma deposition and directional etching. A general overview of these topics will be presented.

[NP1.009] Analysis of V-I characteristics in a nano-crossed-field gap

In the advent fields of nano-technology and vacuum microelectronics, electrode gaps with scales down to nanometer regime can easily be fabricated. On such a microscopic scale, quantum effects such as electron tunneling will become important when the gap spacing is smaller than the electron wavelength. For example, electrons of low emission energy (< 1eV) emitted from the advanced cold cathodes such as field emitter array can have a wavelength larger than 1 microns. In this paper, we study the voltage-current (V-I) characteristics of a nano-size planar electrode gap with a crossed external magnetic field. First, we consider the electrons are emitted into the gap according to the Fowler-Nordheim (FN) emission law. By solving the 1D time-independent Schrodinger equation in the gap region, we obtain the V-I characteristics as a function of 6 parameters: gap spacing D, gap dc voltage V, electron emission energy E, external magnetic field B, and FN emission constants a and b. The V-I characteristics can be divided into 3 regimes according to the magnitude of the electron current density J. When J is low, the V-I characteristics follows the FN emission law, which is material dependent. As J increases, the FN regime transition to the space-charge regime, which is independent of material properties. Finally the V-I characteristics becomes the Child-Langmuir law (quantum extension) of finite magnetic field, which gives the limiting current density that can be transmitted across the gap as a function of D, V, E and B. By using properly constructed dimensionless parameters in our formulation, one can explore parametric dependence in various regimes. Further application and implication of this work will be discussed.

[NP1.010] Nonlinear Optical Diagnosis of Oxide Traps formed during Reactive Ion Etching

Oxide traps generated by reactive ion etching (RIE) are studied using a pulsed femtosecond laser. The second harmonic generation (SHG) signal from the Si/SiO_2 interface is sensitive to charged traps in the oxide. The time evolution of the SHG signal indicates that positive traps predominate. The angular dependence of the polarized signal shows that the electric field generated by the oxide traps alters the symmetry of the sample. The damage is greatest for an oxide thickness of 13 nm (for a plasma DC bias of 300~V). Thicker oxides have smaller SHG signals, presumably because the Fowler-Nordheim tunneling currents induced by plasma charging of the oxide surface are smaller. Very thin oxides also exhibit reduced damage. The time dependent SHG signals depend on the temperature of the samples; these data provide information on the trapping and de-trapping of substrate electrons by oxide holes.

[NP1.011] Electric wind in a positive point-to-ring corona discharge

The electric wind from a corona discharge is a transport phenomenon that occurs in the drift region of the discharge. Due to the non-ionising collisions of the ions with the neutral species, an overall movement of the gas takes place.

The purpose of this study is to determine velocity field of the electric wind. Two steps are to be done to achieve this goal.

In the first step, the ionic concentration from the drift region must be found. The charge carriers are subjected to both convection and thermal diffusion. The ionic concentration is given by the connective diffusion equation, coupled with the Poisson equation. To solve this system of equation one should determine the localisation of the glow region.

The second step uses the ionic concentration as well as the ions' velocity, to determine the partial pressure of the ions using the fundamental equation of kinetic theory of gases. The electric wind velocity is determined as a solution of the Navier-Stokes equation with the pressure term determined as early mentioned.

Considering this model it is possible to find those parameters of the corona discharge, that maximise the energy transfer from the discharge to the gas flow.

[NP1.012] Partially Ionized and Low Temperature Plasmas II (ICPP)

This abstract not available.

[NP1.013] The Flux Control of Cyclotron Ions Incident upon the Substrate by Employing the Ponderomotive Potential

The plasma-assisted deposition is desirable processing technique to grow the epitaxial film such as oxide-metal compound semiconductor on the lower temperature substrate. To control its stoichiometry ratio and orientation more precisely, handling the flux of oxygen and metallic ions incident upon the substrate is required. Here, we propose to employ the ponderomotive potential acting independently on each species of cyclotron ions in order to govern a new method for the flux control of ions selected before the substrate. Our idea is to form the potential structure decelerating the ions before the substrate, resulting in the enhancement of the ponderomotive force on their ions. The preliminary experiment is carried out using the magnetized plasma column terminated by the floating substrate where the localized rotating electrostatic field yielding the ponderomotive potential is impressed. It is found that the ion flux decreases when rf frequency approaches to the cyclotron frequency of helium ion. However, Q-value is small because the ions would not make the cyclotron motion many times until arriving at the substrate. In order to decelerate ions, we try to form the potential hump and/or to reconstruct the ion sheath to the effective electron sheath by the ponderomotive potential.

[NP1.014] Production of Multicharged Ions and Resonance-Surface Configurations of a 2.45 GHz Electron Cyclotron Resonance Source

[NP1.015] A Comprehensive Investigation of Plasma Parameters in a Pressure Dependent Plasma Mode Transition

The control of the electron cyclotron resonance(ECR) plasma is difficult at certain operation parameters. There are several reports of regimes where abrupt and unstable changes in plasma parameters occur as a function of gas pressure or microwave power. These mode changes are not yet well understood, and only a few plasma parameters have been reported simultaneously. In the present work we report extensive measurements of plasma potential, electron and ion temperature, ion energy and ion density over a mode change caused by variation of neutral argon pressure. The parameters were measured over the entire pressure range from 3.0e-4 to 9.0e-3 mbar. We found abrupt drops in density and potential at 5.0e-4 mbar, while the density and the electron and ion temperatures had minima around 1.0e-3 mbar. When the pressure was increased further, the temperatures and density increased rapidly after 1.5e-4 mbar. While the latter increase in density and temperatures might be explained by the transition from underdense to overdense plasma state, the previous drops in both potential, temperature and density cannot be directly attributed to this mechanism. We propose that there might exist a connection between the mode change and competition between excitation and direct ionization of the neutrals.

[NP1.016] The Radial Structure of a Plasma Column Sustained by a Surface Wave

The radial structure of a steady-state surface-wave-sustained argon plasma in a cylindrical dielectric tube surrounded by a metal is described in the context of a hydrodynamic model using three moment equations for electrons, including the electron heat flux, and two moment equations for ions, for which the ion-neutral collision rate is a function of the ion drift velocity. This plasma model is coupled self-consistently to Maxwell's equations for the waveguide TM mode. The radial profiles of the electron density and temperature are obtained for several values of each of the operating parameters (wave frequency, gas pressure, plasma tube radius, surface wave power absorbed per unit of plasma column length). The plasma and field profiles are significantly different from the classical Bessel type and the field profiles of surface waves in homogeneous media. Using a constant mean-free-path ion-neutral collision frequency (rather than the incorrect constant mean-free-time model) has a considerable effect on the profiles. The radial profiles for a typical case are compared for (i) the correct electron heat transport, (ii) isothermal case and (iii) no heat transport . It is shown that heat transport essentially influences onto radial structure of the discharge.

[NP1.017] 2D Modeling of a Large Diameter Plasma Column Produced and Sustained by Surface Waves

A steady-state surface-wave-sustained argon plasma in a cylindrical dielectric tube surrounded by a metal is considered. The study is carried out for surface wave propagating along the circumference of the cylinder and when surface wave skin depth is smaller than the plasma column radius. The spatial structure of the discharge is described in the context of a hydrodynamic model using three-moment equations for electrons and two moment equations for ions. The surface wave electric field distributions are found from Maxwell's equations. The profiles of the electron density and temperature as functions of the operating conditions (gas pressure, wave frequency, plasma tube radius and length, HF power absorbed in plasma column) are obtained. It is shown that the radial plasma density profiles are significantly different from the profiles obtained at constant electron temperature in a plasma column and from the profiles of long tubular surface-wave-sustained discharges. The plasma density profiles obtained from the presented model are compared with one obtained experimentally using a Langmuir probe in [1]. The results show qualitative agreement between our theory and experiment.

[NP1.018] Gas Discharge Sustained by the Electromagnetic Non-Symmetric Surface Wave in Magnetized Heterogeneous Plasma Column

This report is devoted to the study of the non-symmetric high frequency surface wave (SW) properties (dispersion properties, attenuation coefficient and radial wave field structure). Axial plasma density distribution under the condition of the diffusion mechanism of charged particle losses was studied as well. SW considered propagates in magnetized waveguide structure that consists of radial heterogeneous plasma column with radius a, dielectric tube with outer radius b, and vacuum region surrounded by the metal waveguide wall with radius c (a < b < c). External steady magnetic field is directed along the axis of the waveguide system. Plasma is considered in hydrodynamic approximation as cold and weakly collisional medium with constant value of effective electron - neutral collision frequency in the discharge volume. SW characteristics and axial electron density distribution were considered for the case when plasma density axial gradients are rather small. SW properties and plasma density axial distribution were investigated under different external magnetic field values, geometrical parameters of waveguide system and plasma density radial profiles. This work was supported by the Science and Technology Center in Ukraine (STCU, Project number 1112).

[NP1.019] Particle Fluxes in an Electron Cyclotron Resonance Plasma Discharge

In recent years several methods of plasma materials processing have been developed becoming more important and stimulating at the same time the study for achieving greater control of the energy and densities of ions and neutral particles. In addition to other benefitial effects these fluxes can provide the activation energy for compound formation and enhance surface diffusion processes. The enhancement of the surface bombardment by the generated energetic flux produces significant sputtering, surface heating, and a large number of surface defects. In this work, a model is described using the drift kinetic equation approach, to calculate the azimuthal and radial ion fluxes in the materials processing zone of an ECR plasma source. The azimuthal and radial ion fluxes, the mean energy, and density of ions are analyzed separately. The results obtained are in good agreement with experimental reported ones.

[NP1.020] Electron Cyclotron Resonance (ECR) Plasma Thruster Research

A study is being made on an electric propulsion system (EICR Plasma Thruster) which can generate plasma with ECRH (Electron Cyclotron Resonance Heating), accelerate ions with ICRFH (Ion Cyclotron Range of Frequency Heating) via antenna, and adopts a gradient in magnetic field to obtain thrust. The plasma thruster could achieve high power density and long lifetime since this system does not use any electrodes and grids[1].

[NP1.021] Rf Power Absorption in a Short Dense Plasma Column Excited by Various Frame Antennas

[NP1.022] Antennas in Magnetized Plasma Shells

Electrodynamic parameters of a short (1 m length) electric antenna surrounded by a magnetized plasma shell were studied in the frequency range 100 kHz - 2 MHz. It was shown that at certain optimal plasma parameters in the vicinity of the lower-hybrid resonance the efficiency of radiation of the plasma antenna system into the ambient space (and, correspondingly, the efficiency of its reception) was three to four times as high as in the case of no plasma shell present.

[NP1.023] Plasma Rotation and Density Profile Control by Voltage Biased Electrodes in RF Produced Magnetized Plasma

Plasma rotation has been studied in various fields such as plasma application, space plasma, and nuclear fusion related to the enhanced magnetic confinement due to the velocity shear. Controlling the density profile is very important in magnetic confinement devices and plasma sources in the application fields. However, there have been few experiments to demonstrate large changes of density and rotation profiles from a basic viewpoint [1]. Here, we report the control of these profiles, associated with the instabilities, using ten concentric circular rings as biased electrodes in a cylindrical RF produced plasma [1], changing magnetic field configurations (straight field, good and bad (mirror) curvatures). With the positive voltage biasing to the inner (outer) electrodes, a hollow (peaked) density profile was obtained. The azimuthal plasma rotation and high velocity shear in a supersonic regime could also be controlled. Low frequency (3-4kHz) density fluctuations, opposite rotating direction to the edge plasma (ion diamagnetic one), were identified as a drift wave type. [1] S. Shinohara et al., Surf. Coat. Technol. 112 (1999) 20 CJpn. J. Appl. Phys. 38 (1999) 4321.

[NP1.024] RF Wave Propagation in Bounded Plasma with Various Magnetic Field Configurations

RF wave with a frequency between ion and electron cyclotron ones is known as a whistler wave, and spontaneous excitation of this wave by lightning was observed historically and it has been utilized, e.g., as a helicon wave (bounded electromagnetic wave) in the plasma application field. Studies on the RF wave characters under the various magnetic field configurations have been done, concerning the relation between high density plasma production and wave propagation and damping [1]. In order to clarify the detailed wave characteristics in bounded plasma, RF test wave propagation was studied in two-dimensional space, changing the magnetic field configuration (straight, divergent and convergent fields) and the excitation loop antenna diameter (10 and 30 cm) in the afterglow RF produced plasma, 45 cm in diameter and 170 cm in axial length [1]. Numerical calculation results by the finite element method showed good agreements with the observed ones such as the amplitude and phase of three components of the magnetic fields, wave velocity and propagation angle, including standing waves for the low collision case. [1] S. Takechi, S. Shinohara et al., Surf. Coat. Technol. 112 (1999) 15 C Jpn. J. Appl. Phys. 38 (1999) 3716 and L1278.

[NP1.025] RF-Plasma in a Simple Magnetised Torus

The simple magnetised torus (SMT) is a widely used experiment to investigate plasma instabilities and turbulence. The magnetic field is purely toroidal. It was shown, that in such configuration no MHD equilibrium can establish. However, a steady state plasma can be established in a SMT. The common method to produce plasma is a directly heated tungsten wire. The plasma is strongly turbulent in all plasma parameters.However, the heated filament acts as a charge source, the surplus of electrons is transported to the walls. To obviate this effect, wave heating technics can be used. In the TEDDI device inductive resonant RF heating via a helicon antenna is used to produce a plasma. The plasma is steady state. The equilibrium conditions of the RF-plasma are compared to the case of the heated filament. The plasma potential is an order of magnitude higher, leading to a faster ExB-rotation. The possibility of a magneto-electric confinement of the plasma is discussed and the influence of poloidal limiters with different diameter is investigated for both plasma sources.

[NP1.026] Use of Langmuir Probes in High Density Plasmas

[NP1.027] Langmuir Probe Measurements in Low Temperature, Multipolar Plasmas of an Electron Cyclotron Resonance Microwave Ion Source

[NP1.028] Electron-Temperature Control for High-Quality Diamond Formation in a Low-Pressure RF Plasma

[NP1.029] Ionization-driven RF Repetitive Crafted Pulses in a LF Oscillating Plasma

[NP1.030] Low Temperature Plasmas sustained by Internal RF Currents

Power and particle balance in a plasma-filled cylindrical resonator with internal RF currents of different configurations have been studied. RF currents with spatially varying and spatially constant phases have been introduced. The electromagnetic field structure and power density have been investigated invoking electrodynamic and plasma fluid equations. Investigation of power deposition shows that at low plasma densities, the total power deposited onto the plasma raises with plasma density, while at higher densities it starts to decline. It has been demonstrated that conventional diffusion model with uniformly distributed ionization source over the cross-section accurately describes the plasma density profiles in a plasma cylinder. The diffusion model reveals that the electron temperature is an increasing function of plasma sheath's width, while the effective diffusion length is a decreasing function of it. Power and particle balance equations have been used to study the operating parameters of low temperature plasma and their dependence on the input power and gas pressure. It is shown that, introducing the RF currents with spatially varying and spatially constant phases one can generate high-density low temperature plasma with high degree of uniformity.

[NP1.031] Diffraction Effects in Helicon Plasmas

[NP1.032] Two Dimensional Electric Potential with Highly Anisotropic Electron Mobility.

In Hall thrusters electrons are subject to crossed electric and magnetic fields. They undergo collisions in the bulk and at the dielectric walls.

[NP1.033] Atomic Data Requirements for Stationary Plasma Thrusters (SPT).

In the frame of the French Research Group CNRS/CNES/SNECMA/ONERA N0 2232 "Propulsion/Plasma pour Systèmes Spatiaux "; the Laboratoire de Physique des Gaz et des Plasmas participates to the SPT study, being involved so far with spectroscopic measurements, sputtering and modeling concerning the SPT 100-ML prototype and the PIVOINE facility described elsewhere. A description of these and other prototypes together with an overview of the French research and development program in the field of electric propulsion has been recently given (A.Cadiou, M.Lyszyk, M. Dudeck, Research and Development on Plasma Thrusters in France 26th IEPC'99).

[NP1.034] DSMC Neutral and Charged Particle Simulation of a Kaufman-type Ion Thruster

The high specific impulse of the Kaufman-type ion thrusters and their reliability make them well suited for near-earth and deep-space missions. However, due to the complexity of charged/neutral particle transport in these devices, the detailed properties of the ion thruster are not clearly understood. The plume has been extensively studied unlike the phenomena in the main chamber where complex phenomena are suspected to occur. Thus, our particle simulation will investigate the processes involved in an ion thruster designed and tested by the DERA, UK.

[NP1.035] Experimental and Model Study of the DC Discharge in Cylindrical Magnetron in Argon

Our cylindrical magnetron consists of a discharge chamber - anode (110 mm in length, 60 mm inner diameter) and the co-axially positioned cathode (10 mm in diameter). The magnetic field is created by a couple of coils (adjustable from 10 to 40 mT).

[NP1.036] Low Energy Hydrogen amp; Methane Ion Beam Plasmas with Low Electron Temperature

Electrotechnical Laboratory, 1-1-4 Umezono, Tsukuba-shi, 305-8568, Japan-High current low energy convergent ion beam plasmas, whose energies are 50-200 eV, electron temperatures are less than 1 eV and the parameters are controllable, have been developed for the plasma processing. The features of the hydrogen and methane ion beam plasmas are presented. The ion beams are extracted by multi-aperture Molybdenum electrodes of 7 cm x 7 cm. To eliminate the ion space charges, electrons are added to the ion beam, and then the ion beam plasma is attained. The total beam currents of 150 eV of hydrogen and methane ion beams are 170 mA and 100 mA, and the efficiencies are 70 % and 60 %, respectively. The energy profiles of the beam are measured by an energy-mass analyzer. H_3^+ in the hydrogen ion beam and CH_3^+ in the methane ion beam are main ion beam components. The main beam energy of each mass (m/e) is the extracted voltage, but, the energy spectrum has side peaks around the main energy peak. The dissociation, absorption, charge-exchange occur during traveling of 1 m in gas pressure of 0.1 mTorr, and then, the side peaks appear. The mechanisms and the control of them will be presented.

[NP1.037] Instability of a Vacuum Arc Centrifuge

[NP1.038] X-RAY Emission in a Pulsed Vacuum Spark

The soft and hard X-ray emission produced by an electrical discharge in high vacuum is reported. The pulses are produced by a fast 1 kJ line pulser having peak current of the order of tens of kiloamp. The diagnostics employed included time-resolved X-ray detection using pin diodes, voltage monitoring through capacitive probes and current monitoring with a Rogowski coil. Pairs of brass and stainless steel Borda electrodes, well known for their field uniformity, were employed in the tests. The X-ray pin diodes employed were fitted with filters that selectively block the radiation from one of the two electrodes. It is found that when using one each brass and stainless-steel electrodes, the radiation does not vary much if the electrodes are exchanged, meaning there is no preferential sublimation from one of the two. (Electrode sublimation provides the milieu in which the arc grows). Time-resolved detection shows that soon after current start the emission is of soft nature. Much later into the discharge, several microseconds later, the radiation is of a harder nature, thought to originate in arc interruption.

[NP1.039] Characteristics of Plasma Jet by using the Modified Pseudospark Discharge

Pseudospark discharge (PSD) is one of the low-pressure gas discharges. In PSD, the parallel plane electrodes are replaced by that with a circular hole on the axis. This cathode has a cylindrical cavity behind the circular hole. PSD can be formed by using such modified electrodes in the low-pressure region under the Paschen minimum. PSD is characterized as follows: (1)Discharge is formed through the holes of electrode. (2)Breakdown voltage is a function of plasma pressure. (3)Since the hollow cathode effect, high current glow discharge can be formed. (4)Damage of electrode is low. We have been studying the plasma jet by using the modified PSD. In this plasma jet, diameter of anode hole is widened in comparison with that of cathode hole. The force of plasma acceleration is Lorentz force, which is similar to that of MPD arc jet. This plasma jet is characterized as follows: (1)Plasma is spouted at high speed (2)High temperature plasma will be formed due to the high current discharge. (3)Since the damage of electrodes is low, lifetime of the device is long. (4)Strong magnetic field with the plasma jet generates. This strong magnetic field may be possible to transport with plasma jet. In this poster, the space distribution of plasma jet and lifetime of electrode are reported in detail.

[NP1.040] Measurements of Electron Density and Temperature of Micropinch Plasma Formed in Vacuum-Spark Discharge

The parameters of micropinches formed in vacuum-spark discharge are investigated spectroscopically. Single shot x-ray spectra of He-like S, Cl, Ca, Cr, Fe and H-like Ca ions are recorded using crystal spectrograph. The comparison of experimental spectrum with the corresponding theoretical one, generated by codes FLY and FLYSPEC is used to determine the electron density ne and temperature Te of micropinch plasma.

[NP1.041] Formation of Double Layer in Multicomponent Plasma with Negative Ions

[NP1.042] Recombination of H3+ Ions with Electrons at Thermal Energies - Pressure Dependence of the Recombination Rate Coefficient

The literature values for the rate of recombination of the H3+ (H_3^+)ion with electron show considerable discrepancies. In order to contribute to this problem we built for the study of H3+ recombination an experimental system that we call the Advanced Integrated Stationary Afterglow (AISA).

[NP1.043] Plasma Conductivity and Ionization Growth in Flame Breakdown

An investigation into the properties of flame breakdown is reported. A series of DC discharge tests were performed in a set of parallel plane electrodes bridged by flames from a bunsen burner. The experimental setup aims to reproduce the conditions found in waste-disposal reactors where the combined effect of fire and an electrical arc degrade noxious substances. The current was simultaneously monitored in different points of the discharge zone. As the applied voltage is increased, it is found that initially the ionization from the flame controls discharge growth but that in later stages avalanche growth takes over. The slope of the I-V characteristics was used for estimating the Townsend ionization coefficients. The overall plasma conductivity was estimated from both the external circuit measurements and the plasma parameters. The results obtained are compared with previous investigations in which mean discharge resistivity is a relevant parameter, employed for designing applications. The effect of gap separation and height over the burner top were also analyzed. This way it was observed that the temperature profile of the flame dictates the spatial distribution of electrical conductivity and thus of breakdown.

[NP1.044] Self-Organization and Reconnection (DPP)

This abstract not available.

[NP1.045] Spontaneous 3D Magnetic Reconnection in a Unbounded EMHD Plasma

A 3D reverse-field configuration (dipole opposing a uniform field) is established in a large, laboratory plasma (1~m diam, 2.5~m length, B_0 = 5~G, n_e \leq 10^12~cm^-3, kT_e \leq 3~eV). The dipole field, generated with a pulsed loop antenna, is suddenly (\Delta t \leq 1~\mus) switched off and the relaxation of the imposed field is measured in 3D space and time. On whistler time scales (\Delta t \simeq 5~\mus), the separatrix with 3D null points is maintained by plasma currents free of any boundary currents. The dipole field is not unstable to tilting. During the relaxation, closed field lines are spontaneously reconnected to open field lines until the null points coalesce and vanish. The reconnection rate d \Phi_closed / d t is measured. The initially stored magnetic energy is dissipated and convected in the whistler mode. Electron heating and light emission are observed.

[NP1.046] Formation of an EMHD FRC by Merging of two EMHD Vortices

Two loop antennas are inserted in a uniformly magnetized high-beta laboratory plasma (1~m diam, 2.5~m length, n_e \leq 10^12~cm^-3, kT_e \leq 3~eV, \beta \approx 1). The loops are 25~cm offset along the uniform field B_0 = 5~G and produce two reverse-field configurations each with two 3D null points. The dipole fields are simultaneously switched off and the relaxation of the two EMHD vortices is measured in 3D space and time. The vortices merge and form a 50~cm long field-reversed configuration (FRC). It is produced by a cylindrical electron current sheet driven solely by internally stored magnetic energy. In the absence of any boundary currents it remains stable to tilting. The FRC relaxes on EMHD time scales but slower than an individual vortex. The energy is dissipated into the electrons.

[NP1.047] Evolution of 3D Magnetic Null Points in EMHD Plasmas

During the afterglow of a laboratory discharge plasma (1~m diam, 2.5~m length, B_0 = 5~G, n_e \simeq 10^12~cm^3, kT_e \simeq 2~eV). Two field reversed magnetic vortices are created with two pulsed loop antennas with axes antiparallel to B_0 = B_0\hatz. Each vortex produces two 3D magnetic null points (B_total = 0) and a nearly spherical seperatrix surface. The two vortices expand along B_0 and merge into a single field reversed configuration (FRC). During the collision, two simple null points come together and create a single degenerate 3D null point. A Taylor series expansion of the measured magnetic field near this degenerate null point yields negligible linear terms which classifies it as second order. The degenerate null point is unstable and evolves into a circular null line. Coalescence of X and O-type null lines produces the cylindrical current sheet of the FRC. This poster addresses the time evolution of the B-field topology and its relevance to 3D magnetic field line reconnection.

[NP1.048] Study of Magnetic Reconnection Region on MRX

It has been recognized that conventional MHD theories often break down in the reconnection layer, while globally the reconnecting plasma is well approximated by MHD equations. Precise measurement of the profile of the reconnection layer can provide important clues to help understand the non-MHD physics mechanisms of reconnection. The detailed structure of the neutral sheet has been measured in the MRX plasmas [M. Yamada et al., Phys. Plasmas 7, 1781 (2000)]. In agreement with Harris theory [E. G. Harris, Nuovo Cimento 23, 115 (1962)], the width of the reconnection region is on the order of ion skin depth as a simple MHD formulation breaks down in this regime. By treating electrons and ions separately, the generalized Ohm's law describes force balance of an electron flow, E + v_e \times B = \eta_s j + (m_e/e^2) ( \partial v_e / \partial t + v_e \cdot \nabla v_e ) - \nabla P_e/en. In the neutral sheet we expect that the fluctuation components of v_e \times B and v_e \cdot \nabla v_e can make a significant contribution to the force balance and the energy dissipation rate thus affecting the reconnection speed. Recently electrostatic and electromagnetic high-frequency fluctuations have been singled out and measured in MRX. The wave frequencies are between the ion cyclotron and lower hybrid frequencies. Our next goal is to experimentally determine dispersion relations of the fluctuations and to find correlation of these fluctuations with the enhanced resistivity and ion heating. The impact of the MRX data on interpretation of the recent space data will also be presented. Work jointly supported by DoE, NASA, and NSF

[NP1.049] Study of Electric and Magnetic Profiles of Current Sheet in Magnetic Reconnection Experiment

Magnetic reconnection plays an important role in determining the topology of magnetic fields in solar flares, magnetospheric substorms, and relaxation processes in laboratory plasmas. A major puzzle of magnetic reconnection is the observed reconnection rates, which are much faster than predictions by classical theories such as the Sweet-Parker model. In MRX (Magnetic Reconnection Experiment) [Yamada et al., PRL 78, 3117 (1997)], it has been observed [Ji et al., PRL 80, 3256 (1998)] that reconnection happens much faster in the collisionless regime than classically predicted rates and that it is accompanied by strong non-classical ion heating [Hsu et al., PRL 84, 3859 (2000)]. Important clues for the underlying mechanisms may be obtained by precise measurements of both electric and magnetic field profiles using a 29-channel pick-up coil array and a 17-channel Lagmuir probe array with 5~mm resolution, compared to a full width of 3--5~cm for the current sheet. It has been found that the magnetic field profile fits very well to the hyperbolic tangent function, as predicted by Harris' theory [Harris, Il Nuovo Cimento 23, 115 (1962)]. In contrast to Harris' theory, however, a significant in-plane electric field has been detected inside the current sheet. A simple two-fluid theory has been constructed to give a nonlinear differential equation predicting both electric and magnetic field profiles. Detailed measurements and comparisons with theory will be presented. This work is supported by DOE, NASA, and NSF.

[NP1.050] Measurement of electrostatic and magnetic fluctuations in a reconnecting current sheet

Magnetic reconnection is an ubiquitous process which plays an important role in both space and laboratory plasmas. Measurements in the Magnetic Reconnection Experiment (MRX) reveal fast reconnection [H. Ji et.al., PRL 80, 3256 (1998)] and anomalously high ion heating rates [ S.C. Hsu et.al., PRL 84, 3859 (2000)]. Current sheets formed in MRX involve large gradients in density, flow, and magnetic field, all of which can drive unstable fluctuations. Turbulence resulting from these free energy sources could explain the observed anomalous dissipation and ion heating. Double Langmuir and magnetic pick-up probes utilizing wide bandwidth amplifiers built into the probe tips have been used to measure electrostatic and magnetic fluctuations in null-helicity MRX current sheets. We have measured electrostatic fluctuations in the 1-10 MHz range (Ømega_i \approx 300~kHz and ømega_LH \approx 13~MHz) which arise with the formation of the current sheet and have a hollow amplitude profile. Magnetic fluctuations are observed in the 500-800 kHz range, are strong near the end of the reconnection phase, and are peaked in amplitude at the current sheet. Wavelength spectra and possibly dispersion relations will be obtained using multipoint probes. We will present detailed characterization of these fluctuations including any correlation to the reconnection rate.

[NP1.051] Two-Dimensional Structure Measurement in the Magnetic Reconnection Experiment with Planar Laser-Induced Fluorescence

Rapid rates of magnetic reconnection observed in solar flares, the earth magnetosphere and laboratory experiments cannot be explained by classical theory. Several possible dissipation mechanisms, some involving wave-particle interactions, have been proposed in order to explain the rapid reconnection and associated anomalous resistivity. Spatially resolved measurements of fluctuations are necessary to test these theoretical predictions. Planar laser-induced fluorescence (PLIF) is a high-resolution non-invasive technique for visualizing turbulent structures in 2D widely used in fluid dynamics studies, and recently applied to plasmas. We report the first results of PLIF measurements in the Magnetic Reconnection Experiment. The diagnostic setup consists of a tunable pulsed laser, optics for the creation of a sheet beam, and a gated ICCD camera with an interference filter. An image produced with a single short laser pulse provides a 2D map of relative metastable argon ion density in the radial cross-section of the axisymmetric neutral sheet of MRX discharge with the spatial resolution \approx 2 mm. The high resolution and short pulse duration allow the detection of spatial fluctuations of ion density. The data is expected to shed light on the fundamental physics properties of the reconnection region.

[NP1.052] Reconnection Scaling Experiment (RSX) – 3D movies of magnetic reconnection in linear geometry at Los Alamos National Laboratory

The physics of magnetic reconnection is a major unsolved issue in MagnetoHydroDynamics (MHD) and plays a fundamental role in changing the magnetic field topology for many astrophysical and laboratory plasmas. Magnetic flux annihilates and transforms into plasma kinetic energy beyond the scope of ideal MHD. We are presently building the linear Reconnection Scaling Experiment (RSX) at LANL to access the scaling between collisional and collisionless reconnection regimes. Plasma gun technology developed at the Univ. of Wisconsin generates high density (>10^14cm^-3) high current (J\sim300A/cm^2) ohmically heated (T_e>15eV) channels. As the reconnection region sweeps down a 4 meter plasma column we create 3D movies of magnetic reconnection from many repetitive shots. We will attack problem with the experiment, modeling using 3D fluid (NIMROD) and particle simulations of the reconnection region. This approach represents synergistic collaborations across divisions within the LANL community and has a substantial student participation.

[NP1.053] Magnetic reconnection studies on SSX

Three-dimensional magnetic field measurements of the reconnection layer in the SSX experiment are underway. Magnetic probes (B_r, B_\theta, B_z) are arranged in a 5 by 5 by 8 array with 2 cm resolution (600 separate measurements on a single shot). Probe signals are integrated and multiplexed at 10 MHz in groups of 8 so the effective digitization rate is 1.25 MHz. At the same time, energetic particle flux is monitored by sets of identical retarding grid energy analysers (calibrated in a separate cw plasma facility). Finally, temperature and density are monitored with a combination of VUV spectroscopy, laser interferometry, soft x-rays, bolometry and electrostatic probes. Plans for an FRC formation experiment will also be discussed.

[NP1.054] Resistive MHD and particle trajectory simulations on SSX

Formation, relaxation and reconnection of spheromaks in the Swarthmore Spheromak Experiment (SSX) are modeled by a 2-D axisymmetric resistive MHD simulation (TRIM) applied to the SSX boundary and initial conditions. Results are compared to corresponding one- and two-dimensional measurements of magnetic field in the experiment. Both global equilibrium and reconnection zone B-field configurations are reproduced. Several different resistivity implementation schemes and equations of state are attempted. MHD simulation data is used as input for a particle trajectory and energization code. Good agreement between the particle simulation results and corresponding energetic particle data from SSX is observed.

[NP1.055] Investigation of ion heating mechanism of Magnetic Reconnection by Plasma Merging Experiment

The ion heating effect of 3-D reconnection was investigated experimentally. Three-component reconnection effect has been investigated experimentally by varying the external force and the magnetic field component Bt parallel to an X-line. Merging of two plasma toroids in the TS-3 device was used to form a thin current sheet with width of 5-10cm. The direct ion heating by reconnection was observed in those experiments. Ion temperature was found to increase with decreasing toroidal magnetic field toward zero. To investigate the reconnection heating mechanism ,we developed 2-D ion temperature measurement system using Doppler effect. The system is composed of an array of Optical Fibers, a Monochromator and a CCD camera with intensifier. 2-D profile of Ion temperature around reconnection point was measured by using CT(computer tomography).

[NP1.056] Integrability Aspects of a Dynamical System involving Current-sheet Formation in MHD

Integrability aspects of a dynamical system \footnote B.K. Shivamoggi, Phys. Lett. A 258, 131 (1999) involving current-sheet formation near a magnetic neutral line are considered. A study of the singularity structure of the system in question is made and the Painleve property is tested.

[NP1.057] Electron-Inertia Effects on Magnetic Field Reconnection Driven by Perturbations on Boundaries

We consider electron-inertia effects on the magnetic field reconnection in a plasma by perturbing the boundaries of a slab of plasma with a magnetic neutral surface inside. The subsequent evolution of the current sheet formed at the magnetic neutral surface via ideal MHD development is discussed.

[NP1.058] Transient Magnetic Reconnection and Unstable Shear Layers

Results are presented from a study of magnetic reconnection caused by unstable solar wind flow past the Earth's magnetosphere. Flow conditions are considered similar to those in the magnetopause boundary layer. Specifically, subsonic and and sub-Alfvenic flow is considered, which is stable in the direction of the magnetic field, but unstable to the Kelvin-Helmholtz instability perpendicular to the magnetic field. The flow is modeled by the resistive MHD equations in three dimensions with constant resistivity. The equation system is solved using a particle-in-cell method with low computational resistivity. Localized transient reconnection is observed with many of the features associated with Flux Transfer Events, including reversal of the helicity across the equator. Reconnection is shown to increase with decreasing resistivity.

[NP1.059] Comparison of hybrid and Hall-MHD simulations of reconnection

Understanding the field structure and plasma properties near the X-point of magnetic reconnection in a nearly collisionless plasma, such as the magnetotail, remains an important problem, as the recent GEM simulation challenge indicates. Here we compare quantitatively the results of two-dimensional hybrid (particle ions, massless fluid electrons) and Hall-MHD simulations of a thin current sheet, using a Harris equilibrium and a localized resistivity to initiate the reconnection. As in previous simulations, we find comparable reconnection rates in the two simulations, but different plasma properties; e.g., complex non-gyrotropic ion distributions in the hybrid simulations. We carry out a detailed analysis of the plasma and field properties in the two calculations, which are done on the same spatial grid, to compare the electric field in the reconnection region and the resultant forces on the plasma.

[NP1.060] Numerical Study of Co- and Counter-helicity Hall Reconnection

Detailed numerical studies of co- and counter-helicity magnetic reconnection in merging spheromaks have been performed with a parallel implementation of a 2D two-fluid resistive MHD code. We find that the presence of the Hall term can dramatically increase the reconnection rate over that obtained in a single-fluid code with the same resistivity and viscosity. In fact, for reasonable parameters, the presence of the Hall term causes the scaling of the reconnection rate to become independent of the value of the resistivity. This effect is explicable in terms of electron MHD, from which it can be shown that the Hall term causes the geometry of the reconnection region to open from a cusp-like structure to a large angle. The whistler wave dynamics introduced by the Hall term present stringent numerical requirements that are stabilized by the use of an ADI scheme and a fourth-derivative ``hyper-resistivity'' term. Convergence studies verify that the accelerated reconnection observed in the simulations is physical and not numerical in origin.

[NP1.061] Multi_Scale Plasma Simulation of Magnetic Reconnection

In plasma simulation, there are many problems in which electrons contribute to phenomena on ion time and length scales, including magnetic reconnection. Such problems are difficult to simulate accurately because of the wide range of time and space scales that must be resolved with realistic ion/electron mass ratios. Even using implicit plasma simulation methods, in which one solves the full kinetic equations with a time step that is unconstrained by numerical stability, only those time scales that are resolved by the time step are modeled accurately. If one chooses a large time step commensurable with ion time scales, one cannot simultaneously resolve the electron time scales. Only by choosing a time step so small that one resolves all scales can one capture the full physics of electrons and ions, but then the computational effort required may be prohibitive. However, as in magnetic reconnection, the important electron contributions occur in a small subdomain, and only there is a smaller time and space step required. Here we describe the use of adaptive mesh refinement (AMR) in conjunction with the implicit moment method for plasma simulation in two and three dimensions to treat problems of this type. The advantage of this approach is that the same equations are solved at every level of refinement, but the overall solution includes a greatly expanded range of time and space scales compared with those obtained on a uniform mesh. Application to the computation of the evolution of a Harris equilbrium illustrates the technique and the promise of this approach.

[NP1.062] Fast electron reconnection

Fast magnetic reconnection driven by the electron fluid flow through the immobile neutralizing ionic background occurs in the whistler frequency regime (electron MHD). In this regime the plasma can be considered in many physical cases as collisionless or weakly collisional and electron inertia is the major effect leading to reconnection of the magnetic field lines. This regime of magnetic reconnection has quite different underlying physics with respect to the standard MHD picture and can be relevant in the laser plasma interaction. Moreover, it appears suitable for a numerical study of the reconnection instability from the kinetic viewpoint because of its single species nature, allowing us to resolve only the electron time and length scales. In this work the tearing-like fluid derivation, which is limited to the region of validity of the so-called constant-psi approximation, is generalized and the full linear spectrum of the electron MHD reconnection instability is studied in slab geometry. Charge-separation effects are important in the electron MHD reconnection, and lead to the stabilization of the mode in a strongly magnetized plasma via the skin depth renormalization in the poloidal plane.

[NP1.063] On Magnetic Reconnection in a Turbulent MagnetoFluid

Recently, there has been renewed interest in the problem of magnetic reconnection in a turbulent magnetofluid^1. Most models of such a process (see (1), for example), tacitly assume that the turbulence can be given or specified ab-initio, and that reconnection will proceed in the presence of this turbulence without significantly modifying it. Here, we examine the simple Sweet-Parker problem in the presence of turbulent hydrodynamic forcing, but with magnetic fluctuations computed self-consistently via the conservation of mean-square potential (2D) or magnetic helicity (3D). This conservation law introduces important modifications to the conventionally invoked "turbulent resistivity" and "electron viscosity" (aka hyper-resistivity). In 2D, results indicate that Sweet-Parker scaling of the reconnection speed is modified only by a factor of the turbulent Alfvenic Mach number, but that the resistivity scaling remains unchanged. Work on 3D is ongoing and will be discussed.

^1 A. Lazarian and E. Vishniac, ApJ 517, 700 (1999).

[NP1.064] Radial and Poloidal Dynamics of Turbulent Transport in a Self Organized Criticality Model with Sheared Flow

SOC models have been used to describe the dynamics of the transport without relying on the underlying local fluctuation mechanisms and to explain the profile resilience and universal transport behavior observed in many devices. Computations based on a cellular automata model have found that SOC in the absence of sheared flows produces dominant transport scales on the order of the system size rather than on the order of the underlying local fluctuation scales. However the addition of sheared flow to the dynamics leads to a large reduction of the system scale transport events and a commensurate increase in the fluctuation scale transport events needed to maintain the constant flux. This work looks at the radial correlation lengths and the PDF’s of the transport event size both within and outside the sheared region. Additionally we investigate the effect of the shear on the poloidal correlation lengths of these events. The potential ramifications for transport studies will be discussed.

[NP1.065] A Transition in the Transport Dynamics of a SOC System Coupled to Classical Diffusion

The concept of Self-Organized Criticality (SOC) has been advanced as a paradigm[1,2] for turbulent transport in magnetically confined plasmas. In a realistic confined plasma system the transport dynamics is likely to be a combination of coherent anomalous transport (SOC) and a more classical diffusive component. To investigate the interaction between these 2 components, a diffusive term is added to a SOC model and the dynamical regimes as a function of the drive and classical diffusivity are explored. The inclusion of the diffusive term allows access to a regime in which the bursty behavior characteristic of the SOC dynamics is reduced and eventually eliminated. In the intermediate regime both the dynamical character and average dimension of the avalanches change. The implications and possible tests for transport dynamics in various experimental regimes and devices are discussed. 1 P. H. Diamond and T. S. Hahm, Phys. Plasmas 2 (10), 3640-3649 (1995). 2 D. E. Newman, B. A. Carreras, P. H. Diamond et al., Phys. Plasmas 3 (5), 1858-66 (1996).

[NP1.066] Chaotic Scattering and Self-Organization in Spheromak sustainment

The formation and sustainment of a flux core spheromak by electrostatic helicity injection is studied by numerical computation using the resistive MHD equations. An idealized geometry is used in which a cylinder of finite length has magnetized electrodes at the ends, which are at a potential difference V. A critical value V=V_c is reached above which the central column is unstable to a line-tied kink mode. The self-organized nonlinearly saturated state with V just above V_c has a large toroidal volume of flux surfaces, with rotational transform provided by the helical distortion of the central column (stellarator transform). As V is increased further the column becomes more strongly kinked and the tori are destroyed. In both the low and high V cases the field lines exhibit chaotic scattering, showing the usual fractal characteristics of nonhyperbolic chaotic scattering if some invariant tori exist. The distribution of field line lengths L, a diagnostic of chaotic scattering, is studied because of its relevance to confinement and average parallel current density along the field lines. At larger values of V or larger Lundquist number S, a limit cycle appears with almost no invariant tori.

[NP1.067] Self-consistent chaos and coherent structures formation in plasmas and shear flows.

Self-consistent dynamics is studied in the context of the single wave model (SWM), and a self-consistent standard map (SCSM). The SWM is a general, mean-field model that describes the weakly nonlinear dynamics of marginally stable systems interacting through long range forces. The two systems of interest here are Vlasov-Poisson plasmas and shear flows. The SCSM is obtained from a space-time discretization of the SWM, and it corresponds to a standard map globally coupled to a self-consistent mean-field (D. del-Castillo-Negrete, Phys. Lett. A, 241), 99 (1998); Phys. Plasmas, 5, 3886 (1998); CHAOS, 10 (2000); Physica A, 280, 10 (2000). Using these models we study self-consistent resonance overlap, and the role of self-consistent Hamiltonian chaos in the formation of coherent structures.

[NP1.068] Self-Consistent Lagrangian Chaotic Transport in Shear Flows

We present direct numerical simulations on the nonlinear evolution in shear flows instabilities when there are two linear instabilities of differing phase velocities, leading to two chains of islands or 'cat's eyes'. For a symmetric vorticity profile with two peaks, the time asymptotic state consists of two island chains propagating relative to each other. More unstable vorticity profiles lead to larger islands, which can overlap. The typical case has a region of good KAM surfaces near the centers of the islands, where the vorticity is constant on the surfaces of the wave frame streamfunction (BGK modes), and an surrounding area of Lagrangian chaos where the vorticity is flat. The self-consistent influence of the chaotic transport on the amplitude and propagation of the nonlinear waves will be presented. For the antisymmetric case in which the two peaks of the vorticity have opposite signs, the linear modes travel with the same phase velocity, and for some parameters the nonlinear state is a steady reconnected state much like a saturated double tearing mode. For other parameters the time asymptotic state is a limit cycle, and there are good KAM surfaces only in the centers of the reconnected islands. The dependence of the characteristics of the time asymptotic state on the Lagrangian chaos is explored. Since the chaos in both cases is due to a streamfunction that is time dependent but typically not periodic, the 'patchiness' diagnostic of Malhotra, Mezic, and Wiggins [Int. J. Bifurcations and Chaos v. 8 pp. 1053-1093] is used.

[NP1.069] Effect of small dissipation on Twist Maps

Any physical problem treated in the Hamiltonian system should have certain dissipation in actual physical systems. Because the boundary between the Hamiltonian and dissipative systems is strongly discontinuous, if a small dissipation is introduced into the Hamiltonian system, the trajectories should converge to attractors in the dissipative system no matter how small the dissipation is. This discontinuity of the boundary between the Hamiltonian and dissipative systems is examined by investigating the behavior of trajectories, making use of simple two-dimensional twist maps such as drift orbit map in a helical magnetic field similar to the dissipative standard map. With a small dissipation, trajectories in the Hamiltonian systems may change to filamentary set of attractors which are determined as the stable fixed points in the Hamiltonian system. The larger the dissipation is, the faster the convergence becomes. Concentration of trajectories to attractors will be shown by two dimensional basin graphics on the initial phase space.

[NP1.070] Variational Principle for Contour Dynamics (Waterbag) Models

The Vlasov-Poisson system and the two-dimensional Euler equation posses reductions where the dynamics is reduced to that of boundaries of regions of constant phase space density and vorticity, respectively. The Vlasov-Poisson system has solutions that describe nonlinear plasma oscillations, while Euler's equation has special solutions of uniformly distributed vorticity called V-states. We write down the Hamiltonians for these systems and then insert trial functions that possess parameters that approximate the plasma oscillation and V-state solutions. Then, upon variation we obtain approximate as well as exact solutions. Properties of these solutions are examined. For Euler's equation we establish the existence of critical V-states, a fact that was previously only based on numerics. Similar results for Vlasov-Poisson system are discussed.

[NP1.071] WeaklyNonlinear Hamiltonian Dynamics of Vlasov-Poisson Transients

Adapting techniques from finite-dimensional canonical perturbation theory, we study weakly nonlinear dynamics of longitudinal disturbances near stable, homogeneous Vlasov-Poisson equilibria. We derive the continuous analogue of a resonance Hamiltonian that accounts for the lowest order interactions among the modes of the continuous spectrum. The equations generated by this Hamiltonian describe the nonlinear behavior of transient phenomena, often ignored in weakly nonlinear theories of longitudinal waves. We analyze these equations to determine to what extent such neglect is justified.

[NP1.072] A Lagrangian approach to the study of the kinematic dynamo

The problem of the evolution of a passive magnetic field embedded in a chaotic flow---the kinematic dynamo problem---is of great relevance to planetary, solar, and astronomical physics. Progress can be made by transforming to Lagrangian coordinates, which move with the flow. The local growth rates of the magnetic field and of the induced current can then be estimated in terms of the finite-time Lyapunov exponents of the flow, which measure the instantaneous exponential rate of divergence of neighboring trajectories. A numerical method is presented to calculate the Lagrangian derivatives of the magnetic field, which allows comparison of the magnetic energy to the power needed to sustain its growth. It is found that the leading-order growth of the power is less than previously estimated [1], because of constraints on the asymptotic evolution of the flow arising from differential geometry [2].

[1] A.H. Boozer, Astrophys, J. 394, 357 (1992); X. Z. Tang and A. H. Boozer, Phys. Plasmas 7, 1113 (2000)

[2] J.-L. Thiffeault and A. H. Boozer, submitted to Chaos (2000)

[NP1.073] Transition from turbulent to laminar regimes in the reversed field pinch (RFP)

The RFX experiment has shown the possibility for the RFP to develop a non-chaotic region in the core as diagnosed by soft-X ray tomography. These states correspond to a quasi single helicity (QSH) magnetic spectrum where one mode m=1, n=n_sh dominates the others.This motivates the theoretical study of the pure single helicity (SH) states of the RFP which correspond to a laminar dynamo produced by a single mode: these are characterized by a quasi integrable magnetic field, a feature favourable to good confinement. Simulations of visco-resistive MHD reveal a bifurcation from SH to multiple helicity (MH) regimes ruled by the Hartmann number. A continuous change of the order parameter and temporal intermittency are observed suggesting an analogy with a second order phase transition. These SH configurations are compared with helical ohmic states of the RFP obtained by solving the Grad-Shafranov equation. A mechanism of magnetic chaos healing is shown to exist when the magnetic separatrix of the dominant mode of QSH states disappears due to a saddle-node bifurcation.

[NP1.074] Fourier-Beltrami Analysis of Dynamo Magnetic Field

We performed a numerical analyses of the kinematic dynamo field based on the Fourier-Beltrami expansion technique. Since Beltrami function, which is the eigenfunction of curl operator, forms a complete set for the divergence free vector field, we can uniquely decompose any magnetic field into the positive and the negative helicity field using this technique. The objective of this work is to study the characteristic structure of the magnetic helicity generated by dynamo action. We first solve the kinematic dynamo equation for several flow models using high resolution numerical calculation, and numerically expands the solution by Beltrami functions. First we clearly show that dynamo field can grow if and only if the sign of the current helicity, which is created as a result of dynamo process, is same as the kinetic helicity. Secondly, we study the slow dynamo process produced by an integrable flow such as the Roberts cell, and found that the solution of that may be classified into two different classes between the cases those the magnetic Reynolds number (R) is lower and higher than the value to maximize the dynamo growth rate. In the lower R case, the asymmetry between the positive and the negative helicity components, that is the source of dynamo action, exists in the lowest Fourier modes, whereas in the higher R case it shifts to the higher Fourier modes, where the nonlinear coupling is largely affected by the resistive diffusion. Also the coupling between the positive and the negative component is calculated, and it is revealed that the nonlinear coupling between different sign modes is stronger in lower modes. It indicates that the slowing down of dynamo action might be caused by the equipartition of the helicity into different Beltrami modes. Also the result for chaotic flows, those are the candidate of fast dynamo, will be presented.

[NP1.075] The Dynamics of SOC Systems with Classical Diffusion and Parallel Transport

In confined plasmas the transport dynamics is likely to be a combination of coherent anomalous transport (modeled by SOC), a more classical (or neo-classical) diffusive component as well as parallel transport. The parallel transport dynamics are likely to be different then the perpendicular transport dynamics and fundamentally important in regions of open field lines or in devices with open field lines. For different flows, diffusivities and drives a variety of transport regimes are found. In a 2-D system in which the parallel transport is modeled by classical diffusion (or a flow) and the perpendicular transport is modeled by anomalous SOC transport it is preliminarily found that the ratio of the parallel to the perpendicular effective diffusivities can be much greater then 10 and still allow for SOC like dynamics in the perpendicular direction. A spatial transition from SOC like dynamics to diffusive like dynamics is found, setting a radial break point of interest. The implications for transport dynamics in various experimental regimes and devices are discussed.

[NP1.076] Experimental Investigation of Complex Dynamics of Plasma Turbulence and Transport

Theoretical predictions of complex dynamics, such as self-organized criticality (SOC), have led to new insights into the behavior of a wide range of complex systems, such as sandpiles, evolution/extinction models and earthquake fault zones. Recently, complex dynamics have been invoked as a paradigm for understanding turbulent transport in plasmas. In particular, complex dynamical models of turbulent transport make specific predictions regarding power spectra and long range spatial and temporal correlations. In order to test the models experimentally, detailed studies utilizing probe arrays at many axial and azimuthal positions are under way in the linear Large Plasma Device at UCLA. Preliminary edge fluctuation data show frequency spectra with three distinct regions, scaling approximately as f^0, f^-1, and f^-4, in low, intermediate, and high frequency intervals respectively. The f^-1 frequency interval decreases - eventually to zero - as the plasma is scanned from the edge to the core. These observations are consistent with a recently developed complex dynamics model that includes classical diffusion.

*Supported by the National Science Foundation

[NP1.077] Transition to chaos in ion sheath dynamics

Dynamics of externally driven ion sheath in the double plasma device has been investigated by numerically solving a nonlinear differential equation. The global dynamical changes of the system has been examined by evaluating parametric changes of the Poincare map and the maximum Lyapunov exponent. The transitions to chaos caused by the period doubling bifurcation and intermittency are crearly shown by graphical methods. The correlation dimension calculated for a strange attractor agrees well with the experimental one indicates that the chaos is low dimensional, and the system can be described by the low dimensional equation.

[NP1.078] DIII-D Tokamak-CD, MHD, Diagnostics, Data Acquisition (DPP)

This abstract not available.

[NP1.079] Reconstruction of Current Profiles in DIII-D ECCD Discharges

A key element of the DIII-D research program is the use of ECCD to control and sustain the current profile for advanced tokamak study. An important issue is the determination of the ECCD profiles. Previous results obtained from analysis based on a time series of EFIT equilibrium reconstructions using MSE data show that the widths of the ECCD profiles are generally broader than those predicted theoretically. There are indications from ECCD transport simulations that the narrower predicted profiles are consistent with the MSE data and the discrepancy is due to the finite spatial resolution and the smooth basis functions used in the reconstruction. To resolve the discrepancy, various improvements are made to EFIT. These include increasing the spatial grid to 257\times257, an option to optimize the spline knot locations and representations to allow localized features with strong gradient. Detailed comparisons of the reconstructed ECCD profiles using these new EFIT tools against the theoretical predictions will be presented.

[NP1.080] Localized Measurements of ECCD Using MSE Spectroscopy on DIII-D

Electron cyclotron current drive (ECCD) is a valuable tool for current profile control because the generation of localized current at selected radii is easily controlled. Since nearly pure X-mode is required to achieve highly localized deposition, the difference in wave refraction between X-mode and O-mode is used to identify and evaluate the unwanted O-mode component. Two different analysis methods are utilized to verify the localization of the ECCD. First, the plasma current density is separated into inductive and non-inductive components from the evolution of the poloidal magnetic flux; this requires magnetic equilibrium reconstructions with excellent spatial resolution. Second, the measured pitch angles from motional Stark effect (MSE) spectroscopy are compared to simulations of the expected MSE response to localized ECCD. This analysis shows that the width of the ECCD profile is in agreement with theory. The effect of the H-mode edge on the ECCD efficiency and localization will be discussed.

[NP1.081] Electron Cyclotron Heating and Current Drive in DIII-D at High Power Density

Numerical calculations of the electron gyro-orbits passing through a beam of X-mode second harmonic electron cyclotron waves simulating the off-axis ECCD experiments on DIII-D have shown that nonlinear modifications of diffusion from that predicted from RF quasilinear theory sets in at RF injected power density comparable to that used in experiments. Previous work(J.Y. Hsu and S.C. Chiu, Phys.\ Rev.\ Lett.\ 45), 1561 (1980).\ finds that power absorption is increased. Enhanced absorption increases the Fisch-Boozer current while reducing the backward Ohkawa current, leading to increased off-axis current drive efficiency. A direct comparison will be made between QL diffusion coefficients obtained using (1) the standard ray-tracing-based CQL3D QL calculation, and (2) the numerical orbit calculation of diffusion in a toroidal magnetic field due to a coherent diverging Gaussian beam model of the injected power. Nonlinear effects on DIII-D EC current drive will be assessed.

[NP1.082] Ideal and Non-Ideal MHD Stability in High Performance DIII-D Discharges

DIII-D high performance discharges are generally limited by MHD instabilities. Predictions from an array of theoretical, computational, and diagnostic tools are tested against the observations for the principal limiting phenomena in DIII-D. These include disruptions and localized MHD bursts in L-mode NCS discharges, edge instabilities in H-mode high performance discharges, and neoclassical tearing and resistive wall modes in long pulse discharges. In the case of predominantly ideal MHD instabilities, agreement between the code predictions and observed limits can be obtained to within a few percent and the predicted fluctuation profiles and growth rates to within the experimental errors. For non-ideal modes, the predictions can now be productively tested against the observations so that competing effects can be either eliminated or confirmed and, in some cases, the agreement is approaching levels similar to that for ideal comparisons.

[NP1.083] Structure of Resistive Wall Modes in DIII-D

Resistive wall modes limit the performance of DIII-D discharges when beta exceeds the no-wall ideal stability limit. These slowly rotating, n=1 modes are normally detected on DIII-D by a six-coil toroidal array of large-area sensor loops, each covering a 60 degree arc on the midplane. Two new 12-loop arrays of external saddle loops above and below the midplane are now acquiring data routinely. The combined array of 30 saddle loops provides information about both the poloidal and toroidal structure of RWMs. Chord-by-chord comparisons of three identical soft x-ray cameras distributed in a toroidal array confirm the expected kink-like internal structure. The observed poloidal and internal structure will be compared with code predictions.

[NP1.084] Resistive Wall Mode Onset and Active Control in \mboxDIII-D

The n=1 RWM has been clearly observed in DIII-D when the beta exceeds the predicted no-wall beta limit. The RWM onset is strongly affected by plasma rotation and static error fields. At values of beta above the no-wall beta limit, the static error field is observed to be amplified by plasma response leading to a more rapid decay of toroidal rotation and eventual RWM onset. The RWM has been controlled by an active feedback stabilization using six driven saddle coils on the midplane, and powered as three independent n=1 pairs. The RWM amplitude is monitored by a saddle coil sensors measuring the radial magnetic flux through the vacuum vessel under each driver coil. Stabilization of the RWM above the no-wall beta limit has been achieved for more than 30 wall times using ``mode control'' feedback logic and 10 wall times using ``smart shell'' feedback logic. These results are in agreement with predictions of the VALEN finite element 3-D feedback code.

[NP1.085] Modeling of Feedback Stabilization of External MHD Modes in Toroidal Geometry

The intelligent shell feedback scheme(C.M. Bishop, Plasma Phys.\ Contr.\ Nucl.\ Fusion 31), 1179 (1989).\ seeks to utilize external coils to suppress the unstable MHD modes slowed down by the resistive shell. We present a new formulation and numerical results of the interaction between the plasma and its outside vacuum region, with complete plasma response and the inclusion of a resistive vessel in general toroidal geometry. This is achieved by using the Green's function technique, which is a generalization of that previously used for the VACUUM(M.S. Chance, Phys.\ Plasmas 4), 2161 (1997).\ code and coupled with the ideal MHD code GATO. The effectiveness of different realizations of the intelligent shell concept is gauged by their ability to minimize the available free energy to drive the MHD mode. Computations indicate poloidal coverage of 30% of the total resistive wall surface area and 6 or 7 segments of ``intelligent coil'' arrays superimposed on the resistive wall will allow recovery of up to 90% the effectiveness of the ideal shell in stabilizing the ideal external kink.

[NP1.086] Improved Resistive Wall Mode Stability in DIII-D with Optimal Error Field Correction

In the development of an advanced tokamak plasma, beta may be limited by the Resistive Wall Mode (RWM). Sufficient plasma rotation can stabilize the RWM,(A. Boozer, Phys.\ Plasmas 2), 4521 (1995).\ but rotation can be reduced by the torque from uncompensated error fields. In DIII-D plasmas with beta above the ``no wall'' limit, the rotation steadily drops and leads to an n=1 RWM when the rotation decreases below a critical value.(A.M. Garofalo et al.), Phys.\ Rev.\ Lett.\ 82, 3811 (1999). This is consistent with enhanced drag caused by a resonant response to an uncompensated n=1 error field once beta exceeds the no wall limit. A simple torque balance model that includes this effect will be compared with data. The experimental results show that careful error field correction leads to a much longer period of sustained rotation and RWM stability with beta above the no wall limit before the eventual RWM growth is observed.

[NP1.087] Design Study of a Real-Time Resistive-Wall-Mode Identifier using Mirnov Signals for the DIII-D Tokamak*

Real-time identification and control of the Resistive-Wall-Mode (RWM) are crucial for higher performance operation in present and future tokamaks. Identification of the RWM in a tokamak is difficult due to its slow rotation speed and the fact that its growth rate is comparable to the field penetration rate of the vacuum vessel wall. The mode can grow large enough to dramatically degrade confinement or cause disruption in less than one toroidal revolution of the mode. A matched filter method of identifying the n=1 RWM component in DIII-D using signals from toroidal and poloidal arrays of Mirnov probes has been developed. Estimations of the time required for data collection and analysis confirms the feasibility of using this method to produce a real-time mode identifier suitable for use in feedback stabilization of the mode on the DIII-D tokamak. The preliminary design of a real-time RWM identifier for integration with the DIII-D systems will be presented, including data acquisition and analysis hardware.

*Work supported by DOE Grant/Contracts No. DE-FG03-99ER82791, DE-AC03-99ER54463, DE-FG02-89ER53297 and DE-AC02-76CHO3073.

[NP1.088] Parameter Scaling of H-Mode Edge Region Stability

The observed dependence on discharge shape of the edge-localized-mode (ELM) amplitude and frequency in DIII-D tokamak H-mode plasmas has been shown to be consistent with a model for the pressure gradient (P^\prime_edge) stability threshold as a function of toroidal mode number (n).(J.R.\ Ferron et al., Phys. Plasmas 7), 1976 (2000). In this model, based on ideal MHD stability theory, the instability responsible for triggering a Type~I ELM is a coupled kink/ballooning mode with n near the largest value without access to a second stability regime. Here we add to the understanding of this model through additional parameter scaling studies. The calculated sensitivity to the edge current density (J_edge) of the P^\prime_edge stability threshold versus n is used to determine the importance of the uncertainty in the measurement of J_edge. The dependence of the P^\prime_edge threshold on the pressure pedestal width and the total plasma current are compared to experiment. Equilibria characteristic of those at the end of the VH-mode ELM-free phase, when the observed P^\prime_edge can be especially large, are tested for second stability regime access at low values of n.

[NP1.089] Stability and Nonlinear Dynamics of the Tokamak Edge Region, with Parallel Current

Theoretical analysis of edge instabilities which may control the pedestal height and width is complex, in part because the sharp pressure gradients, and consequent large bootstrap currents, near the H-mode edge can destabilize kink, peeling, and ballooning modes over a wide range of toroidal mode numbers (n). An important modification of classical ballooning theory [Conner et al.] allows treatment of the coupled system of edge ballooning and peeling modes at higher n \agt 10. Use of a nonlocal edge ballooning/peeling code \textttELITE together with low-n stability codes allows the study of the ideal MHD edge stability of real tokamak equilibria over essentially the full spectrum of toroidal modes, leading to insights on ELMs and the importance of parallel current. To study the impact of X-point geometry, non-ideal effects, and nonlinear dynamics, simulations are carried out on the electromagnetic Braginskii code \textttBOUT, enhanced to include parallel current terms.

[NP1.090] Sensitivity Studies of Tearing Mode Stability Calculations

For high \beta, highly shaped plasmas in the DIII-D tokamak, the value of the tearing mode stability index \Delta' calculated at a rational surface can depend sensitively on the pressure and current profiles when an ideal mode is near marginal stability in the equilibrium current profile parameter space. Using a single time slice of experimental data and fitting equilibria around a minimum in \chi^2, we show that an estimate of the error in \Delta' will be small when no ideal mode is present. Also, the \Delta' calculation will systematically indicate linear stability to tearing modes (\Delta' < 0) when a global ideal mode is present. Between these regions, near marginal stability for the global ideal mode, a pole in \Delta' exists as predicted by analytic theory [\mu= (-D_I)^1/2 is near 0.5 at the rational surface], and the proximity of the best equilibrium fit to this pole in parameter space is crucial to the accuracy of the tearing mode stability calculation. These results will be applicable to the study of ECCD stabilization of tearing modes, and the accuracy of stability studies of parameterized current peaks at rational surfaces is discussed.

[NP1.091] Scaling of the Critical Beta for Onset of the m/n=2/1 Neoclassical Tearing Mode in Conventional H-Mode Discharges in DIII-D

While m/n=3/2 NTMs have been observed and studied in detail,(R.J. La Haye et al.), to be in the August 2000 Phys.\ Plasmas.\ their consequences are small compared to the m/n=2/1 mode which tends to lock, destroy the H-Mode and cause disruption. The 2/1 modes in DIII-D H-Mode discharges appear to be NTMs in that they are excited as beta is rising, are triggered by a sawtooth crash, ELM or both and have a nearly linear critical beta with rhoistar (the ion gyroradius normalized to the plasma minor radius). Analysis of a 2/1 database in DIII-D will be presented. Preliminary comparison to the polarization\slash inertial theory,(H.R. Wilson et al.), Phys.\ Plasmas 3, 248 (1996).\ particularly of the key issue of island propagation in the local (q=2) E_r=0 quasi-neutrality frame, shows consistency with a stabilizing effect, i.e., a threshold.

[NP1.092] Criteria for Suppression of Neoclassical Tearing Modes

This work reports a theoretical determination of the level of Electron Cyclotron Current Drive (ECCD) needed to suppress Neoclassical Tearing Modes. The ratio of peak driven current density j_cd to the bootstrap current density j_bs is the appropriate figure-of-merit. Altogether there are four criteria. The first governs complete stabilization of NTMs and requires modulated ECCD. The second is the criteria for the existence of two additional fixed points of the island evolution equation, which should limit an NTM to a width comparable to the width of the driven current layer w_cd. The same condition j_cd > 1.6\, j_bs, evaluated for continuous ECCD, fufills both criteria. The third criterion governs the ability to decrease the size of existing, saturated islands. For islands with a saturation width w_sat>w_cd, the criterion is j_cd > 0.16\, j_bs (w_sat/w_cd). The fourth criterion concerns the ability to increase the quantity (-\Delta') by a thin, continuous current drive layer centered on the rational surface. The criterion j_cd > 0.6\, j_bs leads to an island size w

[NP1.093] Simulations of High betaN*H Advanced Tokamak Discharges in DIII-D with the 3D Nonlinear Code NFTC

Nonlinear self-consistent MHD stability simulations of neoclassical tearing modes (NTM) in high \beta_N Advanced Tokamak (AT) are presented. Radially localized electron cyclotron current drive (ECCD) current profile control is considered based on DIII-D discharge #99411 in which \beta_N=3.9 and H_89P-factor=2.9 are reached. Simulations were performed with the full 3D nonlinear code NFTC. Neoclassical terms are included in the basic equations for the magnetic field and pressure. An effective fully implicit numerical scheme allows the transport profile to evolve self-consistently with the nonlinear MHD instabilities and an externally applied source such as ECCD. NTM activity with m/n=2/1 is found in simulations to correspond to experiment. It is shown that magnetic islands could be quickly suppressed by localized ECCD at q=2 before the mode grows substantially. The possibility of q-profile modification by ECCD well before MHD activity to keep q_min>2, so that the discharge evolves stably is also discussed.

[NP1.094] Features of the Kink Mode Structure in Elongated DIII-D Plasma

The unusual behavior of large scale MHD-modes was investigated using joint analysis of Mirnov signals, Electron Cyclotron Emission and Soft X-Ray data. In elongated or divertor plasmas, the internal kink-like n=1 precursor associated with sawtooth or fishbone events can excite coupled structures having m>1, with such amplitude and phase relations that cause an apparent reversal of the pitch of the magnetic perturbation (in plasma frame) or of the direction of mode rotation (in laboratory frame) along the inboard wall. In the case of neoclassical tearing modes, when the m=3/n=2 mode excites coupled modes such as 2/2 and 5/2, this phenomenon is absent. Simple analytic calculations and simulations show that in some cases the coupling of the modes m=1,2,3 gives a reverse phase shift in a restricted sector of the inboard wall, but in this model it is difficult to obtain reverse phase shift in the sector of 180 degrees from the top to bottom. We have no good theoretical explanation of this experimental observation.

[NP1.095] Magnetic Helicity is Conserved at a Tokamak Sawtooth Crash

The sawtooth instability causes sudden changes in magnetic topology during combined neutral beam and fast wave heating in the DIII-D tokamak. Measurements with a Motional Stark Effect diagnostic provide accurate determination of the equilibria before and after the sawtooth reconnection events. The global magnetic helicity \int \vecA\cdot\vecB\, dV changes 0.2\pm0.9% at a sawtooth crash. The local change in helical flux, \chi, is roughly consistent with the Kadomtsev(B.B. Kadomtsev, Sov.\ J. Plasma Phys.\ 1) (1975) 389.\ model within large errors. The volume in which the helical flux changes is 85\pm15% of the volume predicted by Kadomtsev, while the central value of \chi is within 1% of the predicted value.

[NP1.096] Neutral Particle Measurements of the Pitch-Angle Scattering Rate

Three vertically-viewing neutral particle analyzers and one horizontally scanning analyzer(E.M. Carolipio and W.W. Heidbrink, Rev.\ Sci.\ Instrum.\ \bf68) (1997) 304.\ are used to study the pitch-angle scattering rate of beam ions.

As in the classic experiment by Goldston,(R.J. Goldston, Nucl.\ Fusion 15) (1975) 651.\ a short beam pulse creates the intial beam-ion distribution, then the scattering rate is inferred from the charge-exchange signals. The theoretical pitch-angle scattering rate is calculated from measurements of T_e, n_e, and Z_eff. In preliminary analysis, the deceleration rate agrees well with classical theory but the scattering data are inconsistent with a Green's function solution^3 of the Fokker-Planck equation. More detailed simulations with the TRANSP code are planned.

[NP1.097] Predictive Modeling of Halo Currents in Disruptions and Disruption Mitigation Scenarios

The success of recent models of disruption halo currents(D.A. Humphreys, D.G. Whyte, ``Classical Resistivity in a Post-Thermal Quench Disrupting Plasma,'' accepted for publication in Phys.\ Plasmas.)\ and post-thermal quench radiation power balance(D.G. Whyte, D.A. Humphreys, P.L. Taylor, ``Measurement of Plasma Electron Temperature and Effective Charge During Tokamak Disruption,'' accepted for publication in Phys.\ Plasmas.)\ offer the hope of accurately predicting halo currents expected in disruptions mitigated by massive gas injection. Such an ability to reliably predict halo currents in rapid shutdown scenarios is of great importance in the design of next-generation tokamaks. The method is illustrated in application of an extensively validated semi-analytic model to prediction of halo currents expected in unmitigated and mitigated disruptions in the JT-60SU device design and the DIII-D tokamak experiment. Implications of specific halo current predictions and the general method itself are discussed.

[NP1.098] Database Analysis of Evolution to Disruption in \mboxDIII-D

A database consisting of the temporal evolution of several hundred discharges to end-of-shot or disruption was initially constructed last year and initial results were reported.(A.W.\ Hyatt, et~al., Bull.\ Am.\ Phys.\ Soc. 44), 77 (1999). We extend this database and further investigate several preliminary results. One preliminary result is the observation of a region in parameter space where high normalized beta (3 < \beta_N < 4) operation is apparently disruption free. We further investigate whether the disruptivity per unit time diminishes with increasing time after some marker such as maximum \beta_N is reached and whether this is influenced by end-of-shot operational practices such as turning off auxiliary heating or fuel gas flow. We also report on preliminary investigations into parameter space trajectories of discharges that do and do not end in disruption

[NP1.099] Advanced Diagnostic Requirements for the DIII-D Tokamak

A primary focus of the DIII-D advanced tokamak research program is to seek the ultimate potential of the tokamak as a magnetic confinement system. Achieving this potential involves optimizing the plasma cross-sectional shape, current density, and pressure profiles for stability to MHD modes while simultaneously controlling the current density, pressure and radial electric field profiles to minimize the cross field transport of plasma energy. The development of the scientific understanding underlying the simultaneous, nonlinear optimization of shape, current, pressure, and electric field profiles to meet multiple goals is an ambitious goal with potentially very high payoff. Diagnostic measurements play a crucial role in this research both for developing predictive models and for serving as sensors for real-time feedback control. Some of the measurements that are needed are detailed edge current profiles, measurements of electron transport, and measures of fluctuation driven transport. In this paper we will discuss the diagnostic measurement needs for advancing the tokamak concept, possible techniques and plans for diagnostics and diagnostic development on the DIII-D tokamak.

[NP1.100] Offline Methods for Calibration of the Motional Stark Effect Diagnostic

A variety of algorithms and methods to optimize the fitting of Motional Stark Effect data to equilibria generated by EFIT have been implemented and compared in an effort to identify and minimize sources of systematic error in the standard analysis. These methods include a novel calibration technique in which Ohmic, L-mode I_p-ramp shots are used to optimize fitting coefficients. Results based on this method consistently demonstrate lower \chi_MSE^2 throughout a shot when compared to analysis carried out using coefficients generated by the previous method. Sensitivity studies have also been carried out to determine the local and global accuracy of the reconstructed equilibria. These reveal that there is about twice the uncertainty in the inferred value of q for \rho <\,0.15 than for \rho >\,0.15.

[NP1.101] Design of a 2-D X-Ray Imaging System for ECCD/ECH Experiments on DIII-D

During electron cyclotron current drive experiments, the electron cyclotron waves produce an asymmetric electron distribution function which causes an asymmetric plasma resistivity and generates a wave-driven current. Due to the asymmetric bremsstrahlung radiation pattern of energetic electrons, it also produces an asymmetric x-ray emissivity which reflects the energetic electron distribution function. A 2-D x-ray imaging system is designed to measure x-ray emission from the plasma for the study of the energetic electron distribution function during ECCD experiments on DIII-D. The design is based on a high resolution commercial image intensifying tube. This instrument allows detailed studies of ECCD physics at the fundamental level. It also can be used to study electron transport using the energetic electrons as test particles. The design parameters will be chosen based on measurements from a single channel x-ray pulse height spectrometer with a view through the plasma core. A detailed design compatible with the DIII-D environment will be presented.

[NP1.102] Automated Analysis of Reflectometer Density Profiles on DIII-D

Over the last year the DIII-D reflectometer system has been further upgraded so as to operate for the first time as a ``standard'' diagnostic system on DIII-D, i.e., with profile coverage throughout the discharge duration, and automated profile analysis. Specifically, an initial automated between-shots profile analysis capability has been implemented. The automatic analysis has proved robust for edge X-mode measurements, but is not yet as reliable for core O-mode measurements. In addition, the availability of analyzed profiles to end-users has been significantly improved in a modern multi-user environment, using the standard DIII-D ``Reviewplus'' and ``GAprofiles'' data viewing and fitting packages.

[NP1.103] An Alignment Feedback System for the Thomson System at DIII-D

The DIII-D Thomson system measures electron density and temperature with eight pulsed ND:YAG lasers along three 35~m paths through the plasma vessel. The density measurement is especially sensitive to any drift in the laser alignment during operation days at DIII-D. This sensitivity includes the drift of the lasers relative to the collection optics during plasma operations and changes in light levels scattered from the vessel walls during calibrations. A new feedback system is being installed to control the drift of the eight ND:YAG lasers as well as three HeNe lasers used to align the ND:YAG lasers along the three laser paths. The feedback system will use six CCD cameras and motorized mirror mounts to stabilize the alignment HeNe laser along each path. Separate motorized mirrors are used to align the ND:YAG lasers along the same path as the HeNe lasers. The system is currently being installed and tested for operation during the 2001 experimental campaign.

[NP1.104] Design of a Dual High-Power, Long Pulse, Steerable ECH Launcher for DIII-D

ECH launchers typically use a fixed, and a moveable mirror to direct a gyrotron beam to a desired location. Increased power handling capability in an ECH launcher requires adding mass or active cooling to the mirrors. The additional forces on the moveable mirror resulting from either approach typically require a compromise between the power handling capability and the steering capability.

The P2001 launcher designed by PPPL for DIII-D will launch two 800kW beams for 10 seconds every 10 minutes. Poloidal and toroidal steering during and between pulses, and a fast (100deg/sec) poloidal scan capability, will be provided. These conflicting design requirements are satisfied by the use of an innovative steering mechanism. By taking advantage of the geometry of compound Euler rotation angles, the entire scanning range about two axes can be provided with a mechanism that is strong enough to withstand the electromagnetic and inertial forces on a long pulse, high power mirror.

In this paper, the design goals for the P2001 launcher, and the challenges arising from them, are presented. The design of the P2001 launcher is presented, along with calculations to verify its performance and feasibility. Design issues relevant to advanced fusion experiments are highlighted and discussed.

[NP1.105] Data Management and Visualization to Enhance Science Discovery through Advanced Computing in the PSACI Project

The USDOE\slash OFES supported Plasma Science Advanced Computing Initiative (PSACI) is designed to revolutionize fusion research by greatly enhancing simulation and modeling capabilities made accessible by terascale computing. The power of advanced computing to solve critical plasma science problems can be fully exploited only if simultaneously a capable infrastructure is established and effective software tools are made available. This infrastructure includes establishing standardized data structures and access methods, synthetic diagnostics, standard analysis and visualization utilities, and common code interfaces. Work has included support of two PSACI pilot programs: Macroscopic Modeling and Microturbulence Simulation of fusion plasmas. MDSplus provides a standard interface to simulation data from NIMROD, M3D, and GS2. IDL tools act as both synthetic diagnostics and provide interactive scientific visualization for these codes. Demonstrations will be given.

[NP1.106] Enhanced DIII-D Data Management Through a Relational Database

A relational database is being used to serve data about DIII-D experiments. The database is optimized for queries across multiple shots, allowing for rapid data mining by SQL-literate researchers. The relational database relates different experiments and datasets, thus providing a big picture of DIII-D operations. Users are encouraged to add their own tables to the database. Summary physics quantities about DIII-D discharges are collected and stored in the database automatically. Meta-data about code runs, MDSplus usage, and visualization tool usage are collected, stored in the database, and later analyzed to improve computing. Documentation on the database may be accessed through programming languages such as C, Java, and IDL, or through ODBC compliant applications such as Excel and Access. A database-driven web page also provides a convenient means for viewing database quantities through the World Wide Web. Demonstrations will be given at the poster.

[NP1.107] A Linux PC Cluster for Between-Pulse EFIT and Other CPU Bound Analyses at DIII-D

A 12-processor PC Linux cluster, STAR, has been installed to perform between-pulse magnetic equilibrium reconstructions using the EFIT code written in Fortran. The MPICH package implementing Message Passing Interface is employed by EFIT for data distribution and communication. The new system calculates equilibria eight times faster than the previous system yielding a complete equilibrium time-history on a 25~ms time-scale four minutes after the pulse ends. A graphical interface is provided for users to control the time resolution and the type of EFITs. The next analysis to benefit from the cluster will be CERQUICK written in IDL for charge exchange recombination analysis. The plan is to expand the cluster so that a full profile analysis (T_e, T_i, n_e, V_r, Z_eff) will be available between pulses, which lays the ground\-work for Kinetic EFIT or ONETWO transport analyses. The poster will present the description of the cluster and detail of the between-pulse EFIT and the future plans.

[NP1.108] FIRE, ITER, Ignitor and KSTAR (DPP)

This abstract not available.

[NP1.109] FIRE, A Next Step Option for Magnetic Fusion

The next step experiment in magnetic fusion should address both burning plasma physics and advanced toroidal physics in a regimes approaching those expected for a magnetic fusion plasma. The coupling of transport, MHD stability and plasma boundary physics with strong self-heating leads to a self-organized plasma state that presents a challenge for magnetic fusion science. A national design study of a next step option is underway to develop and assess near term opportunities for addressing the issues identified above. The emphasis is on exploring and understanding the behavior of plasmas dominated by alpha heating (Q \geq 5) that are sustained for a duration comparable to characteristic plasma time scales (\geq 10 \tau_E, \sim \tau_SKIN . The study has focused on the technical evaluation of a compact, high-field, cryogenic-copper-magnet, highly-shaped tokamak with the parameters: Ro = 2m, a= 0.525m, \kappa_95 \approx 1.8, \delta_95 \approx 0.4, q_95 > 3 , double-null-divertor, Bt(Ro) = 10 T, and Ip = 6.44 MA and flat top time \sim 20 s ( 25 \tau_E and \ge 1 tau_SKIN). The results of the study and areas needing additional work will be discussd.

Work supported by DOE Contract # DE-AC02-76CH0 3073.

[NP1.110] Global Stability Issues for a Next Step Burning Plasma Experiment

We present analysis which supports the feasibility of a next-step burning plasma experiment. The FIRE design has R = 2 m, a = .525 m, \kappa_95 = 1.77, \delta_95 = 0.4, B = 10(12) T, I = 6.44(7.7) MA, H = 1.2 (1.0) for the reference (high-field) discharge, with monotonic q-profile and sawtoothing ELMy H-mode operation. The primary issues for MHD are associated with (1) the q=1 surface, (2) energetic particle modes (3) edge currents due to the H-mode pedestal,(4) neoclassical tearing modes, and (5) error fields and locked modes. We find (1) the m=1, n=1 mode requires non-linear analysis including energetic-particle effects, (2) \alpha-particle driven Alfven modes, RTAE and KTAE, are expected to be stable for \beta_\alpha 0 < 0.5 %, (3) the predicted critical value for the onset of the NTM is very close to the operating point for the high-field option, and may be mediated by self or active control of seed island width or active island current drive, (4) the nominal self-consistent operating point is stable to external kink modes without a conducting wall and (5) error field requirements need to be revisited. Advanced operating modes with q > 2 everywhere and high-bootstrap fraction also hold promise but need to be further developed.

[NP1.111] Physics Design Guidelines for Estimating Plasma Performance in a Burning Plasma Experiment (FIRE)

The physics design guidelines for a next step, high-field tokamak, burning plasma experiment (FIRE, Fusion Ignition Research Experiment) have been developed as an update of the ITER Physics Basis (IPB) [NF 39, 2137, 1999]. The IPB represented a comprehensive account of the scientific knowledge, existed as of mid-1998, relevant to the design of a reactor-scale tokamak. Physics design guidelines and methodologies for projecting plasma performance in FIRE or in any reactor-class tokamaks are developed from extrapolations of various characterization of the IPB database for tokamak operation and of the understanding that its interpretation provides for both conventional and advanced tokamak operating modes. The plasma performance attainable in FIRE (or any reactor) is affected by many physics issues, including energy confinement, L-to-H-mode power transition thresholds, MHD stability/beta limit, density limit, helium accumulation/removal, impurity content, sawtooth effects, etc. Design basis and guidelines will be provided in each of these areas, along with sensitivities and/or uncertainties involved.

[NP1.112] Transport Comparisons between ITER-FEAT and Compact High-Field Tokamak Reactors

The next generation magnetic fusion reactors under active consideration are ITER-FEAT and compact high-field (CHF) tokamaks. These represent two rather different concepts for achieving a burning plasma. We use the BALDUR predictive transport code with the MMM99 transport model and the alternative OHE transport model(Aaron J.~Redd, et al.,) Phys. Plasmas 5, (1998) 1369. with ITG and new ETG physics. We simulate the proposed reactor designs and compare their fusion performance parameters. ITER-FEAT is a next step large NBI driven reactor with the goal of achieving Q=10 and thus represents a straightforward scaling of our recent study of the transport in JET and DIIID.(P.~Zhu, et al.,) Phys. Plasmas \textbf7 (2000) 2898. Comparisons will be made with CHF reactor designs as well as the original ITER design in order to clarify issues identified in the comparison of these machines types. The off-diagonal turbulent particle pinch will be studied as a key issue in the CHF reactor design simulations.